Court dismisses appeals by failed protection applicants claiming state protection was not available for threats from non-state actors.

APPROVED NO FURTHER REDACTION NEEDED

THE COURT OF APPEAL

CIVIL

Court of Appeal Record Number: 2024/50

High Court Record Number: 2022/34JR

Neutral Citation: [2025] IECA 208

Whelan J. Butler J. Meenan J.

BETWEEN/

Z.M.

APPLICANT/ APPELLANT

- AND -

THE INTERNATIONAL PROTECTION APPEALS TRIBUNAL

AND THE MINISTER FOR JUSTICE AND EQUALITY

RESPONDENTS

Court of Appeal Record Number: 2024/11

High Court Record Number: 2022/915JR

BETWEEN/

N.G.

APPLICANT/ APPELLANT

- AND -

THE INTERNATIONAL PROTECTION APPEALS TRIBUNAL,

THE MINSITER FOR JUSTICE AND EQUALITY,

IRELAND AND THE ATTORNEY GENERAL

RESPONDENTS

JUDGMENT of Ms. Justice Butler delivered on the 21st day of October 2025

Introduction

1. This judgment deals with two appeals which, because they raise identical issues in similar albeit not identical circumstances, were heard together. The appellants in both cases are failed protection applicants who challenge the manner in which the appellate decision maker (“IPAT” or “the Tribunal”) reached the decision refusing their appeals.

2. The similarity between the appeals arises because the claim for protection in both cases concerned threats made to the appellants by criminal elements in their countries of origin, Georgia and Albania respectively, and the assertion that the police and other authorities in those countries were unable to effectively protect the appellants against those threats. Additionally, both Georgia and Albania have been designated by the Minister for Justice as safe countries of origin, the significance of which will be considered further below. The differences, which are not material to the legal issues, arise from the fact that Z.M., the Georgian appellant, was actively involved in the criminal conduct (a feud between two rival factions) which ultimately led to the threats of which he complains whereas N.G., the Albanian appellant, was unwittingly caught up in the criminal conduct of third parties due to his employment as a restaurant worker by one of them.

3. The central issue in the appeal, which was raised in identical terms in both cases, arises because the persecution or serious harm of which the appellants complain is at the hands of “non-state actors” and thus, the focus of the negative decisions against them was on the extent to which effective state protection was available to them against these third parties in their countries of origin. This in turn involved a consideration of the extent to which the availability of state protection was or should have been considered by the decision-maker before concluding, as he did, that the persecution or serious harm feared by the appellants was “well-founded” or constituted a “real risk”.

4. Of itself it might not appear to make much difference whether the availability of state protection is considered as part of the well-foundedness of the fear of persecution (or the reality of the risk of serious harm) or as a subsequent, separate step provided it is appropriately considered by the decision maker before concluding whether or not a person is a refugee or in need of subsidiary protection. However, in this case the appellants contend because the decision maker split the analysis of these issues in the manner in which he did, he did not apply a statutory presumption – sometimes termed an evidential advantage – which he should have applied for their benefit in his analysis of the availability of state protection. The presumption in question under Section 28(6) of the International Protection Act 2015 (“the 2015 Act”) provides that where past persecution is established, it will be a “strong indicator” of future persecution.

5. Finally under this heading, an issue arises as to the content of state protection and whether national authorities must be able to prevent the materialisation of threats of persecution or serious harm in order for the protection of the state to be effective.

6. Both appellants contend that the fact the decision maker had regard to the designation of their country of origin as a safe country was unlawful. In essence the contention seems to be that unless the designation is being formally relied on, presumably to treat the application as manifestly unfounded or for similar reasons, it cannot be invoked at all by the decision maker and particularly not in an analysis of the availability of state protection.

7. Two additional issues are raised by the Z.M. appellant which are not raised by the N.G. appellant. Both issues concern the treatment of the evidence by the decision maker. The first queries the standard of proof applicable to the ascertainment of past events and the second the applicability of the “benefit of the doubt” to the decision maker’s analysis of credibility and plausibility.

8. These judicial reviews were decided against the appellants by two different High Court judges, Hyland J. in the case of Z.M. [2023] IEHC 637 and Phelan J. in the case of N.G. [2023] IEHC 535. Both High Court judges granted leave to appeal to this court under section 5(6) of the Illegal Immigrants (Trafficking) Act 2000 (as amended) in respect of an identical suite of questions concerning the meaning of the terms “well-founded fear of persecution” and “real risk of serious harm” and the relationship of those terms to the concept of state protection. The appellants framed the issue slightly differently in their written submissions, but all of the points raised in the four subparagraphs of the question certified are covered in their analysis of the issues as presented to this court. Z.M. pursued the two additional points outlined above without having been granted leave to appeal on them. However, the respondent accepted that once leave to appeal was granted under section 5(6), the appellant was at large to pursue any grounds of appeal regardless of whether the point had been certified and even where it had been expressly refused by the High Court judge.

9. The outcome of these cases in the High Court was perhaps unsurprising in circumstances where the main issues had been raised in earlier cases and decided against the respective applicants (although some of those applicants did succeed on other, narrower, grounds). In particular, the argument regarding whether the availability of state protection must be considered before a conclusion reached that the fear of persecution is well-founded was made in identical terms to and rejected by Heslin J. in T.A. v. The International Protection Office [2023] IEHC 390. Further, in O.M. v. IPAT [2025] IEHC 51 O’Regan J. considered and determined the standard of proof to be applied to the consideration of past events in an application for international protection.

10. As the issues in the appeal are very dependent on the interpretation of the text of EU and Irish legislative provisions, I propose to consider the relevant legislation, being the Qualification Directive 2004/83/EC, the Procedures Directive 2005/85/EC and the International Protection Act 2015, as amended, in conjunction with the particular issues rather than separately at the outset. I will also address the manner in which the High Court decided each of the issues and the parties’ arguments in relation thereto in conjunction with the particular issues. However, I will commence with a brief overview of the facts in each case and the two decisions under challenge.

11. I have one final preliminary observation. In neither of these cases did the appellant claim a “Convention nexus” which would have brought their application within the scope of the Geneva Convention and which, if successful, would have entitled them to a grant of refugee status. Therefore, in each case the ultimate decision was whether they were entitled to subsidiary protection on the basis of establishing a real risk of serious harm. Because of the way the impugned decisions are structured, the decision maker examined firstly whether each appellant was entitled to refugee status and then, having concluded that they were not, whether they were entitled to subsidiary protection. The legal issues concerning the availability of state protection and the standard of proof are identical whether the test is for refugee status or for subsidiary protection. As it is linguistically awkward to constantly refer to both, I will on occasion default to using the language applicable to one or other of them. However, it should be understood that, unless expressly stated, the analysis in this judgment is equally applicable to both.

The Z.M. Decision

12. Z.M. is a Georgian national in his fifties. His claim for protection is based on threats made to him, via third parties, by a man referred to here as “G.C.”. There is a long history to the appellant’s account of how these threats came to be made. More than 20 years ago a dispute arose between G.C. and Z.L., who was a close associate of the appellant’s being his son’s godfather. In May 2003 Z.L. assaulted G.C. and went immediately to the appellant’s house. He was followed by G.C. and his brother who attacked both Z.L. and the appellant. The appellant sustained a head injury which required hospitalisation. The feud escalated and some days later Z.L. decided to kill G.C.’s family.

13. The appellant originally claimed that he had tried to dissuade Z.L. from doing this and that both he and his cousin followed Z.L. to G.C.’s house to try and stop him. The appellant initially claimed they did not succeed in this objective and remained outside G.C.’s house while Z.L. entered the house and shot and killed G.C.’s mother and brother. The appellant’s account at his oral hearing differed materially from his earlier account. On this occasion he claimed not to have been present at G.C.’s house at all but that he was planning to help Z.L. escape across the border after the murders had been committed. Further, both accounts given by the appellant differ from that accepted by the criminal courts before which he was tried, including the Georgian Supreme Court which expressly accepted the evidence of G.C.’s wife and child which placed the appellant inside G.C.’s house on the night of the murders. The appellant's sentence was reduced as the Supreme Court accepted that although he had come to the locus “for the purpose of murder” he did not implement or take the actions which resulted in the death of the victims. The appellant contends that the testimony of G.C.’s wife and child was false.

14. The feud then escalated and G.C. retaliated by burning the appellant's parents’ house thereby rendering them homeless. In February 2004, G.C. was sentenced to 4 years imprisonment for this arson, but it may be that having already been in custody for seven months, the balance of his sentence was suspended. The judgment of the regional court in this case was submitted by the appellant as part of his claim but the translation of it – and indeed the translation of all of the official Georgian documents submitted – is very poor.

15. On the 7th/8th of February 2004 the appellant's cousin, who appears to have been present the night G.C.’s family members were killed, was murdered, apparently by G.C. and four others. At this point Z.L. was arrested and the appellant handed himself into the police. Both were tried for murder and convicted in early 2005. Z.L. was sentenced to life imprisonment. As noted, the appellant’s initial sentence of 18 years was subsequently reduced to 10 years. G.C. was also arrested, tried and convicted for the murder of the appellant's cousin and sentenced to 23 years imprisonment.

16. As all surviving parties were now in prison, nothing further occurred until the appellant was released from prison in 2012. He claims to have received threatening messages from G.C. through other prisoners as they were released from prison including, in particular, in 2015. At his tribunal hearing the appellant said that he reported this threat to police in 2015 but that no action was taken. The appellant then claims to have been summoned by the head of the local government to attend the police station in March 2019. He claims that the head of the local police told him that his life was at risk from G.C. and that he needed to leave Georgia within 15 days. If he did not leave, he was told he would either have to hire private security or that the only way the police could protect him was to arrest him and put him in custody on false charges. The appellant left Georgia in April 2019.

17. Z.L. was released from prison subsequent to the appellant’s departure and died in a car accident on 27th September 2019 in what the appellant claims were suspicious circumstances. He also claims that the police questioned G.C., who was still in prison, in respect of this death over an extended period which is variously reported as either 10 or 12 days. G.C. was released from prison in December 2019. The appellant claims that he has continued to receive death threats from G.C. communicated to him through third parties including, since his arrival here, through an individual based in this country. He has not reported these threats to the Gardaí.

18. Following an interview held on 14th July 2021, the appellant’s claim for protection was refused by the IPO on 10th August 2021. Whilst the appellant’s account of the various earlier crimes, arrests and convictions was accepted, his account of his interaction with the police in 2019 was considered to be implausible and lacking credibility. His account of being threatened by G.C. was also not believed.

19. The appellant appealed to IPAT which held an online oral hearing in October 2021. The Tribunal issued a decision refusing the appeal and affirming the recommendation of the IPO on 17th November 2021. It should be noted that the same Tribunal member heard the appeal and gave the decision in the N.G. case. Consequently, the two decisions the subject of these appeals are structured and expressed in a similar manner.

20. Following a brief introduction summarising the recommendation of the IPO and the appellant’s grounds of appeal, the decision sets out the allegations made by the appellant in support of this claim under the heading “case, facts and documents”. The claim is summarised at paragraph 2.18 as follows: – “The appellant is afraid that he will be harmed if he is returned to Georgia. He fears harm at the hands of [G.C.]. The Georgian police would not protect the appellant. The appellant would not be safe if he moved to another party of Georgia as Georgia is a very small country and [G.C.] would find him anywhere.”

21. The Tribunal accepted that the appellant was from Georgia. It then proceeded to analyse the “facts and circumstances”, summarising the facts material to the appellant’s claim. It was accepted that the appellant himself had been convicted and sentenced as an accessory to the murder of G.C.’s family members. It was also accepted that G.C. was responsible for the burning of the appellant's parents’ house and the murder of the appellant’s cousin. The Tribunal did not regard the appellant’s suspicion that G.C. was responsible for the death of Z.L. in a car accident as having been established on the balance of probabilities.

22. On the key issue of whether the appellant was advised by the police to leave Georgia because they could not protect him from G.C.’s threats, the Tribunal found the appellant’s account to be wholly implausible. In the view of the Tribunal, the appellant’s account of his dealings with the police in March 2019 ran “strongly counter to the actions of the police in relation to other dimensions of his claim”. It noted the previous investigations and prosecutions which resulted in G.C. serving a lengthy prison sentence and the fact that, on the appellant’s account, G.C. was questioned at length regarding the death of Z.L. It was also noted that no country of origin information was relied on to support the contention that the Georgian police advised Georgian citizens that they cannot protect them and that they should leave the country for their own safety. This element of the claims was held not to have been established on the balance of probabilities.

23. The Tribunal member then proceeded to analyse whether the appellant had demonstrated a well-founded fear of persecution based on the elements of his claim which were accepted. In the event this analysis was largely moot as at para. 5.4 of the decision, the Tribunal concluded that the appellant had not established a nexus with any of the grounds of persecution under the Geneva Convention and therefore regardless of the well-foundedness of his fear, he could not qualify as a refugee. However, the same analysis is relied on in assessing whether those elements, i.e. the material facts which were accepted by the Tribunal, demonstrated a real risk of serious harm which could entitle him to subsidiary protection.

24. The Tribunal accepted that the harm the appellant feared at the hands of G.C. could constitute persecution and/or serious harm based on G.C.’s previous actions in burning his parents’ house and the various threats made between 2012 and 2021. Thus, it held that it had been established “That there is a reasonable chance that if he were to be returned to Georgia the appellant would face a well-founded fear of persecution at the hands of [G.C.]”.

25. The Tribunal then proceeded to consider the question of state protection which is at the heart of this appeal. It concluded that even if a Convention nexus had been established, the claim would fail “in any event on the basis that state protection is available to him in Georgia”. It summarised the issue, using the language of Article 7 of the Qualification Directive, as being “whether reasonable steps are taken by the authorities in Georgia to prevent the kind of persecution at the centre of the appellant’s claim”. Then it looked at country of origin information relating to Georgia emanating from the US State Department, Human Rights Watch and Amnesty International. It was accepted that this material indicated areas in which improvement could be made and that specific allegations of abuse have been made against the Georgian police and security services but, overall, the Tribunal took the view that the country of origin information described “a broadly functioning police force in Georgia” and that it “did not suggest that there were shortcomings or ineffectiveness in the way the Georgian police investigate ordinary crime or preform their normal police functions”. It also had regard to the fact that Georgia had been designated a safe country of origin under S.I. 121 of 2018 which was “indicative of effective state protection in Georgia”. Further, that portion of the appellant’s claim which had been accepted, showed the Georgian police and prosecution service acting reasonably in response to the various crimes which form part of the narrative of the appellant’s claim.

26. These three strands - the country of origin information, the designation of Georgia as a safe country under S.I. 121 of 2018 and aspects of the appellant’s own claim describing the investigation and prosecution of various crimes – brought the Tribunal to the conclusion that there was no basis arising from the appellant’s individual situation to conclude that the state protection generally available in Georgia would not apply to him. Consequently, “adequate” state protection would be available should the appellant be returned to Georgia.

27. The analysis of the appellant’s claim for subsidiary protection follows the same pattern, albeit that it is much truncated as it is dealt with by reference to the earlier findings in respect of persecution. At paragraph 7.3 the Tribunal accepted that substantial grounds had been shown that the appellant faced a real risk of serious harm at the hands of the criminals who attacked and threatened him previously. However, ultimately it concluded that there was effective state protection available should the appellant be returned to Georgia.

The N.G. Decision

28. As it was agreed by the parties that Z.M. would be treated as the lead case, it is not necessary to look at the facts of N.G. in the same level of detail. Fortunately, the facts are also much more straightforward and cover a much shorter time span.

