[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"court-court-of-justice-of-the-european-union":3},{"court":4,"decisions":10},{"id":5,"name":6,"country":7,"level":8,"jurisdiction":9},50,"Court of Justice of the European Union","eu","supreme","general",[11,24,33,42],{"id":12,"ecli":13,"caseNumber":14,"courtId":5,"courtName":15,"decisionDate":16,"publicationDate":16,"fullText":17,"language":18,"source":19,"sourceUrl":20,"summary":15,"slug":21,"metadata":22},"262","ECLI:EU:C:2026:11","C-560\u002F24","","2026-01-15T00:00:00.000Z","OPINION OF ADVOCATE GENERAL\n\nNORKUS 1 Case C‑560\u002F24 [Besthame ( i )]\n\nR.S.\n\nv\n\nMinister for Justice\n\n(Request for a preliminary ruling from the Court of Appeal (Ireland)) \n\n(Reference for a preliminary ruling – Citizenship of the Union – Right to move and reside freely within the territory of the Member States – Directive 2004\u002F38\u002FEC – Family members of a Union citizen – Derived right of residence – Subsequent naturalisation of the family member – Article 35 – Fraud or abuse of rights and\u002For marriage of convenience)\n\n**I.** **Introduction**\n\n1\\. The legal question raised by the present reference for a preliminary ruling is the following: can the competent authorities of a Member State investigate and, if appropriate, make a determination or reach a conclusion that a person who benefited from a derived right of residence in the past as a family member of a Union citizen, under Directive 2004\u002F38\u002FEC, ( 2 ) committed an abuse of rights or fraud, even though that person has, in the meantime, acquired the nationality of that Member State and his or her residence in the Member State is therefore no longer based on that directive?\n\n2\\. That question is posed by the Court of Appeal (Ireland) in proceedings between R.S. (‘the appellant in the main proceedings’), a naturalised Irish national, and the Minister for Justice (Ireland) (‘the Minister for Justice’) concerning the finding made by the latter that, before acquiring Irish nationality, the appellant in the main proceedings, then a third-country national, had contracted a marriage of convenience with a Union citizen or had provided false or misleading information for the purpose of obtaining a residence card of a family member of a Union citizen.\n\n3\\. Consideration of the question will lead the Court to interpret, for the second time, Article 35 of Directive 2004\u002F38. Although the Court has already made mention of the abuse of rights or fraud inter alia in its case-law on the free movement of persons, ( 3 ) the first reference for a preliminary ruling seeking an interpretation of that article is that which gave rise to the judgment in *McCarthy and Others*. ( 4 ) In that judgment, the Court held that a Member State cannot make the right of entry of a third-country national subject to that national obtaining a visa in advance where he or she holds a residence card of a family member of a Union citizen. It stated that that article does not allow measures which, in pursuit of an objective of general prevention, stop family members from entering the territory of a State without a visa. ( 5 )\n\n4\\. However, the present case gives the Court the opportunity to rule on the interpretation of Article 35 of Directive 2004\u002F38 from a different angle. The interpretation sought by the referring court is concerned not with the determination of an abuse of rights as such, but rather with the applicability of that directive to determine it, as well as with the legal consequences which should be drawn from the finding of the existence of such abuse, which was committed in the past and has been established definitively by the competent authorities of a Member State. In the case in the main proceedings, the referring court has found there to have been a marriage of convenience, and therefore that classification of the marriage is not at issue in the present case.\n\n5\\. In that context, the referring court has doubts as to whether Article 35 of Directive 2004\u002F38 is applicable where the person concerned is no longer a ‘beneficiary’, within the meaning of that directive, in the host Member State of the rights conferred by the directive, and, if so, whether that provision allows the competent authorities to investigate whether a marriage of convenience exists and to find that one exists, without other immediate legal consequences being drawn.\n\n6\\. In the present Opinion, I will propose inter alia that the Court find that the situation in which the person concerned is no longer a ‘beneficiary’, under Directive 2004\u002F38, falls within the scope of that directive.\n\n**II.** **Legal framework**\n\n**A.** **European Union law**\n\n7\\. In the context of the present Opinion, I will refer to recitals 25 and 28 of Directive 2004\u002F38 and to Article 2(1) and (2)(a), Article 3(1), Article 15(1) and (3), Article 30(1) and (3), Article 31(1) and (3), Article 35 and Article 36 of that directive.\n\n**B.** **Irish law**\n\n8\\. Directive 2004\u002F38 was transposed into Irish law by the European Communities (Free Movement of Persons) Regulations 2015 (S.I. No 548 of 2015), in the version thereof applicable to the dispute in the main proceedings (‘the 2015 Regulations’).\n\n9\\. Regulation 27 of the 2015 Regulations, entitled ‘Cessation of entitlements’, provides, in paragraphs 1, 2 and 4 thereof: ‘(1) The Minister may revoke, refuse to make or refuse to grant, as the case may be, any of the following where he or she decides, in accordance with this Regulation, that the right, entitlement or status, as the case may be, concerned is being claimed on the basis of fraud or abuse of rights: … (*b*) a residence card … … (2) Where the Minister suspects, on reasonable grounds, that a right, entitlement or status of being treated as a permitted family member conferred by these Regulations is being claimed, or has been obtained, on the basis of fraud or abuse of rights, he or she shall be entitled to make such enquiries and to obtain such information as is reasonably necessary to investigate the matter. … (4) In this Regulation, “abuse of rights” shall include a marriage of convenience …’\n\n10\\. Regulation 28 of those Regulations, entitled ‘Marriages of convenience’, provides: ‘(1) The Minister, in making his or her determination of any matter relevant to these Regulations, may disregard a particular marriage as a factor bearing on that determination where the Minister deems or determines that marriage to be a marriage of convenience. (2) Where the Minister, in taking into account a marriage for the purpose of making a determination of any matter relevant to these Regulations, has reasonable grounds for considering that the marriage is a marriage of convenience, he or she may send a notice to the parties to the marriage requiring the persons concerned to provide, within the time limit specified in that notice, such information as is reasonably necessary, either in writing or in person, to satisfy the Minister that the marriage is not a marriage of convenience. …’\n\n**III.** **The facts of the dispute in the main proceedings, the question referred for a preliminary ruling and the procedure before the Court**\n\n11\\. R.S., who was born in a third country, entered Ireland in the course of 2002 as the holder of a student residence permit. In 2010, 16 days before the expiry of that residence permit, R.S., then a third-country national, married a Union citizen who had exercised her right to move to and reside in Ireland. He was subsequently issued a five-year residence card as a family member of a Union citizen.\n\n12\\. In 2015, R.S. acquired Irish citizenship. Since then, his residence in Ireland has been based on that citizenship. In 2018, R.S. and his wife divorced. In 2019, a third-country national applied for residence in Ireland on the ground that she was the mother of a child, an Irish citizen, of whom R.S. was the biological father. That application prompted an investigation to determine whether the marriage contracted in 2010 was a marriage of convenience.\n\n13\\. By a decision of 18 December 2019, the Minister for Justice ‘revoked’ the residence card issued in 2010 on the ground that, in support of his residence card application, R.S. had produced misleading documents and that the marriage contracted in 2010 had been a marriage of convenience. Following an application for review submitted by R.S., that decision was upheld by a decision of 8 September 2020.\n\n14\\. On 1 February 2022, the Minister for Justice did, however, adopt a new decision further to correspondence with R.S.’s solicitors. By that decision, she rescinded the decision of 8 September 2020, set aside the decision of 18 December 2019 and found that R.S. had produced false or misleading documents or information and contracted a marriage of convenience to obtain a status or a right to which he would not otherwise have been entitled under Directive 2004\u002F38. The Minister took the view that ‘any entitlement or status conferred under [that] directive from [that] marriage … [is] deemed withdrawn from the outset’.\n\n15\\. R.S. brought before the High Court (Ireland) an application for annulment (*certiorari*) of the three decisions of 8 September 2020, 13 February 2020 and 1 February 2022 on the ground that the Minister for Justice had acted ultra vires. As an Irish citizen, R.S. was no longer covered by any provision of the 2015 Regulations or of Directive 2004\u002F38, which could not therefore authorise the Minister for Justice to adopt those decisions. That action was dismissed by the High Court by a judgment of 18 May 2023.\n\n16\\. R.S. then referred the matter to the Court of Appeal, the referring court.\n\n17\\. That court points out, first and foremost, that the decision of 1 February 2022 must be understood as not entailing the revocation or the refusal of a right of residence, but rather as containing a ‘determination’, a ‘finding’ or a ‘conclusion’ as to a past state of affairs or the past conduct of R.S. It adds that that decision suggests that such a determination or finding is conceivable in the context of a future re-assessment of R.S.’s Irish citizenship status, while acknowledging that any future re-assessment would take account of all circumstances and of R.S.’s fundamental rights. With regard to the re-assessment of R.S.’s Irish citizenship, the referring court explains that, by reason of recent decisions of the Supreme Court (Ireland), there is no constitutional procedure that currently allows an investigation to be conducted with a view to revoking Irish citizenship.\n\n18\\. In view of the arguments raised before it, the referring court asks, in the first place, about the scope of Directive 2004\u002F38 and, in particular, of Article 35 thereof.\n\n19\\. In that regard, whilst noting differences between the circumstances of the present case and those of the case which gave rise to the judgment in *Lounes* , ( 6 ) the referring court considers that it could be inferred from that judgment that Directive 2004\u002F38 ceases to apply to a third-country national who is a family member of a Union citizen once that national acquires the citizenship of the host Member State and, as a result, no longer satisfies the definition of a ‘beneficiary’ within the meaning of Article 3(1) of that directive. That said, in the judgment in *Chenchooliah* , ( 7 ) the Court held that that directive was applicable to a decision to expel a third-country national who resided in the host Member State as a family member of a Union citizen, even though that third-country national was no longer a ‘beneficiary’ within the meaning of Article 3(1) of the same directive.\n\n20\\. The referring court wonders whether, by analogy, Directive 2004\u002F38 continues to govern the situation in which the competent authority of a Member State is seeking to determine whether a person initially obtained, by an abuse of rights or fraud, the benefit of a right of residence on the basis of that directive, at a time when that person is no longer a ‘beneficiary’ within the meaning of Article 3(1) of that directive.\n\n21\\. In the second place, the referring court observes that Regulation 27(1) and (2) of the 2015 Regulations authorises the Minister for Justice, first, to revoke, refuse to make or refuse to grant a residence card applied for on the basis of a fraud or an abuse of rights and, second, to investigate the suspected fraud or abuse of rights. Whilst the revocation cannot have retroactive effect, the investigation could cover both current or pending residence card applications as well as residence cards obtained in the past. However, the question arises as to whether the power of investigation can be exercised autonomously, outside any specific action such as the revocation or the refusal of a right of residence, a prosecution for a criminal offence, an expulsion procedure or a procedure for revocation of Irish citizenship.\n\n22\\. In the referring court’s view, such an autonomous power may be implicitly contained in Regulation 27(2) of the 2015 Regulations. It may be justified in the light of the purpose and of the context of those regulations taken as a whole, including the rules governing citizenship of the Union, which require that there be a robust system for the prevention, detection and eradication of fraud and abuses of rights. To be able, if appropriate, to interpret Regulation 27(2) of the 2015 Regulations in a manner consistent with EU law, that court considers it necessary to determine whether Directive 2004\u002F38 applies to a naturalised national of a Member State solely in so far as that directive authorises that Member State to investigate a marriage of convenience entered into at a time when that national enjoyed a right of residence in the Member State under that directive.\n\n23\\. It is in those circumstances that the Court of Appeal, by order of 2 July 2024, received at the Registry of the Court of Justice on 19 August 2024, decided to stay the proceedings and to refer the following question to the Court of Justice for a preliminary ruling: ‘Does Directive [2004\u002F38] apply to a person who previously obtained the benefit of [a] derived [right of] residence in a Member State by virtue of being a spouse of an EU national exercising Treaty rights in the host State but who has more recently become a citizen in the host State and is no longer the beneficiary of any derived benefit in the host State under the directive, solely for the purpose of investigating and (if appropriate) making a determination or reaching a conclusion that he [or she] engaged in a fraud or abuse of rights and\u002For a marriage of convenience in the past within the meaning of Article 35 of the directive in order to obtain a benefit under the directive?’\n\n24\\. Written observations were submitted by R.S., the Minister for Justice, Ireland, the German Government and the European Commission. Oral argument was presented at the hearing of 15 October 2025 on behalf of R.S., Ireland, the German Government and the Commission.\n\n**IV.** **Analysis**\n\n25\\. By its single question, the referring court asks, in essence, whether Article 35 of Directive 2004\u002F38 must be interpreted as allowing the competent national authorities to investigate and, if appropriate, make a determination or reach a conclusion that a person who previously benefited from a derived right to move freely under that directive committed a fraud or an abuse of rights, where the residence of that person in the Member State concerned is no longer based on that directive, including where he or she has in the meantime acquired the nationality of that Member State. ( 8 )\n\n26\\. In their written and oral submissions, the parties to the main proceedings and the interested parties disagree as to the interpretation of Article 35 of Directive 2004\u002F38.\n\n27\\. Generally speaking, the appellant in the main proceedings and the German Government have doubts that that directive is applicable to a person who has benefited in a Member State, as a third-country national, from a derived right of residence as the spouse of a migrant Union citizen, where that person has since then acquired the nationality of the host Member State. They are of the view, in essence, that such a person is no longer covered by the concept of a ‘beneficiary’ within the meaning of Article 3(1) of Directive 2004\u002F38, with the result that that directive no longer applies to him or her. In particular, the German Government takes the view that the determination of the existence of an abuse of rights or a fraud falls within the scope of national law. They therefore propose that the question put by the referring court be answered in the negative. By contrast, the Minister for Justice, Ireland and the Commission argue that Article 35 of Directive 2004\u002F38 is applicable in the present case. They all argue, inter alia, that that provision is an expression of the general principle that abuse of rights is prohibited. They therefore propose that the question referred for a preliminary ruling be answered in the affirmative.\n\n28\\. In order to answer the referring court’s questions, I will begin my analysis with some general considerations (Section A). I will continue by dispelling any doubts as to the applicability of Directive 2004\u002F38 and, in particular, of Article 35 thereof to the situation at issue in the main proceedings (Section B), and conclude by examining the scope of the powers conferred on Member States by that provision (Section C).