29. As an 18-year-old the appellant was hired by A.S. to work at a restaurant which was apparently owned jointly by A.S. and A.K. Two years later in September 2017 there was a falling out between A.S. and A.K. when A.K. accused A.S. of stealing money from the business. There was also some personal issue involved as A.K.’s sister was in a relationship with A.S. at the time. A.K. and two other employees in the business then accused the appellant of having been given the stolen money by A.S. and made repeated threats to kill him if he did not return it. The appellant did not have the money and believes he may have been framed by A.S. The appellant did not return to work because he was scared. A.S. was murdered in December 2017 and A.K. was subsequently arrested for the murder and imprisoned. The appellant did not know if A.K. was convicted or whether he was still in prison. The appellant continued to receive threatening phone calls from the two other employees seeking the return of the missing money. In addition, a threat was made to the appellant's father.

30. The appellant did not report the matter to the Albanian police as he believed the Albanian police to be corrupt, and he did not trust them. He left Albania in June 2018 and travelled to Ireland via Greece - where he spent a year - Italy, and Belgium - where he also spent a year.

31. The appellant was interviewed by the IPO in September 2021. On 2nd February 2022 the IPO issued a recommendation under section 39(3) of the 2015 Act that the appellant's claim for protection be refused. Because the appellant had failed to mention a number of matters that he relied on at interview in the questionnaire comprising his written application and in a lengthy submission made on his behalf by his solicitor and gave only a vague explanation for this failure, his credibility was not accepted. His failure to apply for asylum in Greece, Italy or Belgium also undermined his claim such that the only facts accepted were the appellant's nationality, age, marital status and religion. As these facts did not give rise to an objective basis for fearing persecution or serious harm no analysis was carried out regarding the availability of state protection.

32. The appellant appealed to the Tribunal and had an online oral hearing of his appeal on 25th July 2022. The decision refusing his appeal issued on 3rd October 2022. As previously mentioned, the decision is structured in the same manner as the Z.M. decision. The Tribunal took a different view to the IPO regarding the basic credibility of the appellant’s claim and held (at paragraph 4.3) that as he had given a “broadly consistent and coherent account of his claim” throughout the process and as the claim did not run counter to relevant country of origin information, the benefit of the doubt should be extended to the appellant. Consequently, the material facts of the claim were accepted including the fact that the appellant had been threatened by work colleagues and accused of taking money he had not taken, and that A.S. was murdered by A.K. It was also accepted that objectively these facts meant there was a real chance that if returned to Albania the appellant would face a wellfounded fear of persecution or a real risk of serious harm at the hands of the two work colleagues who had threatened him. Again, there was no nexus with any Convention ground and, although the Tribunal proceeded to consider whether state protection was available to the appellant in Albania, this analysis was ultimately relevant to the question of whether the appellant was entitled to subsidiary protection.

33. The Tribunal found that adequate state protection would be available to the appellant in Albania. In reaching this conclusion it considered the country of origin information regarding policing and prosecution in Albania emanating from the US State Department and Freedom House. A number of problems were noted in this information (including corruption within the police force) which the Tribunal regarded as presenting “a mixed picture” but noted that ongoing steps were being taken by the Albanian government to address these various difficulties. The Tribunal also took account of the designation of Albania as a safe country of origin under S.I. 121 of 2018. A significant weakness in the appellant's case was that he had never reported the threats to the Albanian police so it was not possible to gauge what their reaction might have been had he done so. Further, the person the appellant regarded as the main criminal element involved in the threats against him, namely A.K., had been arrested, prosecuted and imprisoned on various occasions, including significantly in respect of the murder of A.S. At his oral hearing the appellant accepted that this was broadly in line with what would be expected in a country with a functioning police service. Consequently, the Tribunal was satisfied that reasonable steps were taken by the authorities in Albania to prevent the kind of persecution or serious harm at the centre of the appellant’s claim.

Definition of Those Entitled to International Protection

34. The questions certified by the High Court judges for the purposes of these appeals focus on the terms “well-founded fear of persecution” and “real risk of serious harm” and the relationship between these terms and the concept of state protection. Essentially, the appellants’ case is premised on the contention that in order for a fear of persecution to be well-founded or a risk of serious harm to be real the decision maker must have already concluded that there was an absence of effective state protection against that persecution or harm.

35. In this regard the appellants relied on statements by the CJEU in Abdulla v. Bundesrepublik Deutschland Cases C-175/08, C-176-08, C-178/08, C-179/08 at paragraphs

57 and 58 to the effect that the existence of circumstances giving rise to a well-founded fear of persecution “will indicate that the third country does not protect its nationals against acts of persecution”. The court put it even further at para. 57 of Secretary of State for the Home Department v. O.A. C-255/19 stating that the existence of such circumstances establishes rather than merely indicating that the third country does not protect its nationals against persecution. It concluded that “a third country national who is in fact protected against acts of persecution … cannot, for that reason, be regarded as having a well-founded fear of persecution”.

36. There is undoubtedly a close link between the well-foundedness of the fear of persecution or the reality of the risk of serious harm and the availability of state protection against that persecution or harm. In order to understand the argument made by the appellants it is necessary to look at the definitions under which a person may qualify for international protection and the legislative provisions governing state protection. These are found primarily in the Qualification Directive 2004/83/EC. It should be appreciated that at an EU level a number of different directives govern different aspects of the law in relation to international protection. The Qualification Directive deals with the minimum standards to be met in order to qualify as a refugee or to be entitled to subsidiary protection and the content of the protection thereby granted. A separate directive, the Procedures Directive 2005/85/EC, governs the minimum standards in member states for the procedures for the grant or withdrawal of refugee status. The designation of a third country as a safe country of origin is governed by the Procedures Directive rather than the Qualification Directive. Effect is given to both of these directives in Ireland by the 2015 Act. The Qualification Directive was recast in 2011 (2011/95/EU) and the Procedures Directive in 2013 (2013/32/EU) but Ireland has not opted in to these later version of the directives.

37. Article 2 of the Qualification Direction contains at paragraph (c) a definition of “refugee” and at paragraph (e) of a “person eligible for subsidiary protection”. The relevant parts of these definitions are as follows: – “(c) ‘refugee’ means a third country national who, owing to a well-founded fear of being persecuted … is outside the country of nationality and is unable or, owing to such fear, is unwilling to avail himself or herself of the protection of that country …:

(e) ‘person eligible for subsidiary protection’ means a third country national … who does not qualify as a refugee but in respect of whom substantial grounds have been shown for believing that the person concerned, if returned to his or her country of origin, … would face a real risk of suffering serious harm as defined in Article 15 … and is unable, or, owing to such risk, unwilling to avail himself or herself of the protection of that country;”. The definitions are in fact more complex in that a refugee must fear persecution for a reason specified in the Geneva Convention and in order to qualify for subsidiary protection, the serious harm must be of a type detailed in Article 15. However, these elements of the definitions are not relevant to the arguments made on this appeal. The definitions of refugee and a person eligible for subsidiary protection in the interpretation section of the 2015 Act (section 2) are identical as regards the relevant elements. It is important to appreciate that the definition of refugee in Article 2(c) of the Qualification Directive is lifted verbatim from Article 1A(2) of the Geneva Convention. Subsidiary protection on the other hand is an EU law construct recognising the protection afforded by member states to persons who do not qualify as refugees but are nonetheless in need of international protection. While the test to be met to be eligible for subsidiary protection is materially different to that for refugee status, the two definitions are structured in an identical manner. There must be a well-founded fear of persecution or a real risk of serious harm and the person seeking protection must be unable or, because of the fear of persecution or risk of harm respectively, unwilling to avail of the protection of their own country.

38. On the face of it, these definitions each contain two elements both of which must be satisfied in order for a person to qualify for international protection. Firstly, the person must have a well-founded fear of persecution or face a real risk of serious harm if returned to the country of origin. Secondly, they must be unable or, because of the fear or risk, unwilling to avail of the protection of that country. As the respondent correctly pointed out, as a fundamental principle it is presumed that a state can and will protect its own nationals. International protection is a substitute for state protection and host countries are only obliged to provide international protection where, contrary to the general principle, a person is unable to avail of the protection of their own country.

Actors of Persecution and Actors of Protection

39. Article 2(c) and (e) of the Qualification Directive is silent as to the potential source of the persecution or serious harm falling within the scope of the definition of a refugee or a person eligible for subsidiary protection but the protection of which the person is unable or unwilling to avail is that of the country from which the applicant for protection comes. Where the authorities in that state are also responsible for the acts constituting persecution or serious harm, it is relatively straightforward to conclude that the protection of the state will not be available to the person concerned. The position is more complex where the persecution or serious harm emanates from a source other than the state itself. These concepts are fleshed out in Articles 6 and 7 of the Qualification Directive which deal with the actors of persecution or serious harm and actors of protection respectively. These articles provide as follows: - “Article 6: Actors of persecution or serious harm Actors of persecution or serious harm include:

(a) the State;

(b) parties or organisations controlling the State or a substantial part of the territory of the State;

(c) non-State actors, if it can be demonstrated that the actors mentioned in (a) and (b), including international organisations, are unable or unwilling to provide protection against persecution or serious harm as defined in Article 7.

Article 7: Actors of protection

1. Protection can be provided by:

(a) the State; or

(b) parties or organisations, including international organisations, controlling the State or a substantial part of the territory of the State.

2. Protection is generally provided when the actors mentioned in paragraph 1 take reasonable steps to prevent the persecution or suffering of serious harm, inter alia, by operating an effective legal system for the detection, prosecution and punishment of acts constituting persecution or serious harm, and the applicant has access to such protection.” Article 7(3) is not relevant for present purposes.

40. Articles 6 and 7 of the Qualification Directive are implemented in Irish law by sections

30 and 31 of the 2015 Act. Section 30 is a virtually verbatim reproduction of Article 6. The respondents note that section 31 of the 2015 Act is framed slightly differently to Article 7 of the Qualification Directive. Amongst other differences it includes elements deriving from the CJEU’s interpretation of these provisions. Thus, section 31(2)(a) requires the protection be “effective and of a non-temporary nature”, something which is not expressly set out in Article 7 but which the CJEU has held to be necessary in order for state protection to be regarded as available.

41. The cases before the court do not raise any issue concerning protection provided by an international organisation nor the control of a substantial part of the territory of either Georgia or Albania so these elements of the provisions can be disregarded. Nor was it contended that the threats which gave rise to the risk of serious harm emanate from the state or from any state agent. The cases concern a risk of serious harm at the hands of non-state actors, in both cases from criminal elements. Thus, under Article 6(c) in order to establish eligibility for subsidiary protection, the states in question had to be shown to be either unable or unwilling to provide protection against the particular serious harm, i.e. personalised threats from criminal elements.

42. The use of “unable or unwilling” mirrors the use of the same phrase in the definitions in Article 2(c) and (e) save that in Article 2 it refers to an applicant’s inability or unwillingness to rely on the protection of their country, and in Article 6(c) to the inability or unwillingness of the country to provide protection. The standard of protection to be provided under Article 6(c) is expressly stipulated to be that defined in Article 7 – i.e. the taking of reasonable steps to prevent persecution or serious harm. As we shall see the CJEU has made it clear that the same standard applies when considering the availability of state protection under Article 2(c) or (e).

43. As the appeals do not concern an international organisation or those in control of a substantial part of either Georgia or Albania, the issue under Article 7 is whether those states, being actors of protection under Article 7(1)(a), can provide the level of protection required under Article 7(2). It is important to note that Article 7(2) is phrased in terms of protection being “generally provided” when the three elements therein listed are met. The three elements are the taking of reasonable steps to prevent persecution or serious harm; the operation of an effective legal system for the investigation, prosecution and punishment of acts of persecution or serious harm and that that protection is available to the applicant. A

decision maker must nonetheless be live to the possibility that in exceptional cases, even where these elements are met, effective protection might nonetheless not be available to the particular applicant.

44. Although the availability of state protection and an applicant's ability or willingness to avail of it is an issue in every claim for international protection, it assumes particular importance in cases like these where the persecution or harm feared is at the hands of nonstate actors. If a person is suffering persecution at the hands of the authorities in their own country, it is relatively easy to conclude that effective state protection is not available to that applicant from those same authorities. Even where different authorities are involved (for example someone fears persecution at the hands of a local government official and claims that the local police will not act to prevent it) the links between the authorities in many countries are such that even if such a conclusion is not inevitable, it is still a fairly easy one to reach.

45. This is not so where the threat of persecution is from a non-state actor. There is no intrinsic reason to suppose that any country will simply refuse to provide protection to someone who faces persecution or serious harm at the hands of an entity or person for whom the state authorities are not responsible. Therefore, an applicant will usually have to show that the state is unwilling to provide protection to the applicant personally or that it is unable to provide protection although otherwise willing to do so. This is not of course a hard and fast rule, and each case must be examined individually. As we shall see, in neither of these cases does the appellant suggest that the state is unwilling to provide protection to them personally. Instead the arguments made are that the decision maker effectively concluded that state protection was absent when finding that the fear of persecution was well-founded or the risk of serious harm was real; that the presumption or evidential advantage in section 28(6) should have been and was not applied to the consideration of whether state protection was available and, in any event, that the level of protection available is not sufficient.

State Protection – Interpretation of Article 2 of the Qualification Directive

46. There is no dispute between the parties but that international protection need only be provided to the appellants if the protection of their own state is not available to them. The issue is whether the risk of serious harm which in each case was found to have been established, could have been properly found to be a “real risk” without the absence of state protection forming part of that finding.

47. In Z.M. Hyland J. rejected this argument on the basis that the same argument had been rejected by the High Court on two previous occasions by Burns J. in B.A. v. IPAT [2020] IEHC 589 and by Heslin J. in T.A. v. IPO [2023] IEHC 390.

48. In B.A., in a decision structured in a similar manner to these ones, the Tribunal accepted that the applicant, a woman, had a well-founded fear of persecution on the basis of the threat of gender-based violence from her ex-partner if she were returned to her country of origin. However, it went on to conclude that state protection would be available to her. Burns J. did not accept that the finding of a well-founded fear of persecution by the decision maker necessitated a grant of refugee status. She held at paragraph 14 of her decision: - “I cannot accept that interpretation of this finding of the First Respondent and do not follow the logic of the argument. Considering the decision as a whole, it is clear that what the First Respondent meant to convey is that in the absence of a finding by the First Respondent that state protection was available to her, the Applicant had a well-founded fear of persecution from her ex-partner. The First Respondent ordered its decision-making process in a logical fashion: initially determining whether a well-founded fear of the Applicant being subjected to serious harm by her ex-partner existed, and then determining, on foot of the positive finding that it did exist, the question of whether state protection was available.”

49. Heslin J. expanded this analysis when dealing with an argument identical to that made in these appeals. T.A. was an Algerian national who was threatened by an uncle in the context of a family feud. The IPO had accepted the genuineness of his fear of persecution but held that state protection was available to him. The primary reason he failed in his judicial review was because he had not exercised his right of appeal to IPAT and thus, had not exhausted the remedies available to him. This may explain why no appeal was pursued from Heslin J.’s judgment.

50. On the issue of concern here, Heslin J., commencing at paragraph 62 of his judgment, held as follows: - “62. Considerable emphasis was laid by the Applicant on the use of the words “intrinsically linked”, as they appear in para. 56 of the decision of the Court of Justice in OA. However, to say that concepts are intrinsically linked is not equivalent to saying that they are the same concept. Central to the Applicant’s case is that (i) fear of persecution / real risk of serious harm; and (ii) State protection are the same, not different concepts, for the purposes of the 2015 Act and the Directive. Respectfully, I cannot agree. 63.Well–founded fear is certainly an element of the definition of refugee status. However, it seems to me that there is a clear distinction between the concept of well– founded fear and the entitlement to refugee status (whereas the applicant contends that a finding of the former automatically gives rise to the latter). Based on a literal interpretation of the words used (in s.2 of the 2015 Act/Article 2 of the Qualification Directive), it seems to me that State protection is an element of, but distinct from, the definition of refugee.

…

65. It seems to me that the elements of the refugee definition which are linked by the conjunction “and” remain distinct concepts, albeit in the context of a single definition.

66. In other words, the concepts of (i) fear of persecution and (ii) protection are certainly “intrinsically linked” (to cite OA), but this is a close linking of distinct elements making up a unitary definition. In the impugned decision, both these elements were considered by the IPO and I cannot accept that it is permissible for this Court to take issue with the clear and logical approach taken by the IPO when carrying out this consideration.”

51. In N.G. Phelan J. considered herself bound by the decision in T.A. as she had not been persuaded that there was any proper basis for departing from the reasoned conclusions of Heslin J. She regarded it as necessary for the decision-maker to consider as separate elements “both whether the kind of harm apprehended would constitute persecution or serious harm were it to come to pass and also whether the State is safe and there exists effective State protection”. In concluding on this issue, she held at paragraph 20 of judgment: - “The question of State protection only becomes a relevant consideration where a threshold risk of harm (be it persecution or serious harm) capable of triggering an entitlement to protection is found to exist. Furthermore, the nature of the apprehended risk requires to be established in order to properly assess whether effective State protection exists or might reasonably be accessed by an applicant as whether such protection exists may well depend on the particular risk established.”

52. In my view, the observation that it is necessary to consider the nature of the risk that has been found to exist in order to assess whether state protection is available in respect of that particular risk is well made. In that context Phelan J. accepted that in holding the fear to be objectively well-founded, the Tribunal meant only that a threshold risk of harm had been established, and the fear of that harm was reasonably held by the applicant. Proceeding to consider the availability of state protection following such a finding was not a departure from the correct legal test.

53. The appellants face an immediate difficulty because in each case it is clear from the

decision itself that the Tribunal member did not in fact consider the availability of state protection before concluding that the risk of serious harm was real. Instead, he moved through a series of steps which involved analysing the claim to ascertain the material facts relied on by the appellants; analysing each of those facts to ascertain whether they were made out on the evidence; considering whether the established facts were capable of grounding a fear of persecution or a risk of serious harm and then whether there was an objective basis to that fear or risk. At this point the Tribunal member concluded in each case that the fear of persecution was well-founded and the risk of serious harm was real. He then found that there was no Convention nexus in either case, thus excluding the possibility of a grant of refugee status. Only then did he move on to consider state protection. In each case he found it to be available, as a result of which the claims were refused. A similar approach was described by Burns J. in B.A. as “logical”.

54. Further, the approach taken by the Tribunal mirrors the text of the definition in requiring that both a real risk of serious harm and the absence of state protection be shown. The appellants effectively contend that they are entitled to a grant of subsidiary protection by virtue of the conclusion that there was a real risk of serious harm which, on their argument necessarily entailed a finding of an absence of state protection, notwithstanding that in these decisions that conclusion manifestly had not taken account of the availability or nonavailability of state protection.

55. The appellants assert that their interpretation of the terms ‘well-founded fear of persecution’ and ‘real risk of serious harm’ accords with the technical meaning of those terms as used in the Qualification Directive and that the respondents’ interpretation relies on the way in which these terms were used in the decisions under review. In fairness to the respondents, they do not rely on the decisions as establishing that the correct meaning of the terms but rather contend that the terms were correctly applied by the decision-maker. I find it difficult to accept the argument that the term “well-founded fear of persecution” as used in the Qualification Directive encompasses the absence of state protection when the definition of refugee under the Directive requires that there be both a well-founded fear of persecution and an absence of state protection. The definition is framed as comprising two elements and satisfying both of them entitles a person to a declaration that he is a refugee. There is nothing in the text of the provision to suggest that it should be assumed that the second limb is automatically satisfied merely because the first has been shown to have been established. Rather, as Heslin J. put it, the legislation contains a unitary definition which is made up of closely linked but discrete elements.

56. The appellants acknowledge that in order to succeed in their appeals it is necessary that this court overturn not only the two judgments from which the appeals are taken but also the entire line of High Court authority reflected in the decisions in B.A. and T.A. They say that this is required in light of the decision of the CJEU and, even more so the opinion of Advocate General Hogan in the Secretary of State for the Home Department v. O.A. C- 255/19. O.A. was a Somali national who had been granted asylum in the UK in the early 2000s on the grounds that as a member of a minority clan he had suffered persecution at the hands of a militia. In 2016 O.A.’s refugee status was revoked due to what the Secretary of State characterised as a durable change in circumstances in Somalia such that in O.A.’s home area minority clans were no longer persecuted by majority clans and effective state protection was available.

57. Thus, the case concerned the cessation provisions of Article 11(1)(e) of the Qualification Directive which provided that a person would cease to be a refugee if because “the circumstances in connection with which he or she has been recognised as a refugee have ceased to exist” he “can no longer … continue to refuse to avail … of the protection of the country of nationality”. The fact that O.A. concerned cessation provisions is a potentially relevant distinction for two reasons. The decision maker was not trying to decide if O.A. met the definition of a “refugee”; it had been decided some years earlier that he did. Rather, the

decision maker had to decide if, because of the change in circumstances in Somalia, O.A. remained entitled to refuse to avail of the protection of the country from which he had previously fled. The relevance of the case to this decision is further obscured because O.A. successfully contended that in concluding that state protection was available, the decision to revoke his status had incorrectly taken account of the availability of support from non-state actors (family and fellow clan members) instead of looking only at the level of support and protection available from state authorities. That issue does not arise here.

58. The questions referred to the CJEU concerned whether the protection of a refugee’s country of nationality under Article 11(1)(e) was the state protection referred to in Article 2(c), i.e. in the definition of a refugee, such that the criteria to be applied to the Article 11 assessment was the same as those under Article 2(c) and Article 7. A question was also raised (the second question) as to whether the “protection test” must be applied to both elements of the Article 2(c) definition – i.e. the existence of a well-founded fear of persecution and the absence of state protection. The formal answer provided by the court did not expressly address the question concerning Article 2(c). Instead, it ruled that the protection referred to in Article 11(1)(e) regarding the cessation of refugee status was the same as that arising under Article 2(c) read with Article 7 on the grant of such status. It also concluded that social and financial support provided by clan members was of no relevance to the availability or effectiveness of state protection under Article 11 or Article 2.

59. Insofar as the question specifically querying the application of the “protection test” to both limbs of the Article 2(c) definition was addressed, it was rolled by the court into a consideration of the first three questions which were collectively summarised as asking whether such social and financial support provided by private actors could constitute or be relevant to state protection under Article 11, 7 or 2.

60. The significance of all this is that neither the court nor the Advocate General in O.A. was purporting to expressly consider the sequence in which the various elements of Article 2(c) had to be considered when the initial decision to grant or refuse refugee status was made. The observations in O.A. must be viewed in the context of the circumstances of that case and the questions the court was addressing.

61. That said, the passages in which the appellants rely on are found firstly in the opinion of Advocate General Hogan which, unlike the judgment of the CJEU, does address the second question separately. Interestingly he notes (at footnote 25 of his opinion) that the UK government had argued that the need to consider state protection arose at three stages – in considering whether the fear of persecution was well-founded; in considering whether state or non-state actors could provide effective protection against persecution by non-state actors and in considering whether the applicant was unable or unwilling to avail of such protection. He also noted the divergent approaches applied by different courts within the UK. In the circumstances it is perhaps unsurprising that in paragraph 54 of his opinion Advocate General Hogan describes the analysis as ‘overcomplicated’.

62. The paragraphs relied on by the appellant are as follows: -

“54. One can, I think, rather overcomplicate an analysis of what, in the end, is a single concept which accordingly applies identical criteria in the context of the application of both Article 2(c) and Article 7 of the Qualification Directive and, indeed, in turn Article 11(1)(e) of that directive.

55. In examining any application for refugee status, the question must always be whether an applicant has established a well-founded fear of persecution pursuant to Article 2(c) of the Qualification Directive. The use of the term ‘well-founded’ fear in the definition of ‘refugee’ in Article 2(c) of the Qualification Directive requires, inter alia, an analysis of whether the conditions in the applicant’s country of nationality or origin are such as to objectively justify the applicant’s fear of persecution.

56. This test will necessarily require, in my view, an objective examination of whether or not there is protection in the applicant’s country of nationality by actors of protection as defined in Article 7 of the Qualification Directive against persecution and whether the applicant has access to that protection.

57. I therefore agree, in essence, with the Commission’s observation that refugee status falls to be determined by reference to a single protection test which meets the requirements laid down in Article 7 of the Qualification Directive. I would stress, however, that protection in the country of nationality must be available against all actors of persecution as defined by Article 6 of the Qualification Directive.

58. While there is, strictly speaking, no formal definition of ‘protection’ in Article 2 of the Qualification Directive, protection is in fact described in Article 7(2) of the Qualification Directive. …

59. Thus, the continued necessity for international protection (refugee status) in a case such as that in the main proceedings is determined, inter alia, by the ability or otherwise of an actor of protection to take reasonable steps to prevent the persecution of the applicant at the hand of non-State actors by, inter alia, operating an effective legal system for the detection, prosecution and punishment of such acts by, inter alia, non-State actors.”

63. It is important to understand the reason these paragraphs are phrased the way they are. The “single protection test” is referred to in the context of an alternative analysis applied by some UK decision-makers which literally split the definition of a refugee into two elements but also “applied a protection test to the first element that is distinct – and more restrictive – of the protection test under Article 7” (see footnote 31). The protection test that must be met under Article 2(c) is that prescribed by Article 7(2) and no lower standard – such as the social and financial support of fellow clan members – will suffice. Further, the concept of state protection is a single concept which requires identical criteria to be applied to the assessment under both Article 2(c) and (e) and Article 11(1)(e). Something cannot be regarded as effective state protection for the purposes of the cessation of refugee status under Article 11(1)(e) if the same level of protection from the same source would not have sufficed to undermine the well-foundedness of fear of persecution when granting protection. The point in the sequence at which the issue of state protection is considered under Article 2(c) was immaterial in this context.

64. As noted, the CJEU did not separately address the question as to whether the same “protection test” applied to both limbs of the definition of a refugee, the question having been raised because the jurisprudence of some courts in the UK suggested that a narrower test applied when assessing the well-foundedness of the fear under Article 2(c) which might thereafter preclude a broader assessment of the availability of state protection under Article

7. In addressing the first three questions together, the CJEU made the observation at paragraph 57 of its judgment which has already been set out at paragraph 35 above. It describes the “conditions specified in Article 2(c)” being the fear of persecution and protection as “intrinsically linked” in that the protection in question is protection from acts of persecution. It went on to state: - “59. Consequently, for the purposes of determining whether the third country national concerned has a well-founded fear of persecution in his or her country of origin, within the meaning of Article 2(c) of Directive 2004/83, account must be taken of whether there is or is not protection from acts of persecution in that third country.

60. However, the existence of protection from acts of persecution in a third country can permit the inference that there is no well-founded fear of persecution within the meaning of that provision only if that protection satisfies the requirements arising, in particular, from Article 7(2) of that directive.

61. Since the conditions specified in Article 2(c) of that directive dealing with fear of persecution and protection from acts of persecution are, as is stated in paragraph

56 of the present judgment, intrinsically linked, their examination cannot be subject to a separate criterion of protection; their assessment must be made in the light of the requirements laid down in, inter alia, Article 7(2) of that directive.”

65. Again, the purpose of these passages is to emphasise that the standard of protection required under Article 2(c) is that stipulated by Article 7(2). There are not two different protection requirements subject to different criterion. That standard also applies to the protection issue on cessation under Article 11(1)(e). The judgment is not addressing the sequence in which a decision-maker can or should consider various matters arising under Article 2(c), which, as I have previously noted, also involves matters such as whether the fear of persecution is for a Convention reason, which are not the subject of this judgment.

66. At para. 57 of the judgment of O.A. the CJEU cites from its earlier judgment in Abdulla (above) on which the appellants also rely. The respondents argue that in O.A. the CJEU misquotes its judgment in Abdulla. Certainly, as I have noted at para. 35 above, the statement at paragraph 57 of O.A. that the existence of a well-founded fear of persecution establishes that the country in question does not protect its nationals against persecution goes further than the equivalent statement in Abdulla that such circumstances amount to an indication that the third country does not protect its nationals against acts of persecution (at paragraph 58). However, it would not be appropriate for this court to assume that the CJEU had made an error in O.A. and misunderstood the scope or import of its own judgment in Abdulla. That said, I think the point has less significance than the appellants attach to it in the context of these proceedings.

67. Abdulla is ostensibly distinguishable from this case not just because it is a cessation case but because the fear in issue concerned persecution by state actors. However, the appellants rely on Abdulla as a statement of principle and note the application of that principal in O.A. to circumstances which involved persecution by non-state actors. Consequently, they argue that the principle is applicable to these appeals.

68. Abdulla involved Iraqi nationals who had been granted asylum in Germany. The German authorities subsequently sought to revoke their refugee status on the basis of a change of regime in Iraq. The claimants contend that although the persecution which they had originally feared had ceased to exist, they now feared different persecution from the new regime. The questions referred to the CJEU concern the relationship between state protection under Article 7 and under Article 11(1)(e). A question was also referred relating to the standard of proof to be applied when considering whether new and different circumstances were capable of giving rise to a well-founded fear of persecution for somebody whose original grant of refugee status had ceased and whether the standard to be applied was the same as that originally applied under Article 2(c). The response to that question has some relevance to the next issue to be considered regarding Article 4(4).

69. The analysis of the CJEU in Abdulla commences with the observation that the existence of circumstances in a person's country of origin which gave rise to a well-founded fear of being personally the subject of persecution will indicate that the country in question does not protect its nationals against acts of persecution (paragraph 58) and that those circumstances in turn form the reason which make it impossible for the person to avail of the protection of that country (paragraph 59). These are then described as the “determinant factors in respect of the granting of refugee status” (paragraph 60). In the context of cessation, paragraph 66 describes the wording of Article 11(1)(e) as establishing a “causal connection” between the change in the circumstances which originally gave rise to the fear of persecution and the question whether a person can continue to refuse to avail of the protection of their country of nationality.

70. Notwithstanding the appellants’ reliance on these judgments, it is difficult to see why the fact that there is a causal connection between the circumstances giving rise to a fear of persecution (or risk of serious harm) and the availability of state protection, or that these two factors are intrinsically linked, requires that the absence of state protection be established before a finding can be made that there is a well-founded fear of persecution or a real risk of serious harm. In reality, the appellants are asking the court to assume an absence of state protection from the finding that the fear of persecution was well-founded/ there was a real risk of serious harm in the face of the analysis contained in the decisions themselves. The case law makes it clear that the Tribunal was required to consider both the question of serious harm and that of state protection before deciding whether the appellants were eligible for subsidiary protection. In circumstances where the issue of state protection was expressly addressed and the Tribunal concluded that it was available to the appellants, it would be perverse for this court to hold that in finding that a real risk of serious harm had been established the Tribunal had implicitly concluded that state protection was not available.

71. Equally, I am not prepared to hold that the decision was fundamentally flawed because the Tribunal concluded that a real risk of serious harm had been established before addressing the availability or non-availability of state protection. As Phelan J. points out, it is necessary to examine the question of persecution/serious harm first because the analysis of whether state protection is available may depend on the nature of the persecution/serious harm from which protection is needed. I accept that there may be an element of circularity to the analysis involved in that an otherwise well-founded fear of persecution or real risk of serious harm will not suffice to qualify somebody as a refugee or entitled to subsidiary protection if adequate state protection is available in an applicant’s country of origin against that persecution or risk of harm. Nonetheless, the Tribunal member proceeded to consider the appeal in an entirely logical manner consistent with the language of the Qualification Directive as reflected in the 2015 Act.