\n\n**A.** **Some general considerations on the applicability of Directive 2004\u002F38**\n\n29\\. It should be recalled from the outset that, as may be seen from recitals 3 and 4 of Directive 2004\u002F38, that directive aims to facilitate the exercise of the primary and individual right to move and reside freely within the territory of the Member States that is conferred directly on Union citizens by Article 21(1) TFEU, and that it aims in particular to strengthen that right. ( 9 ) Recital 5 of Directive 2004\u002F38 states that the right should, if it is to be exercised under objective conditions of dignity, also be granted to the family members of those citizens, irrespective of nationality. ( 10 ) However, according to settled case-law of the Court, Directive 2004\u002F38 does not confer any autonomous right on family members of a Union citizen who are third-country nationals. Thus, any rights that may be conferred on those nationals by that directive are derived from the rights which the Union citizen concerned enjoys as a result of having exercised his or her freedom of movement. ( 11 ) The Court has repeatedly held that, pursuant to Article 3(1) of Directive 2004\u002F38, Union citizens who move to or reside in a Member State other than that of which they are a national, and their family members as defined in Article 2(2) of that directive who accompany or join them, fall within the scope of the directive *and* are beneficiaries of the rights conferred by it. ( 12 )\n\n30\\. In the present case, given that, following his marriage in 2010 to a national of a Member State residing in Ireland, the appellant in the main proceedings resided in that Member State as the spouse of a Union citizen who had exercised her freedom to move to and reside in a Member State other than that of which she is a national, it is established that *he benefited* from a derived right of residence, under Directive 2004\u002F38, *and fell with the scope* of that directive. However, it is apparent from the order for reference that, in 2015, he acquired Irish nationality and, as a result, has since then resided in Ireland as an Irish national. A few years later, in 2018, the couple divorced. It is therefore quite clear that, since that directive governs only the conditions of entry and of residence of a Union citizen and his or her family members in the host Member State, the appellant in the main proceedings no longer satisfies the definition of a ‘beneficiary’ of the directive within the meaning of Article 3(1) thereof. ( 13 )\n\n31\\. It follows from the foregoing that when he acquired the rights conferred by Directive 2004\u002F38 and over the period to which the investigation into the existence of a marriage of convenience related, that is to say, between 2010 and 2015, the appellant in the main proceedings had the status of a beneficiary of that directive, ( 14 ) as the spouse of a Union citizen who had exercised her freedom of movement by moving to and residing in a Member State other than that of which she was a national. However, by virtue of his naturalisation by Ireland in the course of 2015, ( 15 ) his legal status changed under both national law and that directive, with the result that, in accordance with the judgment in *Lounes* , ( 16 ) when that investigation was opened in 2019, the same directive no longer governed his residence in that Member State. ( 17 )\n\n32\\. Accordingly, must the view be taken that, since the appellant in the main proceedings is no longer a ‘beneficiary’ of the *derived rights of entry and of residence* in the host Member State, within the meaning of Article 3(1) of Directive 2004\u002F38, that directive is no longer applicable?\n\n33\\. I do not believe so.\n\n34\\. As I will demonstrate in what follows, the loss of the status of a beneficiary within the meaning of Article 3(1) of Directive 2004\u002F38 does not presuppose that that directive ceases to apply where, as in the present case, the competent national authorities find there to have been abuse or fraud of the rules laid down in that directive which was *committed when* the appellant in the main proceedings had the status of beneficiary. In that regard, it should be observed that the Court has taken care to mark a distinction between the scope of the directive and the status of beneficiary under that provision. ( 18 ) It therefore follows, in my view, that whilst Union citizens and their family members who exercise the rights conferred by Directive 2004\u002F38 fall within the scope of that directive *and* are beneficiaries of it, within the meaning of the abovementioned provision, persons who no longer exercise those rights because they have lost their status of beneficiaries may, nevertheless, continue to fall within the scope of the directive since, as I will demonstrate in the following points of the present Opinion, that directive contains provisions the application of which is not conditional upon the person concerned having that status at the time of their application. ( 19 )\n\n35\\. That being said, in the following points, I will focus my examination on whether Directive 2004\u002F38, in particular Article 35 thereof, is applicable in the present case.\n\n**B.** **The applicability of Article 35 of Directive 2004\u002F38**\n\n36\\. In order to determine whether Article 35 of Directive 2004\u002F38 is applicable in the present case, it is necessary to determine its scope *ratione temporis* , that is to say, to ascertain whether that provision applies when the person concerned had, in the past, the status of beneficiary under that directive, within the meaning of Article 3(1) thereof, but had already lost that status at the time when the former provision is said to apply.\n\n37\\. According to settled case-law, in interpreting a provision of EU law, it is necessary to consider not only its wording, but also the context in which it occurs and the objectives pursued by the rules of which it is part. ( 20 ) The origins of a provision of EU law may also provide information relevant to its interpretation. ( 21 ) With that in mind, I will undertake a literal, contextual and teleological interpretation of that provision.\n\n*1.*  *Literal interpretation*\n\n38\\. It should be recalled, at the outset, that Article 35 of Directive 2004\u002F38, entitled ‘Abuse of rights’, provides that ‘Member States may adopt the necessary measures to refuse, terminate or withdraw any right conferred by [that] Directive in the case of abuse of rights or fraud, such as marriages of convenience. Any such measure shall be proportionate and subject to the procedural safeguards provided for in Articles 30 and 31.’\n\n39\\. I note, first, that there is nothing in the wording of that provision to support the view that, by the provision, the EU legislature intended to impose a time limit on the possibility for Member States to adopt ‘the necessary measures to refuse, terminate or withdraw any right conferred by [the] Directive in the case of abuse of rights or fraud’. ( 22 ) In other words, that wording does not state that a Member State is required to adopt only ‘the necessary measures’ in connection with a right that currently has legal effects under Directive 2004\u002F38. Providing for such a time limit would amount, in practice, to imposing an additional element not laid down in the wording of that provision. On the contrary, the absence of such a time limit means that that possibility is therefore understood to be an option for Member States to adopt such measures, irrespective of whether the person concerned is exercising or has exercised a right conferred on the basis of Directive 2004\u002F38.\n\n40\\. Second, it is apparent from the wording of Article 35 of Directive 2004\u002F38, which uses the past participle ‘conferred’, that the EU legislature is referring, inter alia, to the rights of which the person concerned is or has been a beneficiary under that directive. ( 23 )\n\n41\\. Third, I note that the use in that provision of the verb ‘refuse’ includes the refusal to grant entry to or residence in the host Member State to the persons covered by Article 3(1) of Directive 2004\u002F38, where those persons acquired the rights conferred by that directive by an abuse of rights or by fraud, and did not therefore satisfy the conditions to be beneficiaries of such rights under the directive. ( 24 ) Consequently, what is at issue here is a (potential) future right of entry or of residence which has not yet arisen, since that person does not satisfy the conditions required by Directive 2004\u002F38 to have the status of ‘beneficiary’ within the meaning of that provision.\n\n42\\. Fourth, I note that the use of the verbs ‘terminate’ and ‘withdraw’ in Article 35 of Directive 2004\u002F38 suggests that that provision covers the termination or the withdrawal of a right conferred by that directive, irrespective of whether or not the effects of that right have been exhausted. ( 25 ) Those two verbs can therefore be retroactive in nature in so far as they may entail the evaporation of both certain future effects and also certain past effects of such a right. ( 26 ) The use of two separate verbs appears, in principle, to reflect the intention of the EU legislature to cover both national measures that allow a right to be terminated *ex nunc* (such as inter alia the termination of a residence card) or to be withdrawn *ex tunc*. ( 27 )\n\n43\\. Therefore, it follows from that interpretation that the choice made by the EU legislature to use the words ‘refuse, terminate or withdraw any right conferred by [that directive]’ clearly expresses the intention to *cover* the right of entry or residence at issue *with regard to its timing* (future, current or past) where that right is ‘conferred by [the directive]’, thus supporting the view that that same directive is applicable regardless of the fact that that right is no longer being exercised.\n\n44\\. I am therefore of the view that the wording of Article 35 of Directive 2004\u002F38 means that that provision must be regarded as being applicable to measures adopted in respect of a right conferred by that directive over a period during which the person holding that right was the ‘beneficiary’, within the meaning of the directive, irrespective of the current position of the person concerned.\n\n45\\. It is, however, necessary to examine whether or not the normative context in which Article 35 of Directive 2004\u002F38 occurs corroborates that interpretation.\n\n*2.*  *Contextual interpretation*\n\n*(a)*  *Internal or schematic interpretation*\n\n46\\. With regard to the context in which Article 35 of Directive 2004\u002F38 occurs, I note, in the first place, that the interpretation set out in the preceding points is not invalidated by the wording of recital 28 of that directive, under which, ‘to guard against abuse of rights or fraud, notably marriages of convenience or any other form of relationships contracted for the sole purpose of enjoying the right of free movement and residence, Member States should have the possibility to adopt the necessary measures.’ The broad wording of that recital, like that of Article 35, does not lay down any time limit for the adoption of such measures by Member States. In that regard, I note that the wording of recital 28 refers to ‘marriages of convenience or [to] any other form of relationships contracted’, thus suggesting that the measures which Member States ‘should have the possibility to adopt’ under Article 35 may be retroactive in nature, since such marriages or relationships have in the majority of cases already taken place when that provision may be applied.\n\n47\\. In the second place, I note that Article 1 of Directive 2004\u002F38, entitled ‘Subject’, provides under point (a) that that directive, inter alia, lays down ‘the conditions governing the exercise of the right of free movement and residence within the territory of the Member States by Union citizens and their family members’. Therefore, since Article 35 of Directive 2004\u002F38 applies to the situations in which the conditions governing the exercise of the right of free movement and residence laid down by that directive *were met only formally, such that those conditions were not respected* , Article 35 of that directive falls within the scope of Article 1 thereof.\n\n48\\. In that regard, as is apparent from the scheme of Directive 2004\u002F38, in particular from Chapters II to IV on the conditions governing the exercise of the rights of residence laid down therein, that directive introduces a system that covers the development of the situation of Union citizens and their family members in the host Member State as well as that of the rights that the same directive confers on them, from the time of their arrival in a Member State other than that of which they are a national until their departure from that Member State. ( 28 ) That system encompasses both the different stages of the exercise of the freedom of movement by a citizen and his or her family members (entry, residence or departure) and the rights of which those persons are beneficiaries and which correspond to each of the stages concerned (right of entry, right of residence, right to retain the right of residence or right of exit). Thus, if, *at a particular time* (including, in the present case, *the time at which the right in question is improperly acquired*), the Union citizen and\u002For his or her family members no longer satisfy the conditions laid down in Directive 2004\u002F38, entailing the loss (even the retroactive loss) of their status as ‘beneficiaries’ within the meaning of Article 3(1) thereof and, therefore, of their rights of entry to and of residence in the host Member State, this does not mean however that other provisions of that directive are not applicable to them. ( 29 )\n\n49\\. In that developing context of the concept of ‘beneficiary’ within the meaning of Article 3(1) of Directive 2004\u002F38, I consider it important to note that Article 35 thereof concerns a right of movement and of residence which has been exercised for a certain period of time, a right from which the person concerned should never have benefited under that directive since he or she only artificially satisfied the conditions to which the exercise of that right is subject. ( 30 ) As the Commission has observed, Article 35 of Directive 2004\u002F38 is applicable to *legal situations arising at the time the right conferred by that directive is obtained* so that such situations may be characterised in the light of the practices or the facts which prove that that right was acquired improperly, ( 31 ) even if the person concerned is no longer exercising that right and no longer has the status of ‘beneficiary’ within the meaning of Article 3(1) thereof.\n\n50\\. In the third and final place, I would add that the ‘Final provisions’ contained in Chapter VII of Directive 2004\u002F38, of which Article 35 forms an integral part, concern the effective implementation of that directive in its entirety. ( 32 ) Without even having to examine those provisions in detail, I note, in particular, that it follows from the inclusion of Articles 35 and 36 of the directive in Chapter VII thereof that they are horizontal provisions relating, respectively, to the power of Member States and to the obligation on Member States to adopt measures in case of infringement of the provisions of the directive. ( 33 )\n\n51\\. It is true that, unlike Article 35, Article 36 concerns Union citizens and their family members who, in principle, lawfully exercise the rights conferred on them by Directive 2004\u002F38, but who have failed to comply with certain administrative formalities laid down therein. ( 34 ) In addition, the measures set out in those two provisions are different in nature: individual or specific measures on the one hand, and general measures on the other. ( 35 ) However, despite those differences, I see no reason why those two provisions are not applicable to cases where the person concerned is no longer a beneficiary of the directive. On the contrary, as I have noted, those two provisions fall within the scope of other provisions of that directive. ( 36 ) To hold otherwise would mean that neither the measures referred to in Article 35 nor the sanctions adopted by Member States pursuant to Article 36 could be applied to a person who no longer benefits from the rights conferred by Directive 2004\u002F38, even though an abuse of rights or a failure to fulfil the obligations imposed by that directive did occur when those rights were acquired or exercised, which would render those provisions meaningless.\n\n*(b)*  *External interpretation or travaux préparatoires*\n\n52\\. I note that in the *travaux préparatoires* for Directive 2004\u002F38, Article 35 was not included in the Commission’s initial proposal. ( 37 ) That article was introduced by the Council ‘in order to clarify that Member States *may* refuse, terminate or withdraw *any right conferred* by [that directive] in the case of abuse of rights or fraud’. ( 38 ) This confirms the intention of the EU legislature, as is apparent from recital 28 of the directive, to *allow* Member States ‘to adopt the necessary measures’‘to guard against abuse of rights or fraud, notably marriages of convenience’, regardless of the current position of the person concerned who is benefiting or has benefited from a right conferred by that directive.