72. Before concluding that a person is not entitled to international protection, or in the case of cessation, is not entitled to retain status already granted, the decision maker must address each of the elements of the definition of refugee or person entitled to subsidiary protection. Where the decision involves an analysis of the availability of state protection against the actions of non-state actors, the “protection test”, which is that prescribed in Article 7(2), must be applied and not any different or lower standard that might be derived from reading Article 2(c) or (e) divorced from the language of Article 7(2). Although the appellants contend that the Tribunal did not apply the test for state protection under Article 7(2) correctly and that the protection of their countries of origin was not in fact available to them (a point to which I will return), they do not contend that the Tribunal purported to apply any different test other than that under Article 7(2). In all of the circumstances this ground of appeal must be rejected.

73. Finally, as this judgment was being prepared my attention was brought to a recent

decision of the Supreme Court in the case of B.D. and Ors v. IPAT (O’Malley J. [2025] IESC 38). The Tribunal had found that the applicants in that case had shown a well-founded fear of persecution but refused to grant them protection on the basis that they were able to avail of state protection in their country of origin (South Africa). Both parties accepted that the

decision regarding the availability of state protection was flawed. The issue on the appeal was whether the Tribunal’s decision was severable such that only the portion which found state protection to be available should be quashed and the case remitted with the benefit of the finding in favour of the applicants on the persecution issue remaining intact or, alternatively, whether the entire decision should be quashed and the matter remitted in full to the Tribunal. The High Court had quashed the entire decision; the Court of Appeal reversed this judgment, severed the state protection element of the Tribunal's decision and quashed only that part. The Supreme Court reversed the Court of Appeal and held that the entire decision should be quashed. Obviously, the question of severance does not arise in these cases unless the appellants were to succeed in their appeals but the decision is nonetheless potentially relevant insofar as it deals with the timing of the conclusions drawn by the decision maker on persecution and on the availability of state protection.

74. The Supreme Court judgment acknowledges (at paragraph 87) that the question of fear of persecution is inextricably entwined with the question of state protection and that the two elements must be assessed by reference to each other. It also examines the decision of the CJEU the X v. IPAT C-756/21 which held that a decision on an application for asylum should be made in two separate stages. The first involves establishing the factual circumstances and considering the evidence supporting the application and the second involves a legal appraisal of that evidence. This relates primarily to the allocation of the burden of proof in international protection applications. The onus is on an applicant for protection to provide the evidence necessary to substantiate the claim, but the appraisal of that evidence must be carried out by the decision maker in cooperation with the applicant. Neither X v. IPAT nor B.D. itself addresses the sequencing of the issues to be determined by the decision maker on the basis of the established factual circumstances.

75. O’Malley J. acknowledged, in accordance with established jurisprudence, that on judicial review in principle severance and partial remittal of different parts of a decision is possible. However, whether that is appropriate in an individual case may depend on the extent to which the invalid part can be properly re-analysed without taking into account the elements involved in the part found to be valid (para.84). The asylum decision in issue involved the assessment of two factors, the well-foundedness of the fear of persecution and the availability of state protection (para.85). She concluded that “the relationship between the two elements to be assessed is too close to separate them in the fashion proposed…” (para.86). Crucially, there is a temporal aspect to a protection decision in that both central elements have to be present at the time the decision is being made. A protection decision is not entirely historical as the decision maker is also concerned with present conditions and the extent to which they provide a basis for either a well-founded fear or an absence of state protection (para.88).

76. The connection identified by O’Malley J. between the two elements of a protection

decision and the need for those elements to be assessed simultaneously on the basis of current and up-to-date information is not determinative of the issue raised by these appellants. Reverting to the observation of Phelan J. in N.G., the state protection to be considered under Article 7(2) of the Qualification Directive, is protection from persecution or serious harm which has been established in respect of the particular applicant. A delay between making a decision on the fear/serious harm issue and a decision on the state protection issue may result in the availability of state protection being considered in respect of a risk of persecution or serious harm which no longer exists, or which no longer exists in the form in which it did when its existence was taken as established. There is nothing in B.D. which suggests that the decision maker cannot conclude that there is a real risk of serious harm without having also concluded that there is an absence of state protection, particularly when the decision is obviously structured in a way which examines these issues sequentially. What B.D. does say is that the analysis of these issues forms part of a single decision-making process which cannot normally be separated so that the different elements are considered in isolation and at a temporal distance from each other.

Article 4(4) – Past Persecution as a Serious Indicator of Future Persecution

77. The purpose of the argument with which I have just dealt in some detail is difficult to understand divorced from its role in providing the basis for the appellants’ second argument. The second argument concerns the application of the presumption or evidential advantage under Article 4(4) of the Qualification Directive and section 28(6) of the 2015 Act. These provisions are identical and provide as follows: - “The fact that an applicant has already been subject to persecution or serious harm, or to direct threats of such persecution or such serious harm, is a serious indication of the applicant’s well-founded fear of persecution or real risk of suffering serious harm, unless there are good reasons to consider that such persecution or serious harm will not be repeated.”

78. The appellants acknowledge that Article 6(c) of the Qualification Directive places an onus of proof on an applicant to demonstrate that the state is unable or unwilling to provide protection where persecution is alleged at the hands of non-state actors. This is consistent with the general presumption that state protection is available in such circumstances and with the onus lying on an applicant for protection to rebut that presumption.

79. In both of these cases the Tribunal found the threats to which the appellants had been subjected constituted persecution and serious harm. In Z.M. the Tribunal expressly held that there was a reasonable chance the appellant would face a well-founded fear of persecution at the hands of G.C. if returned to Georgia (paragraph 5.3) and similarly that there was a real risk of serious harm to the appellant in Georgia (paragraph 7.3). In essence, consistent with the application of Article 4(4), the Tribunal accepted that there was a risk of future persecution/serious harm based on its finding of past persecution/serious harm. A similar exercise was carried out and finding made in N.G. (at paragraph 5.2).

80. The appellants argued that the presumption or evidential advantage created by Article 4(4) should also have been, but was not, applied to the question of state protection. This, it is argued, occurred because the Tribunal impermissibly split the analysis of persecution/serious harm on the one hand and of protection on the other in the manner discussed in the previous section of this judgment. It is also claimed that this impermissible distinction is maintained in an affidavit sworn by an official of the Department of Justice and Equality for the purposes of opposing the application for leave to apply for judicial review in the Z.M. case. The deponent addressed each of the grounds raised by Z.M.. Paragraph 19 of the affidavit deals with this ground. It sets out the conclusions reached at paragraph 5.3 and 7.3 of the Tribunal decision and then proceeds to state: - “I say and believe therefore that the second Respondent has concluded that the Applicant has a well-founded fear of persecution of real risk of suffering serious harm as contemplated by section 28(6) and therefore this ground is misconceived.” The appellants’ counsel characterised this as the respondents treating the presumption as exhausted once the decision maker found that there was a well-founded fear of persecution/real risk of serious harm based on the acceptance of the appellants’ evidence regarding past persecution or serious harm, in fact more specifically here the direct threats to the appellants of persecution/serious harm. I am not persuaded that this characterisation is accurate. It is not so much a question of the presumption being exhausted as of ascertaining exactly what it requires the decision-maker to assume regarding the existence of a continuous or future situation based on proof of past events.

81. Alternatively, the appellants argue that the phrases ‘well-founded fear of persecution’ and a ‘real risk of serious harm’ are used in Article 4(4) as legislative shorthand for the entitlement to refugee status or subsidiary protection. I think this aspect of the appellants’ appeal can be rejected out of hand. If it had been the intention of the EU legislature that subjection to past persecution or serious harm would raise a presumption of entitlement to international protection, then it would no doubt have been expressly provided for in those terms. Given the definitions in Article 2, each of which involve a number of elements, it would be sloppy drafting to say the least to use one of these elements of shorthand for the entire definition. It would make as much sense to read Article 4(4) as meaning that if an applicant had already been the subject of persecution that this is a serious indicator that the persecution was for a Convention reason, something which simply does not logically follow from the existence of past persecution.

82. In her judgment Hyland J. said that it was difficult to understand how the applicant could contend section 28(6) had been breached when he was a beneficiary of a positive

decision on the question of well-founded fear of persecution as clearly, in those circumstances, the Tribunal had not treated the presumption as having been rebutted. Phelan

J. referred to her own previous decisions in N.U. v. IPAT [2022] IEHC 87 and E.S. v. IPAT [2022] IEHC 613 in holding that the finding of a future risk of harm based on previous occurrences reflected an application of section 28(6) and that it was not necessary for the

decision-maker to overtly apply the presumption when in any event the issue was determined in applicant's favour. Thus, N.G.'s application had failed not because the Tribunal concluded despite the rebuttable presumption under section 28(6), that the appellant had not demonstrated a risk of harm but because state protection was available to him in respect of that harm.

83. The appellants rely on three Irish cases in support of the argument that the Article 4(4) presumption applies to state protection. These are the decision of Burns J. in I.L. v. IPAT [2021] IEHC 106 and the two decisions of Phelan J. referred to in the proceedings paragraph namely N.U. v. IPAT and E.S. v. IPAT.

84. I.L. had been threatened by a politician, who was also a commercial rival, whom he had publicly criticised. His account of events in his country of origin was accepted by the Tribunal and it was accepted that the nature of the harm involved constituted persecution. Notwithstanding these findings, the Tribunal went on to conclude that the harm in issue did not reach the necessary level of severity and, specifically, that there was no clear evidence that I.L. would suffer harm on his return. Burns J. granted an order of certiorari on the basis that the Tribunal had not engaged “in any analysis of the rebuttable presumption which the Applicant was entitled to in light of its earlier finding that the Applicant had been subject to threats of serious harm”. She continued at paragraphs 13 and 14 of judgment: - “13. There was an obligation on the First Respondent to engage in an analysis of this rebuttable presumption which it failed to do. Indeed, there is no reference whatsoever by the First Respondent to s. 28(6). This is an error on the part of the First Respondent. Section 28(6) provides a significant evidential presumption to an applicant which can be rebutted by good reason. However, it should be unambiguous from the First Respondent's decision that such a significant evidential presumption was considered by the First Respondent and the good reasons which rebutted the presumption should be stated. …

14. The Respondent argues that good reasons did exist to rebut the presumption and that they are set out and apparent in the decision, although s. 28(6) is not specifically analysed. This is not sufficient to deal with this issue. As already stated, s. 28(6) is a significant evidential benefit which an applicant, who has been found to have been subjected to threats of serious harm, has. It is not appropriate that assumptions and inferences be made as to whether this issue had been considered by the First Respondent, and if so, what the good reasons were for determining that the presumption, which the Applicant is entitled to, has been rebutted.”

85. Later cases suggest that it is not necessary to expressly refer to section 28(6) provided it is clear that the evidential presumption is engaged with by the decision maker. Of course, I.L. was a case where the presumption was not applied, and the decision did not record any “good reason” why the decision-maker considered that the persecution/serious harm would not be repeated or indeed if that was the conclusion reached. Here the Tribunal found in both cases that there was a well-founded fear of persecution and a real risk of serious harm, apparently based on the findings of past persecution/serious harm. Thus, the presumption was applied and so there was no need to explain any “good reason” why it might not have applied.

86. In N.U. the applicant's claim was based on her fear of persecution by her ex-partner who had subjected her to extensive physical violence and the police in her country of origin had failed to take any meaningful action in response to her reported complaints. The Tribunal accepted significant elements of her claim and that she had a well-founded fear of persecution if she was returned but rejected her application on the basis of the availability of state protection.

87. Following I.L., Phelan J. concluded at paragraph 40 of judgment: - “40. Section 28(6) applies in this case to give a significant evidential benefit to the applicant but there is no evidence in the terms of the impugned decision that the Tribunal identified and considered good reasons for determining that the presumption of a future indicator of a risk of persecution had been rebutted (see IL v. IPAT …). I accept the applicant's submission that having found a fear of persecution to be established, there was an obligation on the first respondent to engage in an analysis of this rebuttable presumption under s. 28(6) and to explain the reasons why the Tribunal was satisfied that there was good reason that there would be no future risk, if this is indeed the Tribunal's Decision. No such analysis is carried out by the Tribunal in deciding that state protection would be available.”

88. In E.S. the applicant relied on this paragraph of the judgment in N.U. to argue that as the impugned decision did not refer to the presumption at any point and it did not mention section 28(6), this was evidence that it had not been properly applied. However, as in these cases, the Tribunal had not only accepted that the applicant had suffered past harm but also that he was at risk of future harm. Subsidiary protection was refused on the basis that state protection was available. Consequently, Phelan J. stated at paragraph 51 of her judgment: - “Where the Tribunal accepts that there is a future risk of harm based on previous occurrences, there can be no basis for contending that the Applicant has been deprived of the benefit of s. 28(6) of the 2015 Act. Ultimately this established risk (whether presumptively established and not rebutted or established on the basis of the material) is found to not give rise to a right to protection because State protection is available. The application does not fail because it is concluded, despite the rebuttable presumption under s. 28(6), that the Applicant has not demonstrated a risk of harm. Instead, it fails because it is concluded that State protection is available.”

89. The appellants argued that in the final sentence of paragraph 40 of her judgment in N.U. Phelan J. was applying the presumption under section 28(6) to state protection. I think this may be to read too much into a single sentence, particularly as Phelan J. had already identified a number of gaps in the Tribunal's decision and a number of elements of the applicant’s claim which were simply not dealt with or referred to. The appellants also mischaracterised E.S. as a change of position on Phelan J’s. part from her earlier decision in N.U. Both I.L. and N.U. are clear that where a finding of past persecution/serious harm is made and the presumption under section 28(6) is not applied so as to conclude that there is a well-founded fear of persecution or real risk of serious harm, the decision-maker is required to engage with the rebuttal of the presumption and explain the good reasons which have led to the conclusion that the harm/persecution will not be repeated. E.S. dealt with a materially different situation where not only was a finding made of past persecution, but the decisionmaker also made a finding (whether based on the presumption or not) that there was a wellfounded fear of persecution or a real risk of serious harm. In those circumstances the question of there being a good reason to rebut the presumption simply did not arise.

90. The various international guidance documents to which the appellants referred on this and other issues did not contain any suggestion that establishing past persecution or harm, or direct threats of persecution or harm should be treated as giving rise to a presumption of a future absence of state protection. In my view, it is unlikely that such a recommendation would be made. In cases involving persecution or harm at the hands of non-state actors, Article 6(c) of the Qualification Directive requires an applicant to demonstrate that the state is unable or unwilling to provide protection. Accepting, as I do, that the requirement to “demonstrate” the absence of state protection does not increase the burden of proof on an applicant, it is still an issue on which the applicant bears the primary onus of proof. To presume a future absence of state protection because an applicant has been subjected to past harm would completely remove any onus of proving that absence from the applicant. Instead, the decision maker would be required to address whether there were good reasons to consider the absence of state protection would not be repeated, despite that absence never being established in the first place.

91. It seems that that argument made to this court may have been more focused than that made, at least in N.G., in the court below. The appellants did not contend that the presumption had not been applied or that it had been incorrectly treated as having been rebutted. Instead, the argument concerned the scope of its application and that it should also have been applied to the question of state protection. Presumably, although this is not entirely clear, the appellants’ contention is that once past persecution/serious harm was accepted, then the Tribunal should have assumed (in the absence of good reason) that state protection would not be available in respect of that persecution/serious harm.