\n\n53\\. This is likewise borne out by the Commission’s non-binding guidelines of 2009 ( 39 ) and of 2023, ( 40 ) which refer to the possibility for Member States to adopt the necessary measures in cases of abuse or fraud, under Article 35 of Directive 2004\u002F38, ‘at any point [in] time’, whether in the context of ‘the refusal to confer’ the rights afforded by that directive or ‘the termination or withdrawal’ of such rights. ( 41 )\n\n54\\. Accordingly, the literal interpretation outlined in point 44 of the present Opinion, namely that Article 35 of Directive 2004\u002F38 is applicable to a situation such as that in the main proceedings, is supported by the normative context and the scheme of that directive in which that provision appears.\n\n55\\. That being said, the referring court also asks whether it is possible to rely, by analogy, on the guidance provided in the judgment in *Chenchooliah* ( 42 ) in order to hold that Article 35 of Directive 2004\u002F38 is applicable to a situation such as that at issue in the main proceedings. It is to that question which I will now turn.\n\n*(c)*  *The lessons to be learned from the case-law deriving from the judgment in Chenchooliah*\n\n56\\. In that judgment, the Court observed, first, that the concept of ‘beneficiary’ within the meaning of Article 3(1) of Directive 2004\u002F38 is a ‘dynamic’ concept in that, even though acquired in the past, the status of beneficiary may subsequently be forfeited if the requirements laid down by that provision are no longer met. ( 43 ) Second, the Court stated that that directive does not contain only rules governing the conditions under which one of the various types of residence rights it makes provision for may be obtained and the conditions to be met in order to be able to continue to enjoy the rights concerned. ( 44 ) It added that the directive also lays down a set of rules intended to govern the situation arising in which entitlement to one of those rights is lost, inter alia where the Union citizen leaves the host Member State. ( 45 )\n\n57\\. In that context, the Court, first, found that, in a situation in which a Union citizen returned to the Member State of which he or she is a national and therefore no longer exercises, in the host Member State, his or her right of free movement under EU law, the third-country national, who is the spouse of that Union citizen, no longer has the status of ‘beneficiary’ within the meaning of Article 3(1) of that directive, where that spouse remains in the host Member State and no longer resides with his or her spouse. ( 46 ) Second, the Court ruled that, even though the effect of the loss of that status is that the third-country national concerned no longer has the rights of movement and residence in the territory of the host Member State which that national held for a certain period of time, as he or she no longer meets the requirements to which those rights are subject, that loss does not mean, however, that Directive 2004\u002F38 is no longer applicable where the host Member State takes a decision to expel that person on such a ground. ( 47 )\n\n58\\. It is true that, in paragraph 64 of the judgment in *Chenchooliah* , the Court recalled, relying on paragraph 95 of the judgment in *Metock* , that ‘from the time when a third-country national who is a family member of a Union citizen derives rights of entry and residence in the host Member State from Directive 2004\u002F38, that State may restrict those rights only in compliance with Articles 27 and 35 of the directive’. However, in paragraph 67 of the judgment in *Chenchooliah* , it took the view that the lessons to be drawn from paragraph 95 of the judgment in *Metock* were not applicable to the situation at issue in the main proceedings in the case which gave rise to the judgment in *Chenchooliah* , in which the third-country national was no longer a beneficiary of that directive. ( 48 )\n\n59\\. I do not believe that that statement by the Court can be interpreted as meaning that it sought to link the application of Article 35 of Directive 2004\u002F38 to the period during which the person concerned is a ‘beneficiary’, within the meaning of Article 3(1) thereof. I understand that paragraph rather to mean that the Court simply drew a distinction between the situations at the origin of those two judgments: in the first situation at issue, in the case which gave rise to the judgment in *Metock* , the third-country national concerned benefited from a (current) derived right of residence ( 49 ) in the host Member State, within the meaning of the directive, whereas in the second situation at issue, in the case which gave rise to the judgment in *Chenchooliah* , the third-country national concerned had lost that (past) derived right. ( 50 ) Accordingly, since the two situations were different, the question remained as to whether the loss of the status of ‘beneficiary’ within the meaning of Article 3(1) of Directive 2004\u002F38 meant that an expulsion decision, taken essentially on the ground that that national had been refused a derived right of residence under that directive, was governed not by that directive, but by the national law applicable outside its scope. ( 51 )\n\n60\\. As I have just set out above, the Court held, in paragraph 69 of the judgment in *Chenchooliah* , that that question had to be answered in the negative, whilst – in paragraph 79 of the judgment – isolating the status of ‘beneficiary’ within the meaning of the abovementioned provision from the applicability of other provisions of Directive 2004\u002F38, namely, inter alia, Article 15 thereof. ( 52 )\n\n61\\. In that context, the question arises whether the lessons from that case-law support my contextual and schematic interpretation of Article 35 of that directive.\n\n62\\. In that regard, it should be observed, as the Court noted in paragraph 70 of the judgment in *Chenchooliah* , that Directive 2004\u002F38 ‘does not contain only rules governing the conditions under which one of the various types of residence rights it makes provision for may be obtained and the conditions to be met in order to be able to continue to enjoy the rights concerned’, ( 53 ) but also lays down ‘a set of rules intended to govern the situation arising in which entitlement to one of those rights is lost, inter alia where the Union citizen leaves the host Member State’ (emphasis added).\n\n63\\. It follows from the use of the words ‘inter alia’ in paragraph 70 of the judgment in *Chenchooliah* that ‘the set of rules’ laid down in Directive 2004\u002F38 and ‘intended to govern the situation arising in which entitlement to one of those rights is lost’ is by no means restricted to in a situation in which the Union citizen and\u002For his or her family members leave(s) the host Member State. It is sufficient to recall that, in the context and the scheme of that directive, the legal situations arising when the right conferred by the directive is obtained can evolve up until when they cease to exist *for reasons other than the departure of the beneficiary* of the directive, and inter alia on account of the naturalisation of the person concerned by the host Member State. Such legal situations can also cease to exist in the light of circumstances which prove that that right was acquired improperly or fraudulently, entailing their termination or withdrawal.\n\n64\\. It is therefore arguable that the finding, made by a national authority, of the existence of a marriage of convenience could entail, inter alia, the ‘formal and retroactive withdrawal’ of the derived rights which a third-country national in reality improperly or fraudulently derived from Directive 2004\u002F38, thus entailing the ‘retroactive loss’ of the ‘formal’ benefit of those rights, regardless of whether, as in the present case, those rights were exercised in the past, they are still being exercised or they may be exercised in the future. ( 54 )\n\n65\\. It therefore follows that it is possible to rely, by analogy, on the guidance provided in the judgment in *Chenchooliah* to take the view that Article 35 of Directive 2004\u002F38 is applicable to a situation such as that at issue in the main proceedings.\n\n*3.*  *Teleological interpretation*\n\n66\\. I would point out, in the first place, that, in view of *the objective* pursued by Directive 2004\u002F38, namely that of facilitating the exercise of the primary and individual right to move and reside freely within the territory of the Member States that is conferred directly on Union citizens by Article 21(1) TFEU, ( 55 ) its provisions cannot be interpreted narrowly and must not, in any case, be deprived of a large proportion of their content or their effectiveness. Thus, in so far as Article 35 of Directive 2004\u002F38 allows Member States to restrict the rights of Union citizens and their family members to move and reside freely within the territory of the Member States, that provision makes reference to the procedural safeguards laid down in that directive to ensure a high level of protection. ( 56 ) Accordingly, contrary to what the German Government claims, a restrictive interpretation of that provision would render those procedural safeguards meaningless, thus denying that protection to persons accused of improper or fraudulent conduct, whereas those safeguards are specifically intended to ensure the effectiveness of those rights and, therefore, contribute to the objective of that directive.\n\n67\\. To argue the contrary and take the view that Article 35 of Directive 2004\u002F38 is not applicable where a Union citizen and his or her family members no longer have the status of beneficiaries of that directive would amount, contrary to the will of the EU legislature, first, to disregarding the requirement of proportionality in the adoption of the ‘necessary measures’, as set out in that article, and, second, to infringing Articles 30 and 31 of that directive on the notification of decisions and the access to judicial and administrative redress procedures. Neither the general principle of the proportionality of EU law nor the latter two provisions would apply to the persons concerned, even though the procedural safeguards laid down in those provisions (the objective of which, as I have just noted, is to ensure that the rights conferred by that directive remain effective) were conceived to protect the persons concerned specifically *when those persons forfeit rights* on grounds of public order, public security, public health *or on other grounds* , and are intended inter alia to govern any expulsion measures. ( 57 )\n\n68\\. Therefore, to ensure the effectiveness of Directive 2004\u002F38 and of the rights conferred by it, it must be possible for a person accused of improper or fraudulent conduct to contest that accusation and to be able to defend himself or herself, by availing himself or herself of effective procedural remedies, which also follows from the right to effective judicial protection guaranteed by the first paragraph of Article 47 of the Charter of Fundamental Rights of the European Union. ( 58 )\n\n69\\. With regard, in the second place, to *the purpose* of Article 35 of Directive 2004\u002F38, I must observe, as the Commission noted in its written observations, that it may be that conduct constituting an ‘abuse of rights’ is not apparent when the abuse is committed but comes to light subsequently. The improper nature of a marriage of convenience is detected only after evidence regarding the formally married couple has been collected. ( 59 ) Therefore, an interpretation to the effect that the application of Article 35 is limited to the period during which the person is still a beneficiary of a derived right conferred by that directive would amount to excluding from the scope of that provision a situation such as that at issue in the main proceedings in which the marriage of convenience has gone undetected for a sufficiently long period of time, thus rendering the purpose of the provision redundant.\n\n70\\. In addition, it should be observed that that provision expresses the general principle of EU law that abuse of rights is prohibited. Accordingly, the concept of a ‘marriage of convenience’ must be interpreted in the light of that principle. I will address this point in the next section.\n\n71\\. It follows that the teleological interpretation of Directive 2004\u002F38 argues in favour of the applicability of Article 35.\n\n**C.** **Combating the abuse of rights in the context of Article 35 of Directive 2004\u002F38**\n\n*1.*  *The objections put forward by the German Government*\n\n72\\. As I have already mentioned, the German Government is of the view that Article 35 of Directive 2004\u002F38 is not applicable in the present case. More specifically, it argued in its written observations that the subsequent review of the improper or fraudulent acquisition of a right of residence conferred by that directive falls within the scope of national law, in accordance with the procedural autonomy of the Member States.\n\n73\\. I disagree with that view.\n\n74\\. In the first place, whilst it is indeed true that combating the abuse of rights is a matter for both EU law and national laws, the fact remains that the improper or fraudulent use of the EU legal system must be examined within the framework of that system, in order, inter alia, as the Commission rightly pointed out, to ensure the ‘uniform application’ of the provisions of EU law. ( 60 ) More specifically, I would point out that, in accordance with settled case-law of the Court, Directive 2004\u002F38 does not deny the Member States any power of review over the entry into and residence in their territory of the family members of EU citizens. However, since a family member of a Union citizen who is not a national of a Member State derives rights of entry and of residence from Directive 2004\u002F38 in the host Member State, the latter may restrict those rights only in compliance with Articles 27 and 35 thereof. ( 61 )\n\n75\\. It follows that taking the view, as the German Government does, that the subsequent review of the improper or fraudulent acquisition of a right of movement and of residence under Directive 2004\u002F38 falls within the scope of national law would mean, first, ignoring *when the right conferred* by that directive *was originally acquired* and, therefore, disregarding a key element, namely *the EU legal framework which determined the acquisition of that right*. Such disregard would be contrary, inter alia, to the general principle of legal certainty inherent in the EU legal order. It would be contrary to that principle to accept the approach under which the acquisition of the rights conferred by that directive is subject to EU law, whereas the determination of the abusive nature of their acquisition is governed by national law. Second, such an approach would amount to accepting that each Member State can apply its own criteria to determine whether or not there is an abuse of the rights conferred by the directive, thus undermining the effectiveness of EU law. Directive 2004\u002F38 must therefore apply when examining any improper use of that directive.\n\n76\\. In the second place, as I have just noted above, Article 35 of Directive 2004\u002F38 gives concrete expression to the general principle of EU law that abuse of rights is prohibited. That principle seeks to guarantee the effectiveness of the fight against abusive practices. ( 62 ) In that context, although Article 35 of that directive does not define the concept of an ‘abuse of rights’, it should be recalled that it is an autonomous concept of EU law on which the Court has already had occasion to give rulings. Thus, the Court has clarified that proof of an abuse requires, first, *a combination of objective circumstances* in which, despite formal observance of the conditions laid down by EU rules, the purpose of those rules has not been achieved and, second, *a subjective element* consisting in the intention to obtain an advantage from the EU rules by artificially creating the conditions laid down for obtaining it. ( 63 ) The objective and subjective cumulative conditions for abuse, formulated by the Court in its case-law, allow, in particular, a distinction to be drawn, significantly, between *the lawful use* and *the improper use* of the rights conferred by the directive. ( 64 ) Within the context of that concept, and in the light of recital 28, marriages of convenience can be defined for the purpose of that directive as a marriage contracted, first, without a genuine relationship between the parties, that relationship being purely artificial, and, second, with the sole aim of one spouse benefiting from the right to move and reside freely conferred on Union citizens and their family members by EU law which he or she would not otherwise enjoy. ( 65 ) However, it is important to note that the fact that two people may derive a certain advantage from their genuine marriage, such as, in particular, a right of residence, does not necessarily mean that that marriage is a marriage of convenience and that there has therefore been an abuse of rights. ( 66 )\n\n77\\. That being said, the question remains as to the scope of the powers of review conferred on Member States by Article 35 of Directive 2004\u002F38.\n\n*2.*  *The scope of the powers of review conferred on Member States by Article 35*\n\n78\\. At this stage of my analysis, the answer to the question put by the referring court is therefore partially clear. Article 35 of Directive 2004\u002F38 must be interpreted as allowing Member States to revoke, withdraw or terminate, including retroactively, a right of residence conferred by that directive even though that right has already ceased to produce effects.\n\n79\\. Furthermore, I note that the rights conferred by Directive 2004\u002F38, which were acquired improperly but have ceased to be exercised by the person holding them for some time (here: for a period of at least five years), not only could have produced legal effects in the past but may still continue to have such effects (currently and in the future). If a right conferred by that directive is deemed to have been previously acquired improperly, the national authorities are permitted, pursuant to Article 35 thereof, to adopt the necessary measures to prevent an earlier abuse of rights from serving as the basis for obtaining additional current or future rights. Thus, even if that right of residence is no longer being exercised, which is the case here, *a formal withdrawal* of that right could potentially allow the national authorities to rectify the derived legal effects of that right. ( 67 )\n\n80\\. In that connection, with regard to the discretion enjoyed by Member States in exercising that power of review, it should be observed that it is apparent from the wording of Article 35 of Directive 2004\u002F38 that, by using the verb ‘may’, that provision simply authorises Member States to adopt some or all of the measures mentioned therein, namely to refuse, terminate *or* withdraw a right conferred by that directive. However, nothing in the wording of that provision can be interpreted as indicating that Member States are obliged to adopt one, more or all of the measures set out. It is apparent from the words ‘necessary measures’ contained in the same provision that Member States have some leeway in determining which measures are essential, having regard to the specific abusive practices which constituted the abuse complained of. Thus, the national authorities may find, on an individual and case-by-case basis, that there is an abuse of rights, provided that the measures which they will adopt comply with the general principle of proportionality (suitability, necessity and proportionality *stricto sensu*) and are subject to the procedural safeguards provided for in the directive.\n\n81\\. That said, by its question referred for a preliminary ruling, the referring court still seeks to ascertain whether Article 35 of Directive 2004\u002F38 allows the competent national authorities to investigate a marriage of convenience and, if appropriate, make a determination or reach a conclusion as to the existence of such a marriage.\n\n82\\. In the first place, it should be observed that the implementation of the ‘necessary measures’ set out in Article 35 of Directive 2004\u002F38 necessarily means that Member States have a *power to investigate* an abuse of rights which may legitimately be suspected to have occurred when the right conferred by that directive was acquired. ( 68 ) More specifically, as the Commission argues, that power of investigation allows the national authorities, first, to assess and determine the facts and, second, potentially to classify as abusive for the purposes of that provision, in the light of those facts, the legal situations arising when that right was acquired. It is true that although the existence of an abuse of rights does not constitute *stricto sensu* a measure refusing, terminating or withdrawing a right conferred by the directive, that finding is, however, a necessary stage prior to the adoption of any measure set out in that provision, including, logically, even where that right has already ceased to produce effects. Accordingly, as the Minister for Justice, Ireland and the Commission contend, the power to investigate is an autonomous power of the Member States which may lead to the finding of an abuse of rights or fraud. It therefore falls within the scope of the Member States’ procedural autonomy and must comply with the principles of equivalence and effectiveness, as well as with Articles 7 and 9 of the Charter of Fundamental Rights of the European Union.\n\n83\\. In the second place, I must add, as the parties discussed at the hearing, that a *time limit* on the possibility, for the Member States, to adopt measures pursuant to Article 35 of Directive 2004\u002F38 also falls within the scope of the Member States’ procedural autonomy. ( 69 )\n\n84\\. In the third place, it should be recalled that, since EU law does not lay down *specific sanctions to combat abuse and fraud in connection with rights conferred by that directive* , the possibility of providing for that type of sanction falls within the scope of the domestic legal systems of the Member States, provided that those sanctions are effective, non-discriminatory and proportionate. ( 70 )\n\n85\\. I note, in the fourth and final place, that, at the hearing, in answer to a question put by the Court, Ireland stated that, in order to obtain Irish citizenship by naturalisation, a person must have resided lawfully in Ireland for a certain period of time. However, Ireland made the point at the hearing that the main proceedings do not concern a forfeiture of Irish nationality by the appellant in the main proceedings.\n\n86\\. In any event, and for the sake of completeness, it should be recalled that, as is apparent from settled case-law, if the forfeiture of Irish nationality were to entail the loss of citizenship of the Union and the rights attaching thereto, that forfeiture should have due regard to the principle of proportionality so far as concerns the consequences of the forfeiture for the situation of the person concerned and, if relevant, for that of the members of his or her family, from the point of view of EU law. ( 71 )\n\n**V.** **Conclusion** 1 ( i ) The name of the present case is a fictitious name. It does not correspond to the real name of any party to the proceedings.\n\n87\\. In the light of the foregoing considerations, I propose that the Court answer the question referred for a preliminary ruling by the Court of Appeal (Ireland) as follows: Article 35 of Directive 2004\u002F38\u002FEC of the European Parliament and of the Council of 29 April 2004 on the right of citizens of the Union and their family members to move and reside freely within the territory of the Member States amending Regulation (EEC) No 1612\u002F68 and repealing Directives 64\u002F221\u002FEEC, 68\u002F360\u002FEEC, 72\u002F194\u002FEEC, 73\u002F148\u002FEEC, 75\u002F34\u002FEEC, 75\u002F35\u002FEEC, 90\u002F364\u002FEEC, 90\u002F365\u002FEEC and 93\u002F96\u002FEEC, as amended by Regulation (EU) No 492\u002F2011 of the European Parliament and of the Council of 5 April 2011, must be interpreted as allowing the competent national authorities to investigate and, if appropriate, make a determination or reach a conclusion that a person who previously benefited from a derived right to move freely under that directive committed a fraud or an abuse of rights, where the residence of that person in the Member State concerned is no longer based on the directive, including where he or she has in the meantime acquired the nationality of that Member State.\n\n* * *\n\n( 2 ) Directive of the European Parliament and of the Council of 29 April 2004 on the right of citizens of the Union and their family members to move and reside freely within the territory of the Member States amending Regulation (EEC) No 1612\u002F68 and repealing Directives 64\u002F221\u002FEEC, 68\u002F360\u002FEEC, 72\u002F194\u002FEEC, 73\u002F148\u002FEEC, 75\u002F34\u002FEEC, 75\u002F35\u002FEEC, 90\u002F364\u002FEEC, 90\u002F365\u002FEEC and 93\u002F96\u002FEEC ([OJ 2004 L 158, p. 77](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Foj\u002FJOL_2004_158_R_TOC)), as amended by Regulation (EU) No 492\u002F2011 of the European Parliament and of the Council of 5 April 2011 ([OJ 2011 L 141, p. 1](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Foj\u002FJOL_2011_141_R_TOC), and corrigenda [OJ 2004 L 229, p. 35](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Foj\u002FJOL_2004_229_R_TOC), and [OJ 2005 L 197, p. 34](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Foj\u002FJOL_2005_197_R_TOC)) (‘Directive 2004\u002F38’).\n\n( 3 ) See, in that regard, judgments of 7 July 1992, Singh ([C‑370\u002F90](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A1992%3A296), [EU:C:1992:296](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A1992%3A296), paragraph [24](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A1992%3A296#point24)); of 23 September 2003, Akrich ([C‑109\u002F01](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2003%3A491), [EU:C:2003:491](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2003%3A491), paragraphs [55](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2003%3A491#point55) to [58](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2003%3A491#point58)); and of 19 October 2004, Zhu and Chen ([C‑200\u002F02](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2004%3A639), [EU:C:2004:639](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2004%3A639), paragraphs [34](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2004%3A639#point34) to [37](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2004%3A639#point37)). In that last judgment, the Court simply rejects a Member State’s line of argument based on an alleged abuse of rights. With regard to the case-law in which the Court has mentioned Article 35 of Directive 2004\u002F38 without reference having been made to that provision in the operative part of the judgment, see judgment of 25 July 2008, *Metock and Others* (C‑127\u002F08, ‘the judgment in Metock, [EU:C:2008:449](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2008%3A449), paragraph [75](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2008%3A449#point75)). See also judgment of 26 March 2019, SM (Child placed under Algerian kafala) ([C‑129\u002F18](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A248), [EU:C:2019:248](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A248)), and the Opinion of Advocate General Campos Sánchez-Bordona in that case (points 112 to 117).\n\n( 4 ) Judgment of 18 December 2014, *McCarthy and Others* (C‑202\u002F13, ‘the judgment in McCarthy and Others, [EU:C:2014:2450](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2014%3A2450)).\n\n( 5 ) See judgment in *McCarthy and Others* (paragraph 58 and the operative part).\n\n( 6 ) Judgment of 14 November 2017, *Lounes* (C‑165\u002F16, ‘the judgment in *Lounes* ’, EU:C:2017:862).\n\n( 7 ) Judgment of 10 September 2019, *Chenchooliah* (C‑94\u002F18, ‘the judgment in Chenchooliah, [EU:C:2019:693](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A693)).\n\n( 8 ) Although the question referred for a preliminary ruling concerns, in actual fact, only the loss of the status of beneficiary of a right conferred by Directive 2004\u002F38 by virtue of the acquisition of the nationality of the host Member State, I am of the opinion that, with a view to determining the applicability of Article 35 of that directive, the naturalisation of the person concerned is just one of the scenarios that entails the loss of that status. A beneficiary under the directive can lose his or her status for other reasons, such as in particular departure from that Member State. See point 63 of the present Opinion. Thus, it appears justified, to me, to adopt a general approach which is not limited strictly to the loss of the status of beneficiary by naturalisation in order to interpret that provision, in particular for reasons of the sound administration of justice. See, by analogy, my Opinion in JYSK ([C‑117\u002F24](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A372), [EU:C:2025:372](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A372), point [60](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A372#point60)).\n\n( 9 ) See, inter alia, the judgment in *Metock* (paragraph 82); judgments of 12 March 2014, O. and B. ([C‑456\u002F12](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2014%3A135), [EU:C:2014:135](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2014%3A135), paragraph [35](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2014%3A135#point35)); and of 1 August 2025, Jobcenter Arbeitplus Bielefeld ([C‑397\u002F23](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A602), [EU:C:2025:602](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A602), paragraph [44](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A602#point44) and the case-law cited).\n\n( 10 ) See, inter alia, the judgments in *Lounes* (point 31) and in *McCarthy and Others* (paragraphs 31 and 33 and the case-law cited).\n\n( 11 ) See, inter alia, the judgments in *Lounes* (paragraph 32) and in *McCarthy and Others* (paragraph 34 and the case-law cited).\n\n( 12 ) See, inter alia, the judgment in *Lounes* (paragraph 34 and the case-law cited). In that regard, it should be recalled that the Court has held that ‘if Union citizens were not allowed to lead a normal life in the host Member State, the exercise of the freedoms they are guaranteed by the Treaty would be seriously obstructed’ (see the judgment in *Metock* (paragraph 62)).\n\n( 13 ) See, by analogy, the judgment in *Lounes* (paragraph 43). 14 ( 15 ) At the hearing, R.S. stated that he acquired Irish nationality on 23 April 2016. 16 ( 17 ) See, by analogy, the judgment in *Lounes* (paragraphs 41 and 44). In that judgment, the Court held, first, that the Spanish national, who had been naturalised by the United Kingdom, ceased to fall within the definition of a ‘beneficiary’ within the meaning of Article 3(1) of Directive 2004\u002F38 and, second, that that directive no longer governed her residence in that Member State, as that residence was inherently unconditional. Accordingly, the Court declared that her spouse, a third-country national, did not benefit from a derived right of residence in the Member State in question on the basis of the provisions of that directive. See also Opinion of Advocate General Bot in Lounes ([C‑165\u002F16](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2017%3A407), [EU:C:2017:407](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2017%3A407)).\n\n( 18 ) The Court stated in paragraph 34 of the judgment in *Lounes* that ‘it is clear from the wording of Article 3(1) of Directive 2004\u002F38 that Union citizens who move to or reside in a “Member State other than that of which they are a national”, and their family members … who accompany or join them, fall within the scope of the directive and are beneficiaries of the rights conferred by it’. See point 29 of the present Opinion. 19 ( 20 ) See judgments of 17 November 1983, Merck ([292\u002F82](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A1983%3A335), [EU:C:1983:335](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A1983%3A335), paragraph [12](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A1983%3A335#point12)), and of 1 August 2022, Familienkasse Niedersachsen-Bremen ([C‑411\u002F20](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2022%3A602), [EU:C:2022:602](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2022%3A602), paragraph [51](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2022%3A602#point51)).\n\n( 21 ) See judgments of 3 October 2013, Inuit Tapiriit Kanatami and Others v Parliament and Council ([C‑583\u002F11 P](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2013%3A625), [EU:C:2013:625](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2013%3A625), paragraph [50](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2013%3A625#point50)); of 11 April 2019, Tarola ([C‑483\u002F17](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A309), [EU:C:2019:309](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A309), paragraph [37](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A309#point37)); and of 11 January 2024, Inditex ([C‑361\u002F22](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2024%3A17), [EU:C:2024:17](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2024%3A17), paragraph [43](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2024%3A17#point43) and the case-law cited).\n\n( 22 ) See, inter alia, the French-language version (‘ *refuser, annuler ou retirer* ’), the German-language version (‘ *zu verweigern, aufzuheben oder zu widerrufen* ’), the Spanish-language version (‘ *denegar, extinguir o retirar* ’), the Italian-language version (‘ *rifiutare, estinguere o revocare* ’), the Lithuanian-language version (‘ *atsisakyti, nutraukti ar panaikinti* ’) and the Romanian-language version (‘ *refuza, anula sau retrage* ’).\n\n( 23 ) See, inter alia, the French-language version (‘ *tout droit conféré* ’), the German-language version (‘ *die durch diese Richtlinie verliehenen Rechte* ’), the Spanish-language version (‘ *cualquier derecho conferido* ’), the Italian-language version (‘ *un diritto conferito* ’), the Lithuanian-language version (‘ *bet kokią šia direktyva suteiktą teisę* ’) and the Romanian-language version (‘ *orice drept conferit* ’). 