92. In my view, this is simply illogical and is particularly so in the context of the facts of these cases. There is an evidential logic to treating proof of something that happened in the past as an indicator that it is continuing or might happen again in the future. Because they have fled their countries, people who are making claims for international protection in a foreign country will face an evidential deficit in seeking to establish ongoing or future risks to them in a country in which they are no longer present. For example, if an applicant complains of ill treatment due to his ethnicity at the hands of the police in his country of origin, that applicant is manifestly not the subject of such ill-treatment whilst physically present in this country pursuing his claim for asylum. In a scheme designed to examine such claims it is both reasonable and consistent with the objective of ensuring that protection is granted to those who need it, to require that proof of past persecution of this nature be treated as indicative of likely future persecution.

93. It is in this context that the rationale for the appellants’ pursuit of the argument that the findings of well-founded fear of persecution/ real risk of serious harm must necessarily entail a conclusion that state protection is unavailable becomes clear. If the finding that a real risk of serious harm necessarily encompassed a finding that state protection was unavailable, then under Article 4(4), according to the appellants, that would have to be treated as a serious indicator of a future absence of state protection. When the Tribunal decisions are read as a whole, it is perfectly clear that the finding of a real risk of serious harm did not include a finding of a past failure of state protection. In fact, the opposite is the case particularly in Z.M. where the police and prosecutor’s handling of previous interactions between the appellant and G.C. was regarded as a positive indicator of the availability of state protection in Georgia.

94. I have already held against the appellants on the first point but counsel on their behalf argued that it was not necessary for him to succeed on the first point in order to succeed in his arguments under Article 4(4). Even if the established serious harm did not entail any

conclusion on the absence of state protection, he argued that such absence still had to be presumed because past harm had been established. I do not accept this argument.

95. If the Tribunal had found a past absence of state protection, it might logically lead to the conclusion that state protection would also be absent in the future unless there were good reasons to think otherwise. Note that I am not purporting to make any decision that the text of Article 4(4) would necessarily require a presumption to be applied as regards the lack of state protection in those circumstances. That question simply does not arise on the facts of these cases. There is no evidential nexus between what has been established in these cases, i.e. past threats made by criminals, and what the appellants say must be presumed, i.e. an inability or unwillingness on the part of state authorities to take appropriate action in respect of those threats.

96. The real difficulty the appellants face is that in neither case did the direct threat of harm to which they previously were subject entail any element of an absence of state protection. In N.G.'s case he did not report the threats he had received to the Albanian police before fleeing the country. Realistically how could – or indeed why should – the receipt of threats from criminals which were never brought to the attention of the Albanian authorities be treated as a serious indication that state protection would not be available to N.G. in Albania?

97. In Z.M.'s case the Georgian police and prosecution authorities took active steps against G.C. both in consequence of the burning of Z.M.'s parents’ house and for the murder of Z.M.’s cousin. G.C. received and served a lengthy prison sentence. He was apparently investigated in respect of the death of Z.L. Again, how could the receipt of threats from G.C. be regarded as a serious indication that the Georgian authorities, who had previously acted entirely appropriately in connection with these matters, would not now respond appropriately? There is simply no logical connection between the established threats and the behaviour of the police and prosecution authorities which could require the presumption that the former will be repeated to be accompanied by a presumption of future inaction on the part of the latter.

98. I am satisfied that there was no failure on the part of the Tribunal to apply the presumption for the benefit of these appellants. Although neither section 28(6) nor Article 4(4) were expressly referred to, in each case a finding that there was a well-founded fear of persecution/ a real risk of serious harm was made on the basis of the Tribunal's acceptance that past persecution/serious harm had occurred. As the presumption was applied to the benefit of the appellants, the Tribunal was not required to provide reasons for its rebuttal. The findings made of past harm did not entail any finding of a past absence of state protection. Indeed, in N.G. the Tribunal could not have made such a finding as the state authorities were unaware of the threats. In Z.M. the state authorities had responded appropriately to the criminal behaviour of the person alleged to be making the threats in the past. There was simply no basis in the findings made by the Tribunal for extending the evidential advantage under Article 4(4) to require a presumption be made as to the future conduct of the relevant state authorities which is not grounded in any evidence as to their past conduct.

Designation of Safe Country of Origin

99. The appellants’ third point arises from the fact that both Georgia and Albania have been designated by the Minister as safe countries of origin under Section 72 of the 2015 Act. In both cases the Tribunal had regard to this designation in concluding that state protection was available to the appellants in the countries in question. The appellants argued that the designation of a country as a safe country of origin has a specific legislative purpose and unless designation is used specifically for this purpose, the decision maker cannot have regard to it at all. Consequently, it is contended that the decisions are invalidated by reason of the fact that the Tribunal had regard to these designations.

100. Both Hyland J. and Phelan J. rejected the appellants’ argument on this point. Hyland

J. held that the appellant could not use the terms of section 33 of the 2015 Act to attack the substantive findings of the Tribunal because the purpose of section 33 was to demonstrate that a deciding body could not simply rely on a designation as a safe country of origin without looking at the individual circumstances of the applicant. On the facts she was satisfied that this was “emphatically not what happened” in Z.M. The argument made to Phelan J. appears to have been slightly different to that made to this court. Nonetheless, she noted that the Tribunal's reference to Albania's status as a safe country was in the context of its conclusions on state protection which in turn were based on a further analysis of country of origin information. Further, its designation was “recorded as a fact” but “not relied upon to obviate an analysis of the possibility of State protection by applying the information assessed in the COI to the Applicant's account of his difficulties in his home country”.

101. In order to understand this point it is necessary to look at the provisions of the Procedures Directive which allow a member state to designate a country as safe and at how these have been implemented in this jurisdiction in the 2015 Act.

102. The appellants point firstly to two recitals, numbers 17 and 21, in the Procedures Directive. It is acknowledged that these recitals do not have legislative effect, but the appellants rely on them as establishing a link between the designation of a country as safe and the availability of state protection in that country. The recitals are as follows: - “17. A key consideration for the well-foundedness of an asylum application is the safety of the applicant in his/ her country of origin. Where a third country can be regarded as a safe country of origin, Member States should be able to designate it as safe and presume its safety for a particular applicant, unless he/she presents counter-indications.

21. The designation of a third country as a safe country of origin for the purposes of this Directive cannot establish an absolute guarantee of safety for nationals of that country. By its very nature, the assessment underlying the designation can only take into account the general civil, legal and political circumstances in that country and whether actors of persecution, torture or inhuman or degrading treatment or punishment are subject to sanction in practice when found liable in the country concerned. For this reason, it is important that, where an applicant shows that there are serious reasons to consider the country not to be safe in his/her particular circumstances, the designation of the country as safe can no longer be considered relevant for him/her.”

103. The provisions dealing with safe countries of origin are contained in Section III of Chapter III of the Directive. It is notable that, apart from the three Articles which I describe below, the only other Article in that section, Article 28 deals with “unfounded applications”. That in turn refers back to Article 23 dealing with the “examination procedure”. Article 23(4)(c)(i) allows the examination procedure to be prioritised or accelerated by a member state if, inter alia, an applicant comes from a safe country of origin within the meaning of Articles 29 to 31.

104. Article 29 of the Procedures Directive allows the Council to adopt a minimum common list of third countries regarded as safe countries of origin and sets out the procedure through which such list is to be adopted. Article 30 allows for national designation of third countries as safe countries of origin. A member state may enact legislation which allows for the designation of countries other than those appearing on the Council's common list as safe countries. The minimum requirements to be met before a country can be so designated are set out in Annex II. A country may only be considered safe where “on the basis of the legal situation, the application of the law within a democratic system and the general political circumstances, it can be shown that there is generally and consistently no persecution … no torture or inhuman or degrading treatment …”. The assessment must take account of the “extent to which protection is provided against persecution or mistreatment” by, inter alia, the laws of the country and the manner in which they are applied and the provision of effective remedies against violations of rights and freedoms. This is made an express legal requirement under Article 30(4).

105. Article 31 deals generally with the “safe country of origin concept” and provides as follows: - “1. A third country designated as a safe country of origin in accordance with either Article 29 or 30 may, after an individual examination of the application, be considered as a safe country of origin for a particular applicant for asylum only if:

(a) he/ she has the nationality of that country; …

and he/ she has not submitted any serious grounds for considering the country not to be a safe country of origin in his/her particular circumstances and in terms of his/her qualification as a refugee in accordance with Directive 2004/83/EU.

2. Member States shall, in accordance with paragraph 1, consider the application for asylum as unfounded where the third country is designated as safe pursuant to Article 29.

3. Member States shall lay down in national legislation further rules and modalities for the application of the safe country of origin concept.”

106. Looked at overall, the scheme in the Procedures Directive regarding safe countries is as follows. Countries may be designated as safe at an EU level by the Council or at a national level by member states. In order for a country to be designated as safe by a member state, an assessment must be carried out which looks, inter alia, at the legal and political situation in that country and the extent to which protection against persecution or mistreatment is provided in that country. Even where a country is designated safe, an application for international protection from that country must be individually examined and the country may only be treated as safe as regards the individual applicant where the applicant does not establish serious grounds for not considering the country to be safe in his particular circumstances. Phrasing this in positive rather than negative terms, the designation of a country as safe is always subject to the requirement that each application for protection be individually examined and to the entitlement of an applicant for protection to show that the country is not safe in their particular circumstances. Further, where the Council has designated a country as safe then applications for protection from that country must be treated as unfounded – and under Article 23(4) treating an application as unfounded allows but does not require a member state to prioritise or accelerate that application. Of course, a prioritised or accelerated procedure must still afford the applicant the opportunity of showing that in their personal circumstances the country is not in fact safe. Finally, member states are given a fairly wide discretion to lay down “further rules and modalities” for the application of the concept in national law.

107. One of the difficulties with the safe country concept as set out in the Procedures Directive is that it does not appear at all in the Qualification Directive adopted a year earlier. Consequently, it is not entirely clear how the designation of a country as safe impacts on or interacts with the matters covered by the Qualification Directive, particularly those in Chapter 2 of the Qualification Directive dealing with the assessment of applications for international protection. This chapter includes three of the Articles which have already been considered in this judgment, namely Articles 4, 6 and 7.

108. These provisions of the Procedures Directive are implemented in Irish law by sections

33 and 72, 73 and 73A of the 2015 Act. Section 72 sets out the procedure through which a country can be designated as safe. There was no dispute that this procedure was followed by the Minister and that Georgia and Albania were duly designated as safe. The respondents rely on the fact that in order to do this the Minister conducted an assessment and was satisfied on the basis of the legal and political situation in those countries that there was generally and consistently no persecution, torture or inhuman and degrading treatment and that protection was available against persecution and mistreatment.

109. Section 73 allows for the prioritisation of applications for international protection and under section 73(2)(k) the fact the country is a safe country of origin is a matter to which the Minister can have regard in deciding to accord priority to that application. Similarly, section 73A allows for the acceleration of applications. Neither of these applications were prioritised or accelerated.

110. Section 33 of the 2015 Act is headed “Applicant from safe country of origin” and provides as follows: - “A country that has been designated under section 72 as a safe country of origin shall, for the purposes of the assessment of an application for international protection, be considered to be a safe country of origin in relation to a particular applicant only where—

(a) the country is the country of origin of the applicant, and

(b) the applicant has not submitted any serious grounds for considering the country not to be a safe country of origin in his or her particular circumstances and in terms of his or her eligibility for international protection.”

111. This section is clearly intended to implement Article 31(1) of the Procedures Directive. However, much like the provisions of the Procedures Directive dealing with safe countries relative to the those of the Qualification Directive, it is not entirely clear how section 33 fits into the scheme of 2015 Act as a whole. The applicant contends that section 33 provides for a different and truncated form of assessment where the application for protection will presumptively be refused unless the applicant shows “serious grounds” for considering their country not to be safe. This would be consistent with treating an application from a designated safe country as unfounded and subjecting it to a prioritised or accelerated procedure. Such procedures still require an individualised assessment of the application and are subject always to the entitlement of an applicant to show that the safe country designation is not relevant to their particular circumstances.

112. As noted, neither of these applications were treated as unfounded or subjected to an acceleration or prioritised examination procedure. The appellants had a full opportunity to make their cases and were not required to show “serious grounds” as to why the designation of their countries as safe countries should not apply before their claims were assessed in the normal manner. Consequently, the fact that their countries of origin were designated as safe did not operate as a procedural disadvantage to the appellants in any respect. However, the appellants nonetheless contend that unless section 33 was expressly applied to their cases, the Tribunal was not entitled to have regard to the designation in conducting a “protection analysis” under section 31 (i.e. the equivalent of Article 7 of the Qualification Directive). In fact, the appellants go further and say that the Tribunal was not required to conduct a section

31 protection analysis – and should not have done so - unless the presumption that the country is safe had been rebutted under section 33. Therefore, on the appellants’ construction of the legislation, the decision maker must conduct a separate, preliminary analysis under section 33 in all cases where the applicant for protection is from a designated safe country. Thereafter, the decision maker should only proceed to conduct a full protection analysis under section 31 where the presumption of safety has been rebutted by the applicant. In the context of this construction, the safe country designation should not then be applied as necessarily it has been rebutted before the protection analysis is undertaken.

113. This is obviously not the manner in which the Tribunal proceeded to deal with these claims. Instead, the Tribunal conducted a full analysis of the claim of persecution/serious harm and then of the availability of state protection in the appellants’ countries of origin. In neither case was regard had to the designation until after a finding was made regarding a well-founded fear of persecution/real risk of serious harm. The appellants accept that the full analysis of the protection issue under section 31 is more beneficial to an applicant and therefore preferable to the truncated section 33 analysis. In neither case was the designation of the countries as safe treated as definitively determinative of the availability of state protection as regards to the persecution/harm established. Instead, regard was had to the designation in circumstances where the existence of protection was otherwise evident from the country of origin information and, in the Z.M. case, the appellant's account of the past actions of the Georgian police and prosecution authorities.

114. Again, the appellants’ construction appears to depend, at least in part, on their first argument being accepted – i.e. that a finding of a well-founded fear of persecution/real risk of serious harm necessarily entails a finding that state protection is not available. If that were the case there might be some logic in saying that as state protection has been positively found not to be available, the designation of a country as safe cannot be relied on, even in the tangential way it was relied on in these cases, in order to hold that it is available. However, that is not what occurred in these cases as no findings were ever made by the Tribunal that state protection would not be available to the appellants in either Georgia or Albania.

115. The appellants make a further argument based on the language used in Article 31(1) of the Procedures Directive, which has been set out above. Under that Article a designated safe country “may” be considered as safe for a particular applicant for asylum “only if” the applicant has not submitted serious grounds for considering the country not to be a safe country of origin. Section 33 of the 2015 Act which transposes Article 31(1) is framed in a similar manner save that the designated country “shall” be considered safe unless the applicant submits the requisite serious grounds, and the phrase “only where” is used instead of “only if”. The appellants’ argument seems to be that as a matter of domestic law this confers some sort of procedural exclusivity on a section 33 analysis. I do not agree that this is so.

116. The purpose of section 33 is clearly to ensure that notwithstanding the designation of a country as a safe country of origin, an applicant for international protection may nonetheless present the case that the country is not safe for them in their particular circumstances. Section 33 does not in its terms require a separate and discrete analysis to be conducted and a decision made as to the existence of “serious grounds” as a precursor to the carrying out of the assessment of facts and circumstances required under section 28. The appellants’ argument is predicated on there being a distinct section 31 analysis of the availability of state protection, which will not be carried out at all if in a preceding section

33 analysis, an applicant has not submitted serious grounds for showing that a designated safe country is not safe in their particular circumstances.

117. However, the definitions of actors of persecution or serious harm and actors of protection contained in sections 30 and 31 are not intended to create discrete steps in the

decision-making process but instead inform the assessment of the facts and circumstances pertaining to a claim to be carried out under section 28. Notably, under section 28(4)(a) this assessment of the facts and circumstances requires the decision maker to take account of “all relevant facts as they relate to the country of origin at the time of taking a decision on the application, including laws and regulations of the country of origin and the manner in which they are applied”. Whilst in some instances the laws and regulations of a country may establish the persecution of particular groups, where persecution is alleged at the hands of non-state actors the laws and regulations will likely be relevant to the extent to which state protection is available both generally and specifically having regard to the need for “an effective legal system for the detection, prosecution and punishment of acts constituting persecution or serious harm” under section 31(4).

118. Therefore, there is no separate section 31 protection analysis which must be preceded by a section 33 analysis, and which will only be required if in the course of the section 33 analysis an applicant has submitted serious grounds for suggesting a designated country is not safe. There is an assessment of facts and circumstances pertaining to the application under section 28 in which the availability of state protection, as defined in section 31, will be relevant. The fact that a country has been designated safe will also be relevant to the overall analysis to be conducted under section 28. However, and this is an important proviso, the designation of a country as safe is always subject to the entitlement of an applicant to submit grounds showing that it is not safe for them in their particular circumstances.

119. Although an applicant must be allowed to make that case on an individual basis, because the country has been designated as safe through a procedure which requires an assessment of its general legal and political situation and the availability of protection from persecution, there is a higher onus on an applicant, namely the submission of “serious grounds”, before the designation of the country as safe will be regarded as not applying to them. Apart from (a) requiring an individualised examination of all such applications and

(b) requiring such applications to be considered unfounded (and potentially subjected to a prioritised or accelerated procedure) the Procedures Directive does not stipulate how or when consideration is to be given to the question of whether serious grounds have been submitted that the country is not in fact safe. Therefore, I cannot see that it precludes consideration being given to this issue in the context of a full examination of an applicant's claim for protection particularly where the member state has not purported to apply a prioritised or accelerated procedure to that application.

120. Consequently, I have difficulty in reading Article 7 of the Qualification Directive as requiring consideration of the designation of the country as safe to be completely excluded from the protection analysis to be carried out for the purpose of establishing if somebody is entitled to international protection under Article 2(c) or (e) or the cessation of that protection under Article 11. Apart from anything else, the enactment of Article 7 predates the enactment of the Procedures Directive and, unsurprisingly in that context, on its face Article 7 is completely neutral as to whether such designation is a factor to be considered. Insofar as Article 31(3) of the Procedures Directive affords member states a margin of appreciation regarding the application of the safe country of origin concept, the fact that section 33 of the 2015 Act uses “shall” rather than “may” seems to me to positively require decision makers to have regard to the designation of a country as safe rather than precluding them from doing so unless they have made an anterior decision that the applicant has failed to discharge the burden of submitting serious grounds that it might not be. It would in any event be very difficult for a decision maker to reach a conclusion on the safety of a country in the applicant’s particular circumstances without conducting an assessment of the facts and circumstances relating to that applicant’s application as required under section 28.

121. Further, in these cases the regard that was had to the designation was not used by the

decision maker to shortcut the assessment of the applications nor to exclude consideration of whether Georgia and Albania were in a position to provide protection under section 31/Article 7. Nor were the designations relied on to raise any presumption against the applicants nor to alter or increase the burden of proof which they otherwise bore. It was merely an additional factual matter which supported the conclusion which was available to the Tribunal on the basis of the country of origin information in each case.

Standard of Protection

122. Both appellants argued that the Tribunal had incorrectly concluded that state protection was available to them in circumstances where, according to the country of origin information, the level of protection available would not prevent the third parties acting on the threats they had made to the appellants. In particular, the country of origin information relied on by the Tribunal in each case focused on the existence of a functioning police force and the investigation and prosecution of those involved in criminal activities. Simply put, the appellants argue that in the event the threats were acted on, investigation and prosecution after the event would not protect them from the threatened harm. Z.M., in particular, contended that his life was at risk if he returned to Georgia as the police could not protect him from G.C., now released from prison, and who had already been involved in the murder of his cousin and the burning of his parents’ home.

123. Article 7(2) of the Qualification Directive, set out above, provides that state protection will generally be available where a state takes reasonable steps to prevent persecution and serious harm inter alia by operating an effective legal system for the detection, prosecution and punishment of acts constituting persecution or serious harm and that the applicant must have access to such protection. Section 31(2) of the 2015 Act includes an additional requirement, derived from the case of CJEU, that state protection must be effective and of a non-temporary nature. In neither case is an absolute guarantee required that persecution or serious harm will be prevented in order for acceptable state protection to exist, particularly where the harm threatened is at the hands of non-state actors.

124. The court was referred to a number of authorities addressing what constitutes effective state protection. Clarke J. in Idiakheua v. Minister for Justice [2005] IEHC 150 offered a standard of “reasonable protection in practical terms” which is the standard most often referred to by Irish courts. In Abdulla Advocate General Mazák stated (at para. 54 of his opinion) in relation to the requirements of Article 7(2) of the Qualification Directive: - “These non-negligible, concrete requirements entail the presence of an actor of protection which has the authority, organisational structure and means, inter alia, to maintain a minimum level of law and order in the refugee’s country of nationality. The actor of protection must thus objectively have a reasonable level of capacity and the willingness to prevent acts of persecution as defined by Article 9 of Directive 2004/83.”

125. However, in the same paragraph he notes that “protection need not be absolute in nature, a result which is in any event unobtainable in any society” (at footnote 28). This is consistent with the notion of a minimal level of law and order which the state has a reasonable capacity to deliver. In its judgment in the same case the CJEU observed that verification of the existence of state protection under Article 7(2) would require the competent authorities to assess: - “In particular, the conditions of operation of, on the one hand, the institutions, authorities and security forces and, on the other, all groups or bodies of the third country which may, by their action or inaction, be responsible for acts of persecution against the recipient of refugee status if he returns to that country. In accordance with Article 4(3) of the Directive, relating to the assessment of facts and circumstances, those authorities may take into account, inter alia, the laws and regulations of the country of origin and the manner in which they are applied, and the extent to which basic human rights are guaranteed in that country.”

126. Advocate General Hogan in his opinion in O.A. at footnote 35 to paragraph 59 states that in his view “not only must 'reasonable steps' to be taken, they must also be reasonably effective in their objective”.

127. The appellants did not really take issue with the existence in their countries of origin of an effective legal system for the detection, prosecution and punishment of acts constituting persecution or serious harm. Rather, they argued that this was not sufficient in the context of the threats made to them by known serious criminals. To be effective, they contended, state protection must obviate the risk and, consequently, a state is obliged to take preventative action.

128. This prompted a discussion as to what constituted “reasonable steps” and the extent to which a risk of serious harm must be eliminated or prevented from materialising in order for effective state protection to be available to the appellants. When faced with the difficult question as to what they maintained the Georgian or Albanian authorities would be required to do to provide effective state protection in the circumstances, the appellants were unable to identify any ready answer. Ultimately the only suggestion made was the operation of something equivalent to a witness protection programme for persons who were subject to such threats. They accepted that no such program operates in Ireland, and they were unable to identify any other country, certainly no democratic country, where that level of protection was provided other than exceptionally to persons acting as state witnesses in certain ongoing criminal investigations and trials. They also accepted that a state could not breach third-party rights, for example through a system of preventative detention, in order to prevent the possibility of the detained person committing acts of persecution or serious harm. Thus, in reality the appellants were not seeking a level of protection from this state that would not be available to them in their countries of origin. Rather they were seeking to distance themselves from threats in their countries of origin by physically removing themselves from the location and circumstances in which the threats arose.

129. It was notable that Z.M. claims that threats have been made to him since his arrival in this country through a Georgian national based here. He has not reported these threats to the Gardaí. He regarded this as unnecessary in circumstances where he believes he is only at a risk if he returns to Georgia. He seemed to accept that the Gardaí would not be able to provide him with any greater level of protection than the Georgian police should the individuals through whom the threats are being communicated in this jurisdiction seek to act upon them. Counsel for Z.M. could not confirm that he had made reports to the Georgian police of the threats made by G.C. since his release from prison.

130. The international guidance documents relied on by the appellants did not really support the case they made that the risk of serious harm or persecution must be eliminated or prevented before state protection can be regarded as being available. The Commission's travaux préparatoires to the Qualification Directive (COM/2001/0510 final – CNS 2001/0207) deals in some detail with sources of harm and of protection – i.e. what ultimately became Article 6 and 7 of the Qualification Directive.

131. The protection issue is described as being “whether such a system as a whole offers a sufficient and accessible level of protection to all members of the population. For the system to offer effective protection, the state must be able and willing to operate it, such that there is no risk of persecution of other serious harm being realised.” Thereafter it lists 11 factors to which member states should have regard in their assessment. The document envisages the analysis comprising two stages, namely whether the state has taken or could reasonably be expected to take adequate steps to control or combat the infliction of harm and whether the applicant has reasonable access to the protection afforded by that state.

132. During the hearing of this appeal, I went through each of the eleven factors listed with counsel for the appellants who was unable to identify where either Georgia or Albania would be regarded as having failed to provide the requisite level of protection. The factors included things like state policies with respect to the harm at stake and whether violent attacks are punishable under the criminal law with sentences commensurate with the gravity of the crimes; whether the alleged persecutors had influence with the state officials; the willingness of law-enforcement agencies to detect, prosecute and punish offenders; whether there was a pattern of state unresponsiveness or a denial of state services and whether steps are taken to prevent the infliction of harm. In considering whether an applicant has reasonable access to state protection, the Commission suggests factors to be considered include whether the alleged persecutors are subject to state control, whether the applicant would be refused access to protection and the steps taken by the applicant to secure protection and the state response to those steps.

133. In circumstances where G.C. has been investigated, prosecuted and convicted (twice) in relation to the matters of which Z.M. complains it is difficult to see how any of these factors could be regarded as indicative of an absence of state protection in Georgia. All of the individuals involved in the original feud, including Z.M. himself, were investigated, arrested, prosecuted, convicted and served lengthy prison sentences which is indicative of a pro-active police force and a functioning criminal justice system. Likewise, according to N.G., A.K. was arrested and detained in respect of the murder of A.S. in Albania – the dispute between these two individuals being the source of the allegations of theft against this appellant and the subsequent threats made to him. As N.G had never reported the threats made against him to the Albanian police, they cannot be criticised for any lack of response thereto.

134. Consequently, I cannot accept that the Tribunal erred in finding that there was adequate state protection available to the appellants in either Georgia or Albania. State protection must be effective but is not required to be absolute. The obligation on a state under Article 7 is to take “reasonable steps”. The requirement outlined in Article 7, albeit not necessarily exhaustive, is that a state operate an effective legal system and, in particular, an effective criminal justice system as regards acts constituting persecution or serious harm. The country of origin information demonstrates that both Georgia and Albania do this; the facts of both cases are consistent with them doing so and nothing in the evidence of the appellants (in Z.M.’s case that portion of the evidence which was accepted as plausible) suggested otherwise. Absolute protection, especially where the threat emanates from criminal elements, would in any event be impossible to attain.

Standard of Proof for Past Events

135. The final two issues are raised only in the Z.M. appeal, and both concern the manner in which the evidence in his claim was dealt with by the Tribunal. The first concerns the standard of proof applied to past events – i.e. the appellant’s account of the events which caused him to leave his home country. It will be recalled that part of the appellant's account was accepted, namely the fact that he had received death threats from G.C. and that G.C. had been involved in the burning of his parents’ home and the death of his cousin. Indeed, G.C. was prosecuted and convicted in respect of these two crimes and served a lengthy prison sentence. However, other parts of the appellant's account were not accepted. These were, most notably, the appellant's belief that Z.L. had died in mysterious circumstances and that G.C. was involved in his death and his account of his interactions with the Georgian police and the claim that they had told him he should leave the country.

136. The appellant claims that these elements of his account were incorrectly rejected on the basis that they were not established on the balance of probabilities. He does not contend that his account did in fact reach this threshold, rather he argues that a different and lower threshold should have been applied. In particular, the appellant urges the court to adopt the approach taken by the Court of Appeal in the United Kingdom in Karanakaran v. Secretary of State for the Home Department [2000] 3 All E.R. 449 in applying a standard less than the balance of probabilities to asylum claims in order to ensure compliance with international law.

137. The appellant acknowledges that this issue was expressly considered and determined by the High Court (O’Regan J.) in a test case (O.N. v. Refugee Appeals Tribunal [2017] IEHC 13) by which Hyland J. regarded herself as bound. The appellant argues that O.N. was improperly decided and that this court should not follow it. He points to a number of decisions in this jurisdiction in which either Karanakaran was expressly followed – such as M.A.M.A. v. Refugee Appeals Tribunal [2011] 2 IR 729 (Cooke J.) or in which an approach, which the appellant describes as a Karanakaran approach, was applied, albeit without any reference to the decision in Karanakaran - such as by the Supreme Court in Minister for Justice v. Rettinger [2010] 3 IR 783. The appellant argues that although Rettinger was a European Arrest Warrant (EAW) case, because Article 3 of the European Convention on Human Rights was engaged the principles applied, namely a rigorous examination of all of the material to ascertain if there was a real risk, should apply to the refusal of an asylum claim which may ultimately involve the return of the appellant to his home country.

138. The appellant also argues that the standard of proof applied by the Tribunal was in breach of European law, the European Union Charter, the European Convention on Human Rights and the Constitution but no argument was directed to any specific element of these instruments. The complaint was more a general one that the outcome was unlawful because the appellant faces a real risk of serious harm if returned to Georgia. In this context the appellant argues that the same standard of proof should have been applied to the determination of past events as to the assessment of future risk. In other cases, the respondents have accepted that a lower standard of probability – of approximately 30% (per O.N.) – applies to the assessment of future risk in international protection cases.

139. As previously stated, Hyland J. regarded herself as bound by the settled line of case law reflected in O.N. She also noted that an application for leave to appeal to the Supreme Court in order to argue for a lower standard of proof such as “a reasonable degree of likelihood” or “a reasonable chance” was refused in M.E.O. (Nigeria) v. IPAT [2019] IESCDET 165. The Supreme Court determination records that the intending appellant had not shown that the standard of proof applied was contrary to EU or international law or to the ECHR or domestic law. Further, the indications were that the standard of proof currently applied was “humane and just and, [are] attuned to the difficulties and compassion required in cases of this nature”.

140. The appellant objects to any reliance being placed on the determination in M.E.O. on the grounds that such determinations are not binding and do not have precedential effect. This is of course correct, and Hyland J. herself noted as much but nonetheless regarded it as having some persuasive value. Such determinations routinely include a statement to the effect that the ruling contained therein has “no precedential value as a matter of law”. I do not propose to delve into the distinction between precedential and persuasive value and the extent to which Article 34.5.4 determinations refusing leave to appeal might be regarded as having, at least, the latter. Where leave to appeal is granted by the Supreme Court, any observations made in the determination granting leave will be overtaken by the substantive judgment of that court once the case has been fully argued and decided. Arguably, where leave to appeal is refused the determination may potentially indicate an acceptance of the law as reflected in the existing jurisprudence. Be that as it may, I think it preferable to decide the issue in this appeal without reference to the determination in M.E.O. These appeals have been heard together because they raise similar issues, issues which also arose in Heslin J.’s judgment in T.A. v IPAT (above). In light of the fact that these issues have now arisen again in a number of cases, the unsuccessful party in this appeal may well invite the Supreme Court to reverse the view it took in M.E.O. and to grant leave to appeal from this decision. I should not presume the outcome of any such application.