24 ( 25 ) In English, the verb ‘terminate’ generally means to put an end to an ongoing situation, whereas the verb ‘withdraw’ can simply mean to remove, but can also mean to retract or to revoke. In Italian, the use of the verb ‘ *estinguere* ’ also calls to mind an ongoing situation, whereas ‘ *revocare* ’ suggests the withdrawal of a past declaration or an established situation without however requiring that someone be in possession of what is revoked. The same goes for the German-language version, with the verb ‘ *widerrufen* ’ meaning ‘to retract what has been stated or declared’. 26 ( 27 ) It cannot be ruled out that certain national laws may provide for retroactive termination or limit the scope of termination to effects that have yet to be executed. Here, it is apparent from the order for reference that the decision at issue in the main proceedings concerns the ‘determination’, the ‘finding’ or the ‘conclusion’ drawn from the situation or from the past conduct of the appellant in the main proceedings, in particular the fact that the right was ‘deemed withdrawn from the outset’. Thus, the Minister for Justice appears to be reviewing a past declaration and is not withdrawing a current right from the appellant, but rather the right which had been granted to him in the past and from which the appellant in the main proceedings currently no longer benefits. At the hearing, Ireland explained that such a finding amounts to taking the view that the right of residence, which has in practice been exercised, is deemed to have never existed legally in the light of abusive practices, which is a matter for the referring court to determine. 28 ( 29 ) See points 34 and 49 of the present Opinion. In other words, the system introduced by Directive 2004\u002F38 covers the ‘entire lifecycle’ of the exercise of the freedom of movement by a Union citizen and his or her family members. See Opinion of Advocate General Szpunar in Chenchooliah ([C‑94\u002F18](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A433), [EU:C:2019:433](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A433), points [70](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A433#point70) to [75](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A433#point75)).\n\n( 30 ) See point 47 of the present Opinion. By contrast, since Article 35 of Directive 2004\u002F38 *does not concern* a right of movement and of residence from which the person concerned benefited but which should be restricted on grounds of public policy, public security or public health, it *does not fall within the scope* of Article 1 *(c)* of that directive. Such a restriction on the grounds mentioned is covered by the provisions of Chapter VI, entitled ‘Restrictions on the right of entry and the right of residence on grounds of public policy, public security or public health’. 31 ( 32 ) The provisions in question are, inter alia, provisions on publicity (Article 34), the repeal of earlier directives (Article 38), reports on the application of the directive (Article 39), arrangements for transposition (Article 40), the directive’s entry into force (Article 41) and its addressees (Article 42).\n\n( 33 ) Article 36 of Directive 2004\u002F38, entitled ‘Sanctions’, provides that ‘Member States shall lay down provisions on the sanctions applicable to breaches of national rules *adopted for the implementation of [that] Directive* and *shall take* the measures required for their application. The sanctions laid down *shall be* effective and proportionate. …’ (emphasis added). As a reminder, Article 36 of Directive 2004\u002F38 codifies the related settled case-law of the Court. See, inter alia, judgments of 12 December 1989, Messner ([C‑265\u002F88](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A1989%3A632), [EU:C:1989:632](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A1989%3A632), paragraph [15](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A1989%3A632#point15)), and of 30 April 1998, Commission v Germany ([C‑24\u002F97](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A1998%3A184), [EU:C:1998:184](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A1998%3A184), paragraph [14](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A1998%3A184#point14) and the case-law cited). 34 ( 35 ) The measures laid down in Article 36 of Directive 2004\u002F38 are *general in nature* , whereas, as the Court has observed, measures that *may* be adopted by the national authorities, on the basis of Article 35 of that directive, in order to refuse, terminate or withdraw a right conferred by the directive ‘must be based on an individual examination of the particular case’ and, therefore, are *specific in nature*. See the judgment in *McCarthy and Others* (paragraphs 49 and 52 and the case-law cited). 36 ( 37 ) Proposal for a European Parliament and Council directive on the right of citizens of the Union and their family members to move and reside freely within the territory of the Member States, of 29 June 2001, COM(2001) 257 final ([OJ 2001 C 270 E, p. 150](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Foj\u002FJOC_2001_270_E_TOC)).\n\n( 38 ) Council Common Position (EC) No 6\u002F2004 of 5 December 2003 ([OJ 2004 C 54 E, p. 12](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Foj\u002FJOC_2004_054_E_TOC), in particular p. 32). Emphasis added. 39 ( 40 ) Commission Notice, Guidance on the right of free movement of EU citizens and their families, of 6 December 2023, C(2023) 8500 final (OJ C, C\u002F2023\u002F1392) (‘the 2023 Commission Guidelines’). 41 ( 42 ) That judgment was given in the context of proceedings between a Mauritian national residing in Ireland and the Minister for Justice and Equality (Ireland) concerning a decision to deport her, in accordance with a provision of Irish legislation, following the return of her spouse, a Union citizen, to the Member State of which he was a national, namely Portugal, where he was serving a prison sentence. In accordance with national law, the deportation order automatically imposed an indefinite ban on entry into Irish territory.\n\n( 43 ) See the judgments in *Chenchooliah* (paragraph 62) and in *Lounes* (paragraphs 38 to 42). With regard to the ‘dynamic’ or ‘evolving’ nature of the concept of ‘beneficiary’ within the meaning of Article 3(1) of Directive 2004\u002F38, see the judgment in *Metock* (paragraphs 73, 80 and 99). See also Opinion of Advocate General Szpunar in Chenchooliah ([C‑94\u002F18](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A433), [EU:C:2019:433](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A433), points [47](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A433#point47) to [67](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A433#point67) and [70](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A433#point70)).\n\n( 44 ) See the judgment in *Chenchooliah* (paragraph 70). See, also, point 48 of and footnote 28 to the present Opinion.\n\n( 45 ) See the judgment in *Chenchooliah* (paragraph 70).\n\n( 46 ) See, to that effect, the judgment in *Chenchooliah* (paragraphs 59 to 63).\n\n( 47 ) See, to that effect, the judgment in *Chenchooliah* (paragraph 79). As a reminder, Article 15 of Directive 2004\u002F38, entitled ‘Procedural safeguards’, provides, in paragraph 1 thereof, that ‘the procedures provided for by Articles 30 and 31 shall apply by analogy to all decisions restricting free movement of Union citizens and their family members on grounds other than public policy, public security or public health.’ Paragraph 3 of that article provides that ‘the host Member State may not impose a ban on entry in the context of an expulsion decision to which paragraph 1 applies.’ According to the Court, Article 15 of that directive lays down ‘the rules applicable when a temporary right of residence under the directive comes to an end, in particular where a Union citizen or one of his family members who, in the past, had a right of residence of up to 3 months, or longer than 3 months, by virtue of Article 6 or Article 7 of that directive respectively, no longer satisfies the requirements for the grant of the right of residence concerned and may therefore, in principle, be expelled by the host Member State’ (see paragraph 74 of that judgment). 48 ( 49 ) See the judgment in *Chenchooliah* (paragraph 65).\n\n( 50 ) See the judgment in *Chenchooliah* (paragraph 66).\n\n( 51 ) See the judgment in *Chenchooliah* (paragraph 68).\n\n( 52 ) See point 57 of the present Opinion. As certain legal writers have noted, ‘the protection attached to Union citizenship is thus prolonged: third-country nationals who enjoyed a derived right of residence in the past are more protected than third-country nationals who have never been bound to a Union citizen.’ See Ritleng, D., ‘Scope and meaning of Article 15 of Directive 2004\u002F38: Yes but no: Chenchooliah’, *Common Market Law Review* , Vol. 57, No 4, 2020, pp. 1183 to 1200, in particular p. 1195. 53 ( 54 ) In the present case, the effect of that retroactive loss of the rights conferred by Directive 2004\u002F38, as the referring court states, is that those rights are deemed to have been ‘withdrawn from the outset’.\n\n( 55 ) See, inter alia, the judgment in *Metock* (paragraph 82); judgments of 12 March 2014, O. and B. ([C‑456\u002F12](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2014%3A135), [EU:C:2014:135](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2014%3A135), paragraph [35](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2014%3A135#point35)); and of 1 August 2025, Jobcenter Arbeitplus Bielefeld ([C‑397\u002F23](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A602), [EU:C:2025:602](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A602), paragraph [44](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A602#point44) and the case-law cited). 56 ( 57 ) See footnote 47 to the present Opinion. 58 ( 59 ) See, inter alia, Communication from the Commission to the European Parliament and to the Council – Helping national authorities fight abuses of the right to free movement: Handbook on addressing the issue of alleged marriages of convenience between EU citizens and non-EU nationals in the context of EU law on free movement of EU citizens (SWD(2014) 284 final), COM(2014) 604 final, Section 4.2 (‘the Marriages of Convenience Handbook’). See, also, COM(2014)284 final, Section 4.5, p.42.\n\n( 60 ) See, by analogy, judgment of 23 March 2000, Diamantis ([C‑373\u002F97](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2000%3A150), [EU:C:2000:150](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2000%3A150), paragraph [34](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2000%3A150#point34) and the case-law cited). 61 ( 62 ) The general principle that abuse of rights is prohibited was laid down in the judgment of 14 December 2000, Emsland-Stärke ([C‑110\u002F99](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2000%3A695), [EU:C:2000:695](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2000%3A695), paragraphs [52](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2000%3A695#point52) and [53](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2000%3A695#point53)). The Court recognised it, expressly, as a general principle of EU law in the judgment of 5 July 2007, Kofoed ([C‑321\u002F05](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2007%3A408), [EU:C:2007:408](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2007%3A408), paragraph [38](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2007%3A408#point38)). See, more recently, judgment of 10 November 2011, Foggia – Sociedade Gestora de Participações Sociais ([C‑126\u002F10](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2011%3A718), [EU:C:2011:718](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2011%3A718), paragraph [50](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2011%3A718#point50)). See, in that regard, Tridimas, P.T., ‘The general principles of law: Who needs them?’, *Les Cahiers de Droit* , Vol. 52, No 1, 2015, pp. 419 to 441, in particulier p. 427.\n\n( 63 ) See judgment of 14 December 2000, Emsland-Stärke ([C‑110\u002F99](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2000%3A695), [EU:C:2000:695](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2000%3A695), paragraphs [52](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2000%3A695#point52) and [53](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2000%3A695#point53)), and the judgment in *McCarthy and Others* (paragraph 54 and the case-law cited). See also the Opinion of Advocate General Szpunar in McCarthy and Others ([C‑202\u002F13](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2014%3A345), [EU:C:2014:345](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2014%3A345), points [108](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2014%3A345#point108) to [115](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2014%3A345#point115)).\n\n( 64 ) In particular, in the context of the free movement of persons, the Court has already recalled that the fact that a Member State national has chosen to acquire a professional qualification in a Member State other than that in which he resides in order to benefit there from more favourable legislation is not, in itself, sufficient grounds to conclude that there is an abuse of rights. See, by analogy, judgment of 17 July 2014, Torresi ([C‑58\u002F13 and C‑59\u002F13](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2014%3A2088), [EU:C:2014:2088](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2014%3A2088), paragraph [50](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2014%3A2088#point50)), and Opinion of Advocate General Wahl in those joined cases ([C‑58\u002F13 and C‑59\u002F13](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2014%3A265), [EU:C:2014:265](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2014%3A265), points [91](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2014%3A265#point91) and [92](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2014%3A265#point92)). 65 ( 66 ) See, by analogy, judgments of 9 March 1999, Centros ([C‑212\u002F97](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A1999%3A126), [EU:C:1999:126](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A1999%3A126), paragraph [27](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A1999%3A126#point27)), and of 23 September 2003, Akrich ([C‑109\u002F01](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2003%3A491), [EU:C:2003:491](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2003%3A491), paragraph [55](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2003%3A491#point55)). I note, in that regard, as the Commission has stated, that ‘a marriage cannot be considered as a marriage of convenience simply because it brings an immigration advantage, or indeed any other advantage. The quality of the relationship is immaterial to the application of Article 35’. See the 2009 Commission Guidelines, p. 15. For clues to potential abuse that may trigger an investigation, see the Marriages of Convenience Handbook, p. 7, Section 4.1. See also judgments of the EFTA Court of 9 February 2021, *Kerim* (E-1\u002F20, paragraph 36), and of 23 November 2021, *Q and Others* (E-16\u002F20, paragraphs 59 to 64). 67 ( 68 ) It is apparent from the 2009 Commission Guidelines (p. 16) and from the 2023 Commission Guidelines (p. 71) that Directive 2004\u002F38 ‘does not prevent Member States from investigating individual cases where there is a well-founded suspicion of abuse. However, EU law prohibits systematic checks’. 69 ( 70 ) Such as, in particular, cancelling the effects of a marriage of convenience as a matter of civil law or administrative fines. See, in that regard, the 2009 Commission Guidelines, p. 16, and the 2023 Commission Guidelines, pp. 71 and 72.