141. The respondents’ position on this issue is straightforward. Neither the Qualification Directive nor the Procedures Directive mandate a standard of proof. This is correct. They do deal with the allocation of the burden of proof and raise presumptions which may shift the burden of proof, but they do not address the standard of proof to be met in order for the burden of proof to be discharged. Consequently, in the absence of an EU standard, the principle of national procedural autonomy applies. It is open to a member state to apply a standard of proof that is consistent with its own legal system. The extent of a member state’s autonomy in this regard is limited by the principles of equivalence and effectiveness but the appellant does not make the case that these principles were not complied with by the application of the standard of proof of the balance of probabilities. Consequently, the standard of proof applied to the determination of past events in this case – namely the balance of probabilities coupled with the benefit of the doubt in appropriate circumstances – was lawful. Counsel for the respondent also notes that the decision in Karanakaran predates the adoption of the Qualification Directive and the Procedures Directive and the development of a comprehensive acquis of European law on international protection. Therefore, the Karanakaran judgment needs to be specifically considered in that context.

142. Given the structure of the appellants’ arguments, the starting point for this analysis must be the decision in Karanakaran itself. The applicant in that case was a Sri Lankan Tamil and a failed asylum seeker. Expert evidence had been adduced on his behalf raising a number of different points and that evidence was rejected on the basis that it was because it was too speculative and insufficient to establish that the applicant's return to Sri Lanka would be unduly harsh. Further, it was held that it would be open to the applicant to relocate internally in Sri Lanka.

143. Before looking specifically at the principles set out in the judgment of Brooke L.J. on which the appellant relies, it should be noted that the evidence given by the applicant in Karanakaran was in fact accepted. His account was that he was from Northern Sri Lanka, his home and family had been repeatedly attacked by government forces, family members had been injured and killed by the army and that he was harassed by the “Tamil Tigers” who sought to recruit him for their cause. At a time prior to the introduction of subsidiary protection, his claim was rejected at first instance on the basis that his family was not specifically targeted but rather was caught up in a situation of conflict in the area in which they lived and that internal relocation in Colombo was open to him. Expert evidence before the Immigration Appeals Tribunal dealt with the question of whether it was possible for a Tamil to live safely in Colombo and suggested that the Tamil population there was subjected to arbitrary arrest and detention by the security forces and harassment by the Tamil Tigers. This evidence was characterised as speculation by the Tribunal, but it appears that the applicant's personal account of his experiences was again accepted.

144. I have set this out because I think it important to understand that on its facts Karanakaran was a case about the assessment of future risks and not the determination of past events. The judgment ranges more widely as some of the authorities considered – including Kaja v. Secretary of State for the Home Department [1995] Imm AR 1 – looked specifically at the latter rather than the former. In Kaja it had been held that in assessing future risk, a decision maker was not bound to exclude evidence which did not reach the threshold being probably true but to which the decision-maker was prepared to attach some credence. Brooke L.J. noted that this had been interpreted as the application of a lower standard of proof to past events although this interpretation does not appear to be strictly accurate.

145. The court also examined a line of Australian jurisprudence culminating in the decision of Sackville J. in Minister for Immigration and Multicultural Affairs v. Rajalingam [1999] FCA 719. The principles to be derived from this jurisprudence are summarised by Brooke L.J. (at page 468 of the report) as follows: - “(1) There may be circumstances in which a decision maker must take into account the possibility that alleged past events occurred even though it finds that those events probably did not occur. The reasons for this is that the ultimate question is whether the applicant has a real substantial basis for his fear of future persecution. The

decision maker must not foreclose reasonable speculation about the chances of the future hypothetical events occurring.

(2) Although the civil standard of proof is not irrelevant to the fact-finding process, the decision-maker cannot simply apply that standard to all fact finding. It frequently has to make its assessment on the basis of fragmented, incomplete and confused information. It has to access the plausibility of accounts given by people who may be understandably bewildered, frightened and, perhaps, desperate, and who often do not understand either the process or the language spoken by the decisionmaker/investigator. Even applicants with a genuine fear of persecution may not present as models of consistency or transparent veracity.

(3) In this context, when the decision-maker is uncertain as to whether an alleged event occurred, or finds that although the probabilities are against it, the event may have occurred, it may be necessary to take into account the possibility that the event took place in deciding the ultimate question (for which see (1) above). Similarly, if the non-occurrence of an event is important to the applicant's case, the possibility that the event did not occur may need to be considered by the decision-maker even though it considers that the disputed event probably did occur.

(4) Although the “what if I am wrong?” terminology has gained currency, it is more accurate to see this requirement as simply an aspect of the obligation to apply correctly the principles for determining whether an applicant has a “well-founded fear of being persecuted” for a Convention reason.

(5) There is no reason in principle to support a general rule that a decision-maker must express findings as to whether alleged past events actually occurred in a manner that makes explicit its degree of conviction or confidence that its findings were correct. …

(6) If a fair reading of the decision-makers reasons as a whole shows that it “had no real doubts” that claimed events did not occur, then there was no warrant for holding that it should have considered the possibility that its findings were wrong.”

146. Personally, I have to admit that I find these statements of principle somewhat convoluted and expect that a decision maker might have great difficulty in attempting to apply them, especially first instance decision makers who will usually not have legal qualifications. Brooke L.J.’s summary of what is required is a little more straightforward: - “This approach does not entail the decision maker … purporting to find “proved” facts, whether past or present, about which is it not satisfied on the balance of probabilities. What it does mean, on the other hand, is that is must not exclude any matters from its consideration when it is assessing the future unless it feels that it can safely discard them because it has no real doubt that they did not in fact occur (or, indeed, that they are not occurring at present).”

147. I find the observations of Sedley L.J. who concurred in the outcome of Karanakaran of more assistance. In analysing the exercise being undertaken by a decision maker he said (at p.477 of the report): - “The civil standard of proof, which treats anything which probably happened as having definitely happened, is part of a pragmatic legal fiction. It has no logical bearing on the assessment of the likelihood of future events or (by parity of reasoning) the quality of past ones. It is true that in general legal process partitions its material so as to segregate past events and apply the civil standard of proof to them: so that liability for negligence will depend on a probabilistic conclusion as to what happened. But this is by no means the whole process of reasoning. In a negligence case, for example, the question will arise whether what happened was reasonably foreseeable. There is no rational means of determining this on a balance of probabilities: the court will consider the evidence, including its findings as to past facts, and answer the question as posed. More importantly, and more relevantly, a civil judge will not make a discrete assessment of the probable veracity of each item of the evidence: he or she will reach a conclusion on the probable factuality of an alleged event by evaluating all the evidence about it for what it is worth. … So it is fallacious to think of probability (or certainty) as a uniform criterion of fact-finding in our courts: it is no more than the final touchstone, appropriate to the nature of the issue, for testing a body of evidence of often diverse cogency.”

148. Sedley L.J. summarised his view as follows (at p.479 of the report): - “The question whether an applicant for asylum is within the protection of [the Convention] is not a head-to-head litigation issue. Testing a claim ordinarily involves no choice between conflicting accounts but an evaluation of the intrinsic and extrinsic credibility, and ultimately the significance, of the applicant's case. It is conducted initially by a departmental officer and then, if challenged by one or more tribunals which, though empowered by statute and bound to observe the principles of justice, are not courts of law. … Such decision-makers, on classic principles of public law, are required to take everything material into account. Their sources of information will frequently go well beyond the testimony of the applicant and include in-country reports, expert testimony and – sometimes – specialised knowledge of their own (which must of course be disclosed). No probabilistic cut-off operates here: everything capable of having a bearing has to be given the weight, great or little, due to it. What the decision-makers ultimately make of the material is a matter for their own conscientious judgment, so long as the procedure by which they approach and entertain it is lawful and fair and provided their decision logically addresses the Convention issues. Finally, and importantly, the Convention issues from first to last are evaluative, not factual. The facts, so far as they can be established, are signposts on the road to a conclusion on the issues; they are not themselves conclusions. How far this process truly differs from civil or criminal litigation need not detain us now.” He ultimately describes that the process as “a unitary process of evaluation of evidential material of many kinds and qualities against the Convention’s criteria of eligibility for asylum.” For that reason, he did not have regard to any question of either the burden or standard of proof arising on the critical issue of internal relocation.

149. In Rettinger the Supreme Court considered the standard of proof that should apply in European Arrest Warrant cases when a person facing surrender complains of the danger of being subjected to inhuman or degrading treatment in the issuing member state (Fennelly J. at paragraph 64). Two judgments were delivered. Denham J. set out a series of principles to be applied (at p.801 of the report). These require, inter alia, the court to conduct a rigorous examination as to whether there is a real risk and to examine the foreseeable consequences of sending the person to the requesting state. However, mere possibility of ill-treatment is not sufficient to establish a respondent's case.

150. The appellant argues that the second of these principles – “a court should examine whether there is a real risk, in a rigorous examination”- reflects the Karanakaran standard. I am reluctant to treat Rettinger as having somehow silently imported the Karanakaran line of jurisprudence into Irish law, especially as Karanakaran predates Rettinger by about a decade. If this was Denham J.’s intention, it might be expected that she would have expressly stated that she was doing so. Further, the third of Denham J.’s principles notes that the burden lies on the respondent (i.e. the person to be surrendered) and the discharge of that burden requires the respondent “to adduce evidence capable of proving that there are substantial grounds for believing that if he (or she) were returned to the requesting country he, or she, would be exposed to a real risk of being subjected to treatment contrary to Article 3…”. Given that the materialisation of a future risk can never be “proved” as such, the obligation to adduce evidence to establish substantial grounds for believing that the risk is real seems to me to reflect a somewhat higher standard than that which the appellant contends for here.

151. Fennelly J. noted that the relevant section, section 37(1)(c) of the European Arrest Warrant Act 2003 required no more than that there be reasonable grounds for belief that were the person to be surrendered he or she would be tortured or subjected to inhuman or degrading treatment. His conclusions as to the applicable standard were as follows (p. 817 of the report): - “It is neither possible nor appropriate to prescribe too narrowly the way in which a judge should assess the evidence, whose quality necessarily varies enormously. The credibility of any evidence will depend on all the circumstances of the individual case. In the final analysis, it is a matter for the judge to decide whether the necessary substantial grounds have been established for believing that there is a real risk of illtreatment contrary to Article 3. He does not have to be convinced that ill-treatment would probably occur. There is no shifting of the burden of proof in the absence of evidence of substantial grounds. The judge may, however, regard the failure of the applicant to respond to evidence produced as significant. In that situation, he may reasonably expect the applicant to produce some evidence.”

152. The respondent points out, and I accept, that the Supreme Court was not considering the standard of proof applicable to past events in Rettinger. The issue was whether the surrender of the respondent to Poland would give rise to a risk of inhuman or degrading treatment because of the conditions in the prisons in that country. Although the respondent had attested to the prison conditions he had previously experienced, the issue was really a systemic one to which international reports and decisions of the Polish Supreme Court and the European Court of Human Rights were relevant. Further, although it was dealing with circumstances where Article 3 of the ECHR was engaged due to the risks the respondent claimed he faced if returned to his country of origin, the Court did not have to determine the standard of proof applicable to such risks in the abstract because the statutory provision set a standard of “reasonable grounds for belief”.

153. The appellant correctly observes that Cooke J. followed the decision in Karanakaran in M.A.M.A. (above). However, I think it is instructive to look at how Cooke J. anticipated the analysis of future risk might look when an applicant’s evidence regarding past events has not been accepted. He stated at paragraph 17 of his judgment: - “This Court accepts as correct the approach to the standard of proof outlined in this case law. The sole fact that particular facts or events relied upon as evidence of past persecution have been disbelieved will not necessarily relieve the administrative

decision-maker of the obligation to consider whether, nevertheless, there is a risk of future persecution of the type alleged in the event of repatriation. In practical terms, however, the precise impact of the finding of lack of credibility in that regard upon the evaluation of the risk of future persecution must necessarily depend upon the nature and extent of the findings which reject the credibility of the first stage. This is because the obligation to consider the risk of future persecution must have a basis in some elements of the applicant's story which can be accepted as possibly being true. The obligation to consider the need for "reasonable speculation" is not an invitation or pretext for gratuitous speculation: it must have some basis in, and connection to, the apparent circumstances of the applicant.”

154. In the following passages of his judgment Cooke J. explains how the rejection of an applicant’s evidence as to past events may impact on the assessment of future risk. For example, if the decision maker rejects the applicant's evidence regarding his nationality or ethnicity or that they come from the region or place in which persecution is said to exist there may be no obligation to speculate as to the risks attending repatriation in the case. In contrast, if an applicant's evidence as to his personal involvement in certain events is not accepted but it is accepted that he comes from the place in which those events occurred and has a relevant ethnicity or religion etc, the decision maker will still have to consider a risk of future harm based on a person having that ethnicity or religion in that location. Therefore, notwithstanding his acceptance of the Karanakaran line of jurisprudence, Cooke J.’s application of it does not reflect what the appellant contends is required. Instead of assessing the risk to the applicant based on the possible truth of elements of his account which were rejected, Cooke J. regarded two things as required. Firstly, even if an applicant’s account of his personal involvement in events is not credible, the decision maker still has to assess the risk to that applicant of future persecution or serious harm in their country of origin. Secondly, that assessment should be based on those elements of the applicant’s account which have been believed which, in M.A.M.A., included his membership of a minority tribe in the Darfur area of Sudan.

155. In O.N. the applicant contended the appropriate standard of proof applicable to the determination of facts and/or the acceptance of the history of events as disclosed by an individual applicant was the same as that applied to the assessment of future risk, namely a “reasonable degree of likelihood” or a standard of approximately 30%. The respondents contended that the appropriate standard was the balance of probabilities coupled with affording the applicant the benefit of the doubt in appropriate circumstances. The respondents accepted that a lower standard should be applied to the assessment of future risk. They also acknowledged that, depending on the circumstances, some facts which may have been discounted in a fact-finding exercise may nevertheless be taken into account in the evaluation of future risk.

156. O’Regan J. looked at Articles 2, 3 and 4 of the Qualification Directive, the UNHCR Handbook and guidance and then conducted a review of the relevant case law, including the decisions in Karanakaran and MAMA. The domestic legislation in force at the time the decisions were made was the Refugee Act 1996, which has since been replaced by the 2015 Act. She noted Article 4(5) of the Qualification Directive which allows for the acceptance of an applicant’s statements which are not supported by documentary or other evidence where certain conditions are fulfilled. These conditions include at (c) that “the applicant’s statements are found to be coherent and plausible and do not run counter to the available specific and general information relevant to the applicant’s case” and at (e) that “the general credibility of the applicant has been established”. Article 4(5) is transposed by section 28(7) of the 2015 Act and the reference to the benefit of the doubt relates to the application of this section by decision-makers. In certain cases, it may provide a significant benefit to an applicant who is unable to adduce specific evidence to support their account of events in their country of origin but where the presentation of their application otherwise fulfils the relevant criteria.

157. This application of the benefit of the doubt is evident in the N.G. case. The IPO rejected N.G.’s account of the threats received by him on the basis that his general credibility was not accepted. His account was regarded as vague and lacking specificity, unreasonable and undermined by his delay in leaving Albania and in making his application for protection. The Tribunal on the other hand regarded N.G. as having given a “broadly consistent and coherent account of his claim throughout the protection process” and that the claim “does not run counter to any relevant COI”. Consequently, the Tribunal member extended the benefit of the doubt to N.G., accepted the material facts of his claim and proceeded to assess it on the basis that he had received threats from criminal elements as he had described. If anything, the application of the benefit of the doubt to N.G. was generous as he did not fulfil all five conditions in Article 4(5)/section 28(7) in that he had not applied for protection at the earliest possible opportunity.