\n\n( 71 ) See judgments of 2 March 2010, Rottmann ([C‑135\u002F08](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2010%3A104), [EU:C:2010:104](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2010%3A104), paragraphs [39](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2010%3A104#point39), [41](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2010%3A104#point41), [42](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2010%3A104#point42), [45](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2010%3A104#point45), [55](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2010%3A104#point55) and [56](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2010%3A104#point56)), and of 12 March 2019, Tjebbes and Others ([C‑221\u002F17](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A189), [EU:C:2019:189](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A189), paragraphs [30](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A189#point30) to [32](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A189#point32) and [40](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A189#point40)). See also judgments of 21 October 2020, Stadt Duisburg (Maintenance of rights acquired after naturalisation) ([C‑720\u002F19](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2020%3A847), [EU:C:2020:847](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2020%3A847)); of 18 January 2022, Wiener Landesregierung (Revocation of an assurance of naturalisation) ([C‑118\u002F20](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2022%3A34), [EU:C:2022:34](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2022%3A34)); and of 5 September 2023, Udlændinge- og Integrationsministeriet (Loss of Danish nationality) ([C‑689\u002F21](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2023%3A626), [EU:C:2023:626](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2023%3A626)).","en","cjeu","https:\u002F\u002Feur-lex.europa.eu\u002Flegal-content\u002FEN\u002FTXT\u002FHTML\u002F?uri=CELEX:62024CC0560","ecli-eu-c-2026-11",{"caseNumber":23},"62024CC0560",{"id":25,"ecli":26,"caseNumber":27,"courtId":5,"courtName":15,"decisionDate":16,"publicationDate":16,"fullText":28,"language":18,"source":19,"sourceUrl":29,"summary":15,"slug":30,"metadata":31},"266","ECLI:EU:C:2026:16","C-748\u002F24","OPINION OF ADVOCATE GENERAL\n\nSPIELMANN 1 Case C‑748\u002F24 [Kotaňák ( i )]\n\nOkresná prokuratúra Bratislava III\n\nCriminal proceedings\n\nv\n\nAC\n\nintervening party:\n\nLZ\n\n(Request for a preliminary ruling from the Mestský súd Bratislava I (Bratislava I City Court, Slovakia)) \n\n(Reference for a preliminary ruling – Area of freedom, security and justice – Judicial cooperation in criminal matters – Fundamental rights – Presumption of innocence and rights of the defence – Order of a court of first instance discontinuing criminal proceedings on the grounds that the acts under examination are not criminal in nature – Annulment by the higher court – Alleged criminal nature of those acts according to the higher court – Appropriate measures to remedy the breach of the presumption of innocence)\n\n**I.** **Introduction**\n\n1\\. As a fundamental principle of criminal procedure in States governed by the rule of law, the presumption of innocence, as Robert Badinter notes, ‘is first expressed in a question of procedure, the burden of proof … However, the accused person must be guaranteed protection of the presumption of innocence at each stage of the proceedings … More importantly, in the absence of conclusive evidence, any doubt should benefit the accused’. ( 2 ) Thus, at each stage, it is necessary to allow ‘the law to hear the reasoning that leads to innocence’. ( 3 )\n\n2\\. By its question for a preliminary ruling, the referring court asks the Court to clarify the scope of the procedural guarantees covered by the presumption of innocence, in the context of preliminary decisions of a procedural nature taken by judicial authorities before the main proceedings to determine whether the accused person is guilty.\n\n3\\. The request for a preliminary ruling, made in the context of criminal proceedings against AC for slander, relates more specifically to the interpretation of Article 4 and Article 6(1) of Directive (EU) 2016\u002F343 ( 4 ) and Article 48 of the Charter of Fundamental Rights of the European Union (‘the Charter’).\n\n**II.** **Legal context**\n\n**A.** **European Union law**\n\n4\\. Article 48 of the Charter, entitled ‘Presumption of innocence and right of defence’, provides, in paragraph 1, that ‘everyone who has been charged shall be presumed innocent until proved guilty according to law’.\n\n5\\. Recitals 9, 10, 16 and 48 of Directive 2016\u002F343 are worded as follows: \n\n‘(9) The purpose of this Directive is to enhance the right to a fair trial in criminal proceedings by laying down common minimum rules concerning certain aspects of the presumption of innocence and the right to be present at the trial. (10) By establishing common minimum rules on the protection of procedural rights of suspects and accused persons, this Directive aims to strengthen the trust of Member States in each other’s criminal justice systems and thus to facilitate mutual recognition of decisions in criminal matters. Such common minimum rules may also remove obstacles to the free movement of citizens throughout the territory of the Member States. … (16) The presumption of innocence would be violated if public statements made by public authorities, or judicial decisions other than those on guilt, referred to a suspect or an accused person as being guilty, for as long as that person has not been proved guilty according to law. Such statements and judicial decisions should not reflect an opinion that that person is guilty. This should be without prejudice to acts of the prosecution which aim to prove the guilt of the suspect or accused person, such as the indictment, and without prejudice to judicial decisions as a result of which a suspended sentence takes effect, provided that the rights of the defence are respected. This should also be without prejudice to preliminary decisions of a procedural nature, which are taken by judicial or other competent authorities and are based on suspicion or on elements of incriminating evidence, such as decisions on pre-trial detention, provided that such decisions do not refer to the suspect or accused person as being guilty. Before taking a preliminary decision of a procedural nature the competent authority might first have to verify that there are sufficient elements of incriminating evidence against the suspect or accused person to justify the decision concerned, and the decision could contain reference to those elements. … (48) As this Directive establishes minimum rules, Member States should be able to extend the rights laid down in this Directive in order to provide a higher level of protection. The level of protection provided for by Member States should never fall below the standards provided for by the Charter or by the [European Convention on Human Rights, signed in Rome on 4 November 1950 (ECHR)], as interpreted by the Court of Justice and by the European Court of Human Rights [(ECtHR)].’\n\n6\\. Article 1 of that directive provides: ‘This Directive lays down common minimum rules concerning: (a) certain aspects of the presumption of innocence in criminal proceedings; (b) the right to be present at the trial in criminal proceedings.’\n\n7\\. According to Article 3 of that directive: ‘Member States shall ensure that suspects and accused persons are presumed innocent until proved guilty according to law.’\n\n8\\. Article 4 of that directive, that article being headed ‘Public references to guilt’, provides, in paragraphs 1 and 2: ‘1. Member States shall take the necessary measures to ensure that, for as long as a suspect or an accused person has not been proved guilty according to law, public statements made by public authorities, and judicial decisions, other than those on guilt, do not refer to that person as being guilty. This shall be without prejudice to acts of the prosecution which aim to prove the guilt of the suspect or accused person, and to preliminary decisions of a procedural nature, which are taken by judicial or other competent authorities and which are based on suspicion or incriminating evidence. 2\\. Member States shall ensure that appropriate measures are available in the event of a breach of the obligation laid down in paragraph 1 of this Article not to refer to suspects or accused persons as being guilty, in accordance with this Directive and, in particular, with Article 10.’\n\n9\\. Pursuant to Article 6 of Directive 2016\u002F343, entitled ‘Burden of proof’: ‘1. Member States shall ensure that the burden of proof for establishing the guilt of suspects and accused persons is on the prosecution. This shall be without prejudice to any obligation on the judge or the competent court to seek both inculpatory and exculpatory evidence, and to the right of the defence to submit evidence in accordance with the applicable national law. 2\\. Member States shall ensure that any doubt as to the question of guilt is to benefit the suspect or accused person, including where the court assesses whether the person concerned should be acquitted.’\n\n10\\. Article 10(1) of that directive provides: ‘Member States shall ensure that suspects and accused persons have an effective remedy if their rights under this Directive are breached.’\n\n**B.** **Slovak law**\n\n11\\. Paragraph 373(1) and (2) of zákon č. 300\u002F2005 Z. z. Trestný zakon (Law No 300\u002F2005 establishing the Criminal Code) of 20 May 2005 (‘the Criminal Code’), which defines the criminal offence of slander, provides: ‘(1) Whoever communicates false information about another person which is capable of considerably damaging the respect of fellow citizens for such a person, their career and business, their family relations, or which causes them other grievous harm, shall be punished by a prison sentence of up to two years. (2) A prison sentence of one to five years shall be imposed upon an offender if they commit an act referred to in paragraph 1 … (c) publicly; …’\n\n12\\. Paragraph 327(1) of zákon č. 301\u002F2005 Z. z., Trestný poriadok (Law No 301\u002F2005 establishing the Code of Criminal Procedure) of 24 May 2005 (‘the Code of Criminal Procedure’) states that ‘the court to which the matter was returned for a new hearing and decision shall be bound by the legal opinion which the court of appeal pronounced in its ruling, and shall be obliged to proceed with the acts and the taking of evidence ordered by the appeal court’.\n\n**C.** **The dispute in the main proceedings, the questions referred for a preliminary ruling and the procedure before the Court**\n\n13\\. On 18 November 2020, the prosecutor filed an indictment with the Okresný súd Bratislava III (Bratislava III District Court, Slovakia) against AC for slander, an offence under Paragraph 373(1) and (2)(c) of the Criminal Code, which he is accused of having committed by posting several videos online containing a series of false allegations.\n\n14\\. By order of 8 November 2021, the Okresný súd Bratislava III (Bratislava III District Court) discontinued the proceedings on the ground that the conduct referred to in the indictment did not constitute a criminal offence.\n\n15\\. The prosecutor brought an appeal against that order. By order of 8 February 2022, the Krajský súd v Bratislave (Regional Court, Bratislava, Slovakia) set aside the order of the court of first instance and referred the case back to that court so that it could hear and decide the case again by re-examining all the evidence.\n\n16\\. By order of 3 October 2022, the Okresný súd Bratislava III (Bratislava III District Court) again dismissed the indictment against AC.\n\n17\\. By order of 18 April 2023, the Krajský súd v Bratislave (Regional Court, Bratislava) again set aside that dismissal decision and referred the case back to the court of first instance.\n\n18\\. Following a reorganisation of the Slovak judicial system, jurisdiction to hear the case in the main proceedings was transferred, on 1 June 2023, to the Mestský súd Bratislava I (Bratislava I City Court, Slovakia), which is the referring court.\n\n19\\. Before that court, AC argued that the rulings of the appeal court infringe the presumption of innocence, in so far as they leave no room for any finding other than a guilty verdict, even though it was not necessary for the appeal court to assess whether the charges against him were well founded. AC therefore considers that the referring court should disregard the rulings of the appeal court.\n\n20\\. The referring court observes that, under the Slovak rules of procedure, it is bound by the rulings of the appeal court. However, it considers that those rulings infringe the presumption of innocence of AC and that national regulations do not provide for appropriate measures to ensure respect for that presumption.\n\n21\\. In those circumstances, the Mestský súd Bratislava I (Bratislava I City Court) decided to stay the proceedings and to refer the following questions to the Court of Justice for a preliminary ruling: \n\n‘(1) Does the right to respect for the presumption of innocence pursuant to Article 48(1) of the [Charter] in conjunction with Article 4(1) and Article 6(1) of [Directive 2016\u002F343], and also the principle of proportionality and the rights of the defence pursuant to Article 48(2) of the [Charter], preclude a court of higher instance, in appeal proceedings brought by a prosecutor against an order handed down by a court of first instance discontinuing criminal proceedings, from setting out in the grounds of its ruling, before the case has been decided on the merits and without the court taking evidence in the case, the following findings of fact and law: “The false information of a purely intimate nature, which the accused shared with a large number of followers via his post, is – due to its content – liable to cause serious damage to the aggrieved person’s romantic, family and social relations, and to undermine trust in her … The evidence taken during the pre-trial investigation and on which the [court of first instance] also based its findings shows beyond doubt that the accused has been using lies about her and attacking her, which has placed her in a very unpleasant situation in her relations with her family and acquaintances. The intimate relationship between her and SB is a figment of the accused’s imagination, and the events to which the accused referred in his statements never took place … As regards the determination of which of the published statements were false, they are undoubtedly statements whose veracity was ascertained in the pre-trial investigation based on the evidence taken, and which concerned the intimate relationship between the aggrieved person and the witness SB, and the sexual practices which the accused clearly described in his post”? (2) Is the answer to the first question influenced by the fact that neither national legislation nor national practice require the court of higher instance to carry out an assessment of the factual and legal issues in the grounds for the order setting aside the ruling, and that those grounds could be limited to a simple statement that the ruling handed down by the lower court was flawed because the court of first instance is required to take evidence at the main hearing and to rule accordingly? (3) Is the answer to the first question influenced by the fact that the court of higher instance, when deciding on the prosecutor’s appeal, ruled solely on the basis of the evidence taken during the pre-trial investigation, which had not yet been heard before a court of first instance? (4) In the event that the first question is answered in the affirmative: should the following actions and rulings be regarded as appropriate measures for ensuring respect for the presumption of innocence within the meaning of Article 4(2) of [Directive 2016\u002F343]? (i) the actions of a national court of first instance, which – invoking the principle of the primacy and effectiveness of EU law – disregards the findings of fact and law made by the court of higher instance in the grounds for its ruling to the extent that they are contrary to EU law, even though those findings would otherwise have been legally binding under national legislation, and itself decides the case after the evidence has been properly taken; or the actions of a national court of first instance, which – invoking the principle of the primacy and effectiveness of EU law – disregards the findings of fact and law made by the court of higher instance in the grounds for its ruling to the extent that they are contrary to EU law, even though those findings would otherwise have been legally binding under national legislation, and itself decides the case again, handing down the same ruling to discontinue the criminal proceedings that had already been set aside once by the court of higher instance; or (ii) the recusal of the judges of the court of higher instance from the proceedings on the grounds of their lack of impartiality due to the fact that they had failed to observe the presumption of innocence, based on the plea of lack of impartiality entered by the suspect?’\n\n22\\. Written observations were lodged by the defendant AC, the Hungarian and Slovak Governments and the European Commission. AC and the Commission presented oral argument at the hearing on 16 October 2025.\n\n**III.** **Analysis**\n\n**A.** **Admissibility of the questions**\n\n23\\. The Slovak Government submits, primarily, that the request for a preliminary ruling is inadmissible in its entirety. It claims, first, that the request does not mention the factual and legal circumstances necessary to provide a useful answer or the reasons why it needs an answer from the Court, and, second, that the interpretation sought is not relevant to rule on the dispute in the main proceedings.\n\n24\\. According to settled case-law, questions on the interpretation of EU law referred by a national court in the factual and legislative context which that court is responsible for defining, and the accuracy of which is not a matter for the Court to determine, enjoy a presumption of relevance. The Court may refuse to rule on a question referred by a national court only where it is quite obvious that the interpretation of EU law that is sought is unrelated to the actual facts of the main action or its object, where the problem is hypothetical, or where the Court does not have before it the factual or legal material necessary to give a useful answer to the questions submitted to it. ( 5 )\n\n25\\. The Court has already emphasised that it may be necessary to answer questions referred in order to be able to provide referring courts with an interpretation of EU law that enables them to settle procedural questions of national law before they can rule on the substance of disputes pending before them. ( 6 )\n\n26\\. In the present case, by its first three questions, the referring court asks, in essence, whether the provisions of EU law referred to preclude a criminal court, hearing an appeal against a preliminary decision of a procedural nature, from taking a position on the incriminating evidence by determining whether the constituent elements of the criminal offence at issue are present, when it was not required to do so under national law and not all the evidence was taken at that preliminary stage of the proceedings. The referring court also questions the manner in which the appeal court took a position with regard to the presumption of innocence. In the fourth question, the referring court asks what the procedural consequences would be if the first three questions were answered in the affirmative.