158. At the conclusion of her judgment, O’Regan J. set out a series of principles which emerged from all of the material which she considered. She noted there was no single approach to the standard of proof in asylum cases which was universally accepted and no approach mandated internationally by the Geneva Convention or under EU law. She acknowledged that the UK, which had previously applied a balance of probability standard, is now applying a “reasonable degree of likelihood” standard. She noted that different common law countries applied variously a balance of probabilities test (e.g. Canada) or a lower standard (e.g. the U.S.A.). She distinguished M.A.M.A. on the basis that it was dealing with the standard required when looking at the reality of the risk of future persecution or serious harm and not the standard required in respect of past facts or the credibility of an applicant. Consequently, these judgments did not, in O’Regan J.’s view, import a “reasonable degree of likelihood” test in respect of past facts or the applicant's credibility into Irish law.

159. Consequently, she concluded that the principles of equivalence and effectiveness were safeguarded by the application of a standard of proof being the balance of probabilities coupled with, where appropriate, the benefit of the doubt. This judgment has been followed in a number of subsequent cases and is generally regarded as the authoritative decision of the High Court on the issue.

160. The appellant argues that O.N. was wrongly decided and should not be followed by this court. The main argument made was simply that the standard in Karanakaran was the correct standard and should be applied in this jurisdiction. In many respects the argument went significantly further than Karanakaran - counsel suggested that even if the decisionmaker was satisfied to the extent of 95% that an applicant’s account was untrue, that meant there was still a 5% risk which could constitute a real risk of serious harm.

161. In my view, the use of percentages in this manner is not especially helpful. It is important to bear in mind the distinction between proof of past events which, as a matter of fact, either have or have not occurred and “proof” of future risk. In principle the occurrence of past events can be established through evidence consisting of oral and written statements, documentary material supporting those statements and, in the context of international protection, international reports reflecting the experiences, observations and research of independent third parties where the applicant’s account is consistent with those reports. It may be difficult for an individual applicant who is outside of their country of origin and who may have left suddenly and in traumatic circumstances to adduce all the evidence that would normally be required to establish the veracity of their account. The application of the benefit of the doubt facilitates such an applicant where their account is otherwise coherent, plausible and not inconsistent with the information generally available regarding their country of origin. Where the benefit of the doubt is applied, the effective standard of proof in traditional, common law terms will be less than the balance of probabilities - otherwise there would be no benefit to an applicant since they would in any event have proved the requisite elements of their claim.

162. Strictly speaking in evidential terms the likelihood of a risk materialising in the future cannot be “proved” as the factual events which would amount to such materialisation have not yet occurred. Instead, what is required of a decision-maker is an assessment of the likelihood of the risk materialising in the future. The past history and the events described by an applicant will be central to this assessment as will independent information concerning the applicant’s country of origin. As in the case of any risk assessment, a number of factors must be weighed in the balance including, in the context of this case, the availability of state protection in Georgia. It is not really accurate in my view to characterise this forwardlooking assessment as involving the application of a standard of proof, although I acknowledge that for convenience this language is frequently used. What is important is that the law acknowledges that a rigorous assessment of the risk is required.

163. Reliance was also placed by the appellant on a 2023 publication by the European Union Agency for Asylum (EUAA) entitled “Judicial Analysis on Qualification for International Protection (Directive 2011/95/EU)”. (Note that Ireland has not opted into the recast Qualification Directive 2011/95/EU.) This publication looks at what it terms the “Standard of proof/level of conviction” applied in different member states. It commences by acknowledging that the principle of procedural autonomy leaves this matter to individual member states. It also notes that diverse terminology is used regarding the standard of proof/level of conviction in different member states.

164. The authors appear critical at the application in Ireland of two different standards of proof in the assessment of an application for international protection which they describe as “one in connection with the establishment of the (past and present) material facts and circumstances of an application and another in assessing whether the established material facts and circumstances give rise to a future ‘well-founded fear of persecution’ or a future ‘real risk of serious harm’”. It is not clear from the text why this could be regarded as unsatisfactory in the context of the Irish common law system save for the fact that civil law jurisdictions, which in any event treat evidence and proof materially differently, do not necessarily make this distinction. Somewhat ironically, if Ireland were to apply a probability of not more than 50% to both elements of the analysis (as applied in Germany), the overall effect would be to increase the standard of proof (insofar as assessment of future risk can be described as being subject to a standard of proof) which applicants currently face.

165. There are inherent difficulties, evident in the publication, when comparing civil law and common law jurisdictions. For example, the authors note that “in civil law jurisdictions the judge must be convinced that the applicant in fact fulfils the criteria for qualification as a refugee or eligibility for subsidiary protection.” To common lawyers “convinced” suggests the application of a higher standard than requiring the decision maker to be “satisfied” whether on the balance of probabilities or otherwise. They then go on to say the “material standard of their conviction” is defined by the recast Qualification Directive and that, in the case of a refugee, that standard is that the fear of persecution is “well-founded”. The notion of a well-founded fear persecution is of course an intrinsic element of the definition of a refugee. To say that the fear must be well-founded tells the common lawyer little about the standard that must be applied in order to conclude that the fear is actually well-founded. In fact, it appears that the standard applied to the future element of the assessment in the countries referred to (namely Germany and Slovenia) is higher than that applied in Ireland (a probability of less than 50% compared to approximately 30%).

166. It is difficult to understand the difference the contended for standard would have made to the determination of this appellant’s claim. Much of his factual account was accepted and a forward-looking risk assessment was conducted on the basis that he faced a real risk of serious harm based on that account. When pressed, the suggestion made by counsel on his behalf was that the appellant’s speculative view that G.C. was responsible for the death of Z.L. in a car crash would or should have been accepted if a lower standard of proof had been applied. The logic of this is not apparent. There was no evidence at all adduced that G.C., who was in prison at the time, was involved in the car crash which led to the death of Z.L. The Tribunal decision notes that the appellant “has never gone further than claiming to have a suspicion that G.C. was behind the death of Z.L.” and the appellant “accepted he had no hard evidence to support this suspicion”. Even if a lower standard of proof were to have been applied, as there was no evidence at all to support the appellant’s suspicion it would still not have been capable of being regarded as proved.

167. In the circumstances I am not satisfied that the appellant has shown that the standard of proof applied to his application was legally incorrect. I am also not satisfied that the application of a lower standard of proof to the determination of the past or historic elements of this claim would have made any difference in the particular circumstances.

Plausibility/Benefit of the Doubt

168. The final issue raised by Z.M. concerns the lawfulness of the finding made by the Tribunal that it was implausible that the Georgian police had told him they could not protect him and that he should leave the country within fifteen days on pain of being framed by them for a crime he did not commit. The time devoted by the appellant to this aspect of his case in oral argument was limited and reliance was placed instead on his written legal submissions. Those submissions make a very broad argument considering, in a general way and by reference to academic material, the difference between plausibility and credibility and the extent to which something can be said to be “counter” to available country of origin information if it is simply not mentioned in that information.

169. It is clear from the Tribunal decision that the appellant was cross-examined by the presenting officer at the hearing specifically on the plausibility of this element of his evidence. It was put to the appellant that not only was his account inherently implausible, it was inconsistent with other aspects of his claim which detailed the “Georgian police force and prosecution service meticulously investigating and prosecuting various crimes in the manner that one might expect if a criminal justice system is functioning in a reasonable manner”. The appellant’s only response was to repeat his account of his interactions with the Georgian police. The Tribunal emphasised the reluctance with which it made implausibility findings and acknowledged the risk of speculation and conjecture. Nonetheless it concluded that the appellant’s contention was inconsistent with his own account of the actions of the Georgian police in connection with these matters which led to G.C. serving a lengthy prison sentence and indeed being further questioned whilst in prison. The applicant’s account was not supported by any country of origin information which suggested that the Georgian police advise citizens to leave the country for their own safety or threaten to imprison citizens for crimes they have not committed. Consequently, the Tribunal found this element of the appellant’s evidence to be implausible and this aspect of his claim not to have been made out on the balance of probabilities.

170. In his pleaded case the appellant contends that the Tribunal’s disbelief of his account of his interactions with the Georgian police was irrational or based on conjecture. It is also pleaded that the appellant was entitled to the benefit of the doubt on this issue because his account did not run counter to country of origin information, and section 28(7) of the 2015 Act/Article 4(5) of the Qualification Directive should have been applied. I have briefly considered the benefit of the doubt in the context of the standard of proof at paragraphs 156 and 157 above.

171. Hyland J. rejected these arguments. She was satisfied that the Tribunal had carefully addressed the question of plausibility including expressly considering the appellant’s responses at the hearing when it was put to him that his account was not plausible. She did not accept that the Tribunal had failed to take account of any relevant aspect of the appellant’s case including what he had said on the issue in his initial interview. The approach taken by the Tribunal in using the past actions of the Georgian police as a lens through which to evaluate the appellant’s claim was rational. As regards section 28(7) of the 2015 Act, she held the application of that section to provide the appellant with the benefit of the doubt did not arise in circumstances where he did not meet the condition set out in subparagraph (c), i.e. his statements were not found to be plausible.

172. I acknowledge that credibility and plausibility are distinct concepts although, in my view, there is a significant overlap between them. In international protection cases the credibility of an applicant relates to their inherent trustworthiness – in other words, can they be believed? The same can be said of other material relied on in support of a claim. Material emanating from bodies such as Amnesty International or Human Rights Watch is regarded as generally credible because of the international standing of those bodies over a lengthy period and their expertise in the analysis of human rights conditions in various countries. Material emanating from other sources may have to be scrutinized more closely because the

decision maker is not familiar with the source itself. Plausibility relates to the extent to which the account given is likely to be true which will frequently depend on the extent to which it is consistent with what is generally known about the country in question. In very broad terms, credibility is concerned with the source of the information whereas plausibility is concerned with its content. An otherwise credible witness might give evidence which is untrue because of flaws in their memory, understanding or perception of what they have witnessed. I accept that something might be implausible but nonetheless true and that something might be perfectly plausible and yet untrue.

173. In the context of international protection decision-making, the credibility of something will frequently correlate to its plausibility in light of other available information including information from independent sources. Applicants may be questioned during an interview or an oral hearing, as indeed this appellant was, and the credibility of their account may be tested before the decision-maker. However, cross-examination for the purposes of testing an applicant’s personal credibility in the way that occurs in inter partes litigation will usually neither arise nor be appropriate in international protection cases. Language and cultural differences mean that conclusions should not be lightly drawn based on an applicant’s demeanour. It is the content of the answers rather than the manner in which they are delivered which is important. The questioner, in this case the presenting officer, will rarely be in a position to call contrary evidence which directly contradicts the applicant’s account and cannot avail of procedural mechanisms such as discovery which might provide material upon which the applicant’s account can be undermined through cross-examination. Consequently, questioning will likely be based, as it was here, on the inherent implausibility of elements of an applicant’s account and its consistency or inconsistency with known country of origin information.

174. In this case a striking feature of the appellant’s account was that it went significantly further than simply claiming that the police had told him they could not protect him and included an allegation that the police had threatened to frame him for a crime he did not commit if he did not leave the country within fifteen days. The suggestion that a national police service would threaten to frame individuals for crimes they had not committed would not be consistent with a broadly functioning police service and would constitute a serious irregularity in the operation of the national legal system. Were such threats to be made by the Georgian police, it is reasonable to expect that they would be reflected in the various international reports submitted to the Tribunal. The absence from those reports of any suggestion that the Georgian police engage in such practices is notable and, in my view, cannot logically be relied on as supporting the plausibility of the appellant’s account.

175. The appellant argues that his account does not run counter to those reports as no mention is made in those reports of such a practice. In my view, this inference is misconceived. Bodies such as the US State Department, Amnesty International and Human Rights Watch (whose reports are referenced in the Tribunal decision) strive to produce material which provides a comprehensive overview of the human rights situation in the country with which they are concerned. Invariably such reports will include sections on the operation of local police and prosecution services as these are not only areas of potential abuse but are also important in the context of assessing the extent to which a state is both able and willing to protect its own citizens. Omission of something which, if it was occurring, would be a serious concern is generally a reasonable indication that it is not occurring. In this case the reports were not uniformly positive in their description of Georgian law enforcement services – particularly as regards the policing of political protest and the attitude of the security forces to the LGBTQ+ community – but there was no suggestion that the police force deliberately frames innocent people or that they were unwilling to provide protection to citizens from criminal behaviour. In that sense the appellant’s account ran counter to rather than being consistent with the available information regarding the police force in Georgia.

176. Finally, even if the analysis in the preceding paragraph were incorrect, it should be noted that not only are the conditions in section 28(7) cumulative, the factors listed in subparagraph (c) are also cumulative. The section does not require an applicant to be given the benefit of the doubt if their account does not run counter to the available country of origin information simpliciter. An applicant’s statements must be coherent and plausible and not run counter to the available country of origin information before sub-paragraph (c) is fulfilled. It would be a significant truncation of the provision if it were to be read as requiring a statement which did not run counter to the available country of origin information to be treated as plausible by virtue of that fact alone and regardless of its inherent plausibility or its consistency with the other evidence in the claim.

177. In the context of the particular facts, it is important to bear in mind that Z.M.’s evidence on this issue was not regarded as implausible solely because it was not supported by countryof-origin information. It was also inconsistent with his own account of the operation of the Georgian police and prosecution services in relation to events arising from the feud with G.C. The IPO had found this aspect of the appellant’s account to be implausible, so the appellant was aware that it was an issue he needed to address in his appeal. He gave evidence in person and had a full opportunity to deal with the issue. As outlined above, it was expressly put to him at the oral hearing by the presenting officer that his account of his exchanges with the local police was not plausible and he was unable to put the matter any further in response. His evidence, which was disbelieved, was not consistent with the available information about the Georgian police service. There was no other evidence which supported this aspect of his account. A decision maker is entitled not to accept evidence which it regards as implausible, which is inconsistent with the other evidence in the case and which is not supported by country of origin information. I also reject this ground of appeal.

Conclusions

178. It will be apparent from the analysis above that all of the grounds of appeal advanced by the appellants are rejected. I should add for the sake of completeness that even if I were incorrect in my conclusions on the central issue in the appeal (whether a finding that there was a real risk of serious harm necessarily imported a finding that state protection would not be available to the appellants in respect of that harm), I would nonetheless uphold the trial judges’ decisions to refuse to grant relief on this ground.

179. This is because it is clear from the decisions under challenge that the Tribunal did not consider state protection as part of its analysis of whether there was a real risk of serious harm but considered it subsequent to that initial finding and concluded in each case that because state protection was available the appellants did not satisfy the criteria to be granted subsidiary protection. The appellants’ argument is a technical one as to the sequencing of the analysis of a claim for protection and the point in that sequence at which the availability of state protection should be considered. If the appellants were correct, then logically the only difference it would make is that the Tribunal should not have formally concluded that there was a real risk of serious harm before determining whether state protection was available. It does not change the validity of the Tribunal’s conclusion that state protection was available in both cases. On the appellants’ logic, this could only mean that the Tribunal should not have reached the conclusion that there was a real risk of serious harm. Either way, the appellants did not satisfy the criteria for eligibility for subsidiary protection because the protection of their own state was available to them in respect of the harm they feared. In those circumstances even if the appellants’ argument was correct, as judicial review is a discretionary remedy I would decline relief on the basis that, however it is expressed, the Tribunal’s analysis necessarily leads to the same conclusion and the rejection of their claims.

180. In circumstances where the appellants have been unsuccessful in their appeals, the respondents would normally be entitled to an order for their costs in each appeal. However, as both appeals were heard together and the same lawyers appeared on both sides in both cases, I propose making an order that the respondents should recover their costs in both appeals but that those costs be limited so as to include only a single set of costs in respect of the hearing of the appeals. If either party wishes to contend for a different order, they have

14 days within which to file written submissions (limited to 1,000 words) on costs. The opposing party will then have 10 days within which to file a reply (also limited to 1,000 words). If no submissions are received within this time frame the court will proceed to make the order as proposed.

181. My colleagues Whelan J. and Meenan J. have read this judgment in advance of its delivery and have indicated that they agree with it.