\n\n27\\. In my view, even though it largely referred to the arguments of the defence, the referring court has adequately specified the factual and legal circumstances necessary to give a useful answer to the questions referred. I would add that the content of the request for a preliminary ruling was sufficient to allow the Commission, the Hungarian Government and AC, as well as the Slovak Government in the alternative, to exercise the right to submit observations conferred on them by Article 23 of the Statute of the Court of Justice of the European Union.\n\n28\\. Moreover, the Slovak Government submits that the interpretation sought is not relevant for resolving the dispute in the main proceedings. It argues that any breach of the presumption of innocence by the appeal court would not prevent the referring court from dismissing the case on other grounds or on the same grounds, but on the basis of additional facts.\n\n29\\. However, since Slovak law requires the referring court to comply with the rulings of the appeal court, it seems to me that the Court’s answers to the first three questions are intended to provide it with useful guidance for assessing the conformity of the appeal court’s decision with the presumption of innocence and that the answer to the fourth question will, if necessary, allow it to infer the appropriate procedural consequences. In that respect, it does not seem evident that the interpretation of EU law sought bears no relation to the object of the main action.\n\n30\\. Nevertheless, in the third part of the fourth question, the referring court asks whether the procedural consequences of the appropriate measures within the meaning of Article 4(2) of Directive 2016\u002F343 include the possibility of recusal of the appeal court judges.\n\n31\\. It is apparent from both the observations of the Slovak Government and the answers to the questions put at the hearing that the referring court does not have jurisdiction to give a ruling on the recusal of the judges of the higher court thus invoked. ( 7 )\n\n32\\. Therefore, the third part of the fourth question is hypothetical and therefore inadmissible.\n\n33\\. Accordingly, except for the third part of the fourth question, the request for a preliminary ruling seems to me to be admissible.\n\n**B.** **Substance**\n\n**1.** **The first, second and third questions**\n\n34\\. By its first three questions, which it is appropriate to examine together, the referring court asks, in essence, whether Article 4(1) and Article 6(1) of Directive 2016\u002F343, read in conjunction with Article 48 of the Charter, must be interpreted as precluding a criminal court, hearing an appeal against a preliminary decision of a procedural nature, from taking a position on the incriminating evidence by determining whether the constituent elements of the criminal offence at issue are present, when it is not required to do so in order to give a ruling and not all the evidence was taken.\n\n**(a)** **The inapplicability of Article 6 of Directive 2016\u002F343**\n\n35\\. The referring court refers, in its first question, to Article 6(1) of Directive 2016\u002F343, according to which ‘Member States shall ensure that the burden of proof for establishing the guilt of suspects and accused persons is on the prosecution … without prejudice to any obligation on the judge or the competent court to seek both inculpatory and exculpatory evidence, and to the right of the defence to submit evidence in accordance with the applicable national law’.\n\n36\\. The Slovak Government and the Commission submit that that provision, which relates to decisions on guilt, is inapplicable in the main proceedings.\n\n37\\. I also share that view.\n\n38\\. As the Court has already held with regard to decisions on the continued detention on remand pending trial of an accused person, ( 8 ) it follows from Article 4 of Directive 2016\u002F343 that that directive distinguishes between, on the one hand, judicial decisions on guilt, which necessarily occur at the conclusion of the criminal proceedings, and, on the other hand, other procedural acts, such as acts of the prosecution and preliminary decisions of a procedural nature. The reference to establishing guilt in Article 6 of Directive 2016\u002F343 must therefore be construed as meaning that the aim of that provision is to govern the allocation of the burden of proof only in the adoption of judicial decisions on guilt.\n\n39\\. In the present case, the questions referred relate to a *preliminary stage* of the Slovak criminal proceedings, during which a single judge must determine whether a given indictment constitutes a sufficient basis for the opening of criminal proceedings or whether, conversely, the case should be discontinued or dealt with by other means.\n\n40\\. As with decisions on, for example, the continued detention on remand pending trial of an accused person, ( 9 ) such a procedural decision cannot be classified as a judicial decision on the guilt of the accused person; on the contrary, it is a ‘judicial decision, other than [that] on guilt’, within the meaning of Article 4 of Directive 2016\u002F343. The fact that, as the applicant submits, the statements of fact and in law made in the appeal decision are binding on the court of first instance, to the extent, inter alia, that it finds that the constituent elements of the offence (in the present case, slander) are present, does not alter that finding.\n\n41\\. Accordingly, since the dispute in the main proceedings does not concern the substance of the question whether the accused person is guilty, it is not necessary to answer the questions referred on the basis of Article 6(1) of Directive 2016\u002F343.\n\n**(b)** **The scope of the presumption of innocence**\n\n42\\. The first three questions referred for a preliminary ruling require us to consider the scope of the presumption of innocence with regard to the content of a preliminary decision, as well as the conditions for the adoption of that decision.\n\n*(1)*  *Content of the preliminary decision (first question)*\n\n43\\. The referring court asks whether the criminal court hearing an appeal against the preliminary decision of a procedural nature infringed the presumption of innocence by expressing itself as it did. To that end, it quotes an excerpt from the statement of reasons of the appeal court.\n\n44\\. I should point out that it is clear from Article 4(1) of Directive 2016\u002F343 that Member States are to take the necessary measures to ensure that, for as long as a suspect or an accused person has not been proved guilty according to law, judicial decisions in particular, other than those on guilt, do not refer to that person as being guilty, without prejudice to preliminary decisions of a procedural nature which are taken by judicial authorities and which are based on suspicion or on incriminating evidence.\n\n45\\. That provision must be read in the light, inter alia, of Article 48(1) of the Charter ( 10 ) and Article 3 of Directive 2016\u002F343 on the presumption of innocence, as well as recital 16 of that directive, according to which respect for the presumption of innocence is without prejudice to preliminary decisions of a procedural nature, which are taken by judicial or other competent authorities and are based on suspicion or on elements of incriminating evidence, provided that such decisions do not refer to the suspect or accused person as being guilty. According to that recital, before taking a preliminary decision of a procedural nature, the judicial authorities might first have to verify that there is sufficient incriminating evidence against the suspect or accused person to justify the decision concerned, and that the decision could contain reference to that evidence.\n\n46\\. In the absence of specific guidance in Directive 2016\u002F343 and in the case-law concerning Article 48(1) of the Charter as to how to determine whether a person is presented as being guilty in a judicial decision, we should turn primarily to the case-law of the ECtHR. ( 11 )\n\n*(i)*  *The case-law of the ECtHR*\n\n47\\. The presumption of innocence governs the criminal proceedings in their entirety ( 12 ) and the scope of Article 6(2) ECHR, ‘thus extended from the outset beyond its natural domain, will gradually extend further, particularly *ratione temporis* , that is to say at the stages of the proceedings where the person concerned is no longer or is not yet “accused”’. ( 13 ) It thus consists of a first part, namely a as procedural guarantee during the criminal trial, ( 14 ) and a second part once the criminal proceedings have ended, specifically aimed at protecting the reputation of the person concerned. ( 15 ) In view of the elements of the dispute in the main proceedings, I will focus here solely on the first part.\n\n48\\. On that basis, the presumption of innocence will be infringed if a statement by a figure of public authority, such as a judge, ( 16 ) concerning a person charged with a criminal offence reflects an opinion that he or she is guilty, before that person has been proved guilty according to law. Therefore, judges should not start with the preconceived idea that the accused has committed the offence with which that person has been charged; the burden of proof is on the prosecution, and *in dubio pro reo* – any doubt should benefit the accused. ( 17 ) The presumption of innocence may thus be infringed, even in the absence of a formal statement, if an official statement by a public authority or the reasoning of a judicial decision suggests that the judge considers the person concerned to be guilty. ( 18 )\n\n49\\. The ECtHR emphasises the importance of the *choice of words* used by public authorities in their statements before a person has been tried and found guilty of an offence. ( 19 ) However, a distinction must be made between decisions or statements that reflect an *opinion* that the person concerned is guilty and those which merely describe a state of *suspicion*. The former violate the presumption of innocence, whereas the latter have been found on several occasions to be in conformity with the spirit of Article 6 ECHR. ( 20 )\n\n50\\. In the judgment in *Marziano v. Italy* , for example, while the preliminary investigations judge did make a prognosis, he merely noted that, although there were plausible reasons to suspect the person concerned of having committed the offence at issue, there were other indications suggesting that the prosecution would have little chance of winning in court. ( 21 ) The ECtHR held that that decision described a ‘state of suspicion’ without any finding of guilt, and so was not contrary to the presumption of innocence.\n\n51\\. In the judgment in *Lavents v. Latvia* , the judge presiding over the case had given several interviews to the press, stating, inter alia, that she did not yet know ‘whether the judgment [would] lead to a conviction or partial acquittal’, expressing her astonishment at the fact that the defendants persisted in pleading not guilty to all the charges, which violated the presumption of innocence. ( 22 ) The fact that the statements in question were formulated as a question or expressed doubt was not sufficient to evade the application of Article 6(2) ECHR. According to the ECtHR, what matters for the purposes of the application of that provision ‘is the actual meaning of the statements in question, not their wording’.\n\n52\\. In *Gutsanovi v. Bulgaria* , ( 23 ) the applicant complained about the reasoning of the court’s decision ordering his continued detention on remand pending trial. The ECtHR held that the judge of the regional court had to satisfy himself that there were still plausible reasons to suspect the applicant of the commission of a criminal offence and that the words used ( 24 ) had gone beyond merely describing a state of suspicion, which entailed the violation of Article 6(2) ECHR.\n\n53\\. Although the choice of words is important, it should also be recalled that whether a statement of a public official is in breach of the principle of the presumption of innocence must be determined in the *context of the particular circumstances* in which the impugned statement was made. ( 25 ) In cases of ‘unfortunate’ language, it is necessary to look at the context of the proceedings as a whole and their special features to determine whether the statement gives rise to a violation of Article 6(2) ECHR. ( 26 ) In *Daktaras v. Lithuania* , ( 27 ) the impugned statements were made by a prosecutor in the course of a reasoned decision at a preliminary stage of the proceedings, rejecting the applicant’s request to discontinue the prosecution. The Court found that the assertion that the applicant’s guilt had been ‘proved’ by the evidence in the case file was ‘unfortunate’, but that, having regard to the context, the prosecutor was referring not to the question whether the applicant’s guilt had been established by the evidence – which was clearly not one for the determination of the prosecutor – but to the question whether the case file disclosed sufficient evidence of the applicant’s guilt to justify proceeding to trial. The presumption of innocence was not considered to have been breached in that case.\n\n54\\. Nevertheless, *Daktaras v. Lithuania* , in which the impugned statements were made by a prosecutor, is an opportunity to recall that judges’ statements are subject to a more thorough examination than the statements made by investigative authorities such as the police and the prosecutor’s office. ( 28 )\n\n55\\. Last, I should also mention the recent judgment in *Ravier v. France* , in which the ECtHR dismissed the violation of the presumption of innocence by the Conseil d’État (Council of State, France), in expressing its view that the applicant had ‘engaged in fraudulent conduct intended to undermine the fairness of the election’, because the Council of State’s decision had remained in the purely electoral sphere. ( 29 )\n\n56\\. The referring court must therefore take into account both the judicial authorities’ choice of words and the particular circumstances in which the words were used, such as the nature and context of the proceedings in question.\n\n*(ii)*  *Application in the present case*\n\n57\\. It should be recalled that, in proceedings under Article 267 TFEU, which is based on a clear separation of functions between the national courts and the Court of Justice, any assessment of the facts in the case is a matter for the national court. However, in order to give the national court a useful answer, the Court may, in a spirit of cooperation with national courts, provide it with all the guidance that it deems necessary. ( 30 )\n\n58\\. In the first place, I should point out that the indictment here relates to the criminal offence of slander, the three constituent elements of which are, under Slovak law: (i) disclosure; (ii) the fact that it involves false information; and (iii) the fact that it is ‘capable of considerably damaging the respect’ for the person concerned, particularly in their professional or family life. ( 31 )\n\n59\\. In the second place, as regards the choice of words, it is clear from certain passages of the statement of reasons that the appeal court seems convinced that the constituent elements of the offence are indeed present in the case at hand. It is stated, for example, that ‘false information’ was ‘shared with a large number of followers’ and the evidence taken during the pre-trial proceedings and on which the Okresný súd Bratislava III (Bratislava III District Court) also based its findings ‘shows beyond doubt’ that the accused ‘has been using lies’ about the aggrieved person.\n\n60\\. It seems to me that those words must be regarded as going beyond mere suspicion and as establishing, on the basis of the incriminating evidence uncovered during the investigation, the guilt of the accused person with regard to the false nature of the information disclosed.\n\n61\\. In so doing, the appeal court may be considered to have taken a position on the incriminating evidence by determining whether the constituent elements of the criminal offence at issue are present, using words that go beyond mere suspicion and that clearly reflect an opinion that the accused person is guilty.\n\n62\\. In the third place, as regards the nature of the decision at issue and its context, it should be emphasised that it is true that the guilt of the accused person was not a matter to be determined at that preliminary stage of the proceedings. It was not a question of whether the facts of the case were proven, but whether, if so, they could be classified as a criminal offence under national law.\n\n63\\. Yet that reasoning cannot justify remarks infringing the presumption of innocence, or it would be sufficient to point out that the guilt of the accused person is not yet at issue to exclude any possibility of a breach of the presumption of innocence at that stage. Such an approach cannot be accepted, since the guarantee of the presumption of innocence, as we have seen, governs criminal proceedings in their entirety and not solely the examination of the merits of the charge. ( 32 )\n\n64\\. I should add that, as we have also seen, judges, and in particular investigating judges, must pay particular attention to the choice of words in their decisions. ( 33 )\n\n65\\. Furthermore, in the present case, it should be noted that, in the context of the preliminary procedure, the appeal court seems to be the court of last instance, ( 34 ) that its decision is binding on the court of first instance at that preliminary stage and that its choice of words can also have an influence on the lower court which is to rule on the substance of the case, lending its decision significant weight in the entire proceedings, regardless of whether the accused person is eventually sent before the criminal court or the civil court.\n\n66\\. Moreover, there is no doubt that expressions that respect the presumption of innocence can easily be used at that stage of the proceedings, ( 35 ) while being sufficiently convincing about the presence of the constituent elements of the offence so as to justify, if necessary, the referral of the accused person to the criminal court.\n\n67\\. In addition, I am not convinced by the Slovak Government’s argument that it is appropriate to distinguish between, on the one hand, expressions intended to state the facts inferred from the case file and evidence that is intended to corroborate the legal opinion concerning the criminal nature or otherwise of the acts of slander at issue, and, on the other hand, expressions related to guilt in substance, which could imply that the presumption of innocence has been infringed. Such an approach suggests that the decision contains passages that are severable from each other having regard to respect for the presumption of innocence. Although it is true that an unfortunate word or expression may possibly be understood as being, in certain circumstances, accidental, judges have a duty not to give the impression that they are acting with the preconceived idea of the accused person’s guilt. To my mind, that means that they must choose the words of their entire statement of reasons with special attention and care. ( 36 )\n\n68\\. Admittedly, it is possible that insinuations or ambiguous passages having regard to the presumption of innocence may be considered accidental or balanced by other passages of the statement of reasons reflecting the fact that the court is not of the opinion that the accused person is guilty. ( 37 ) Nevertheless, comments implying that a judge who is not ruling on the substance of the case believes that the accused person is guilty may be sufficient to vitiate the entire statement of reasons, which must be assessed on a case-by-case basis.\n\n69\\. Therefore, although the right to respect for the presumption of innocence provided for in Article 4(1) of Directive 2016\u002F343, read in the light of Article 48(1) of the Charter, does not preclude a higher court, ruling in an appeal brought by the prosecutor against the decision of the court of first instance to discontinue the proceedings, from taking a position on the constituent elements of the criminal offence at issue in order to justify its decision whether or not to annul the discontinuance, it is only on condition that it does not refer to the accused person as being guilty, which it is for the referring court to assess.\n\n70\\. Is that observation altered by the fact that *the appeal court was not required* under national law to carry out an assessment of the factual and legal issues in the grounds for its decision (second question), even though *not all the evidence was taken* (third question)? That is what we now need to examine.\n\n*(2)*  *The conditions for adopting a preliminary decision (second and third questions)*\n\n71\\. I should point out that Directive 2016\u002F343 confines itself, in accordance with recital 10 thereof, to establishing common minimum rules on the protection of procedural rights of suspects and accused persons, in order to strengthen the trust of Member States in each other’s criminal justice systems and thus to facilitate mutual recognition of decisions in criminal matters. Accordingly, in the light of the minimal degree of harmonisation pursued therein, Directive 2016\u002F343 cannot be interpreted as being a complete and exhaustive instrument intended to lay down all the conditions for the adoption of preliminary decisions.\n\n72\\. The Court has already had occasion to rule on the conditions for the adoption of preliminary decisions with regard to the presumption of innocence, including decisions on the continued detention on remand pending trial. Such decisions may thus be based on suspicions or incriminating evidence, provided that they do not refer to the detained person as being guilty. ( 38 ) Similarly, where the competent court examines the reasonable grounds for believing that the suspect or the accused person has committed the offence with which he or she is charged, in order to give a ruling on the legality of a pre-trial detention decision, that court may compare the elements of incriminating and exculpatory evidence presented to it and give reasons for its decision, not only stating the evidence relied on, but also ruling on the objections of the defence counsel of the person concerned, provided that that decision does not present the person detained as being guilty. ( 39 )\n\n73\\. In addition, the Court has held that the discretion of the Member States in adopting the measures necessary for the purposes of Article 4(1) of Directive 2016\u002F343 included the possibility of referring to a plea-bargaining agreement between the prosecutor and an accused person who pleads guilty, and mentioning other accused persons, on condition, however, that the latter are not referred to as guilty. ( 40 ) The same reasoning was applied, *mutatis mutandis* , with respect to a decision by which a national court accepted a ‘guilty plea’ from one of the co-defendants. ( 41 )\n\n74\\. Accordingly, Article 4(1) of Directive 2016\u002F343, as interpreted by the Court, leaves the Member States a degree of latitude as regards the conditions for the adoption of a decision taken at a preliminary stage by a national court, and in particular as regards the procedures for hearing the evidence and the extent of the statement of reasons which it is required to provide in response to the arguments submitted to it, such questions being covered solely by national law.\n\n75\\. In those circumstances, in the first place, the referring court questions the assessment of the factual and legal issues carried out by the appeal court in the grounds of its decision, even though national law did not require it to do so (second question).\n\n76\\. In that regard, apart from the fact that the referring court does not cite any provision of its national law, I am of the opinion that such detailed grounds do not alter the interpretation of the provisions at issue with regard to the presumption of innocence.\n\n77\\. First of all, as the Slovak Government points out, the appeal court rules by way of an order and, according to Paragraph 176 of the Code of Criminal Procedure, any order must, as a court decision, be duly reasoned.\n\n78\\. Moreover, and in any event, for the purposes of the sound administration of justice, judicial decisions, including at that preliminary stage of the proceedings, must adequately state the grounds on which they are based, provided that they do not contain wording prejudging the guilt of the accused person. A sufficiently detailed statement of reasons seems capable of providing the accused person with better information on the charges against him or her and the reasons why the facts on which the criminal proceedings are based could constitute offences justifying the referral of the substance of the case to the criminal court. ( 42 )\n\n79\\. I would add that such a statement of reasons, provided that it respects the presumption of innocence, may be considered particularly appropriate in a case such as the one at issue in the main proceedings, in which the appeal court adjudicates for the second time on the order for dismissal from the court of first instance.\n\n80\\. It does not seem, therefore, that the provisions and principles of EU law invoked by the referring court must be interpreted as precluding a detailed statement of reasons, provided that it does not prejudge the guilt of the accused person.\n\n81\\. In the second place, the fact that the appeal court ruled solely on the basis of evidence taken in the investigation procedure – that is to say, at a stage when not all the evidence had been taken (third question) – also has no effect on the answer to the first question.\n\n82\\. I should point out that, at the preliminary stage of the procedure at issue in the main proceedings, it is a question – as is apparent from the observations of the Slovak Government – of determining, with regard to the investigation file, whether the indictment examined is a reliable basis for the continuation of the proceedings, whether the investigation was conducted in accordance with the law and whether a trial hearing may take place or whether a different approach is needed. The court may therefore close the case if the offence at issue is not criminal, which may be justified by mere doubts or contradictory evidence. The court’s assessment focuses here on the question whether the classification of the acts at issue falls under criminal law.\n\n83\\. Hypothetically, only the evidence adduced in the course of the pre-trial judicial investigation may be used at this stage. This is expressly provided for in Article 4(1) of Directive 2016\u002F343, when it mentions the possibility for the judicial authorities or other competent authorities, when making preliminary decisions of a procedural nature, as is the case here, to rely on suspicion or incriminating evidence.\n\n84\\. The presumption of innocence cannot prevent a court hearing an appeal against a preliminary decision from relying on the evidence available to it at that stage of the proceedings to adopt its decision, provided that it does not refer to the accused person as being guilty.\n\n85\\. In the third place, the referring court seems to have doubts concerning respect for the rights of the defence, guaranteed by Article 48(2) of the Charter, on the ground that the appeal court’s decision was adopted without the accused person being able to respond.\n\n86\\. I recall in that regard that, according to the long-standing and consistent case-law of the ECtHR, the guarantees of a fair trial under Article 6 ECHR are applicable as soon as there is a ‘criminal charge’ within the meaning of that provision. They may also be relevant before a case is sent for trial if and in so far as the fairness of the trial is likely to be seriously prejudiced by an initial failure to comply with them. ( 43 ) The guarantees of a fair trial, including the rights of the defence, thus apply from the pre-trial judicial investigation stage. ( 44 )\n\n87\\. In principle, however, the assessment takes place throughout the proceedings and according to the circumstances of the case. ( 45 )\n\n88\\. As regards the rights of the defence in appeal proceedings, it is apparent from the information in the file that the accused person was able to submit observations in response to the prosecutor’s observations during the appeal stage. Moreover, as regards the rights of the defence concerning respect for the presumption of innocence in the order from the appeal court, it must be stated that the proceedings are to continue before the referring court, before which the accused person will be able to put forward his arguments.\n\n89\\. In those circumstances, it seems that the referring court’s doubts concerning respect for the rights of the defence may be dispelled.\n\n**(c)** **Conclusions on the first three questions**\n\n90\\. In my opinion, it is appropriate to answer the referring court that the right to respect for the presumption of innocence provided for in Article 3 and Article 4(1) of Directive 2016\u002F343, read in the light of Article 48(1) of the Charter, does not preclude a criminal court, hearing an appeal against a preliminary decision of a procedural nature, from taking a position on the incriminating evidence by carrying out an assessment in fact and in law as to whether the constituent elements of the criminal offence at issue are present, when it was not required to do so under national law and not all the evidence was taken at that preliminary stage of the proceedings, provided that it does not refer to the accused person as being guilty, which it is for the referring court to assess.\n\n**2.** **The fourth question**\n\n91\\. In view of my proposed answer to the first question and since it will be for the referring court to determine whether there has been a breach of the presumption of innocence in the case in the main proceedings, I consider it appropriate to answer the fourth question, the purpose of which is to identify the measures to be adopted in such a case.\n\n92\\. By the first two parts of the fourth question, ( 46 ) the referring court asks whether, in the event that the presumption of innocence is found to have been infringed, the fact of disregarding the rulings of the appeal court setting aside an order that no further action be taken, before adjudicating either after the proper taking of evidence (first part), or by issuing the same decision as the one previously set aside by the appeal court (second part), would constitute an appropriate measure within the meaning of Article 4(2) of Directive 2016\u002F343.\n\n93\\. I should point out that, although Article 4(2) of Directive 2016\u002F343 requires that provision be made for ‘appropriate measures’, it does not specify the nature of those measures and merely refers ‘in particular’ to Article 10 of that directive, which implies that those measures could, inter alia, take the form of effective remedies available to the accused person within the meaning of that provision. Furthermore, in the light of the minimal degree of harmonisation pursued therein, Directive 2016\u002F343 cannot be interpreted as being a complete and exhaustive instrument intended to lay down all the conditions for the adoption of appropriate measures. ( 47 )\n\n94\\. Accordingly, in the absence of EU rules on the matter, ‘appropriate measures’ should be defined by the Member States in accordance with their procedural autonomy, within the limits resulting from the principles of equivalence and effectiveness. ( 48 )\n\n95\\. As a preliminary point, I note that the referring court does not mention the national remedies put in place in that regard.\n\n96\\. The Slovak Government refers, for its part, to four remedies in the event of a breach of the presumption of innocence by a higher court, namely: (i) the possibility of raising a plea of lack of impartiality of the judges, if the criminal case is subsequently assigned to them to rule on the substance of the case; (ii) the possibility of a liability claim against the State; (iii) the possibility of a constitutional challenge; and (iv) the possibility of lodging a claim of fault-based liability against the judges who committed such a breach. ( 49 )\n\n97\\. The Commission, for its part, maintains that it is necessary to examine first what available measures could place the suspects or accused persons in the same position in which they would have found themselves had the breach not occurred. Further, the Commission submits that it is only if national law does not afford any such opportunity that the question arises as to whether the decision of the higher court should be left unapplied.\n\n98\\. I note, however, that it is apparent from the question put by the referring court that it concerns not the appropriate measures taken *by* *the Member State* to guarantee effective remedies in the event of a breach of the presumption of innocence, but how the referring court ensures the effectiveness of respect for that presumption in the event of a breach in that regard by the higher appeal court.\n\n99\\. Accordingly, it seems to me that, without prejudice to the appropriateness or otherwise of the remedies under Article 4(2) of Directive 2016\u002F343 put in place in national law, their existence does not preclude a court faced with a breach of the presumption of innocence at the preliminary stage of criminal proceedings, as an ordinary court in matters of EU law, from having to adopt certain measures to ensure that it is respected. ( 50 )\n\n100\\. In the first place, the referring court raises the possibility of disregarding a decision of a higher court that it considers contrary to the presumption of innocence.\n\n101\\. It seems to me that such a measure is based less on Article 4(2) of Directive 2016\u002F343 concerning appropriate measures than on the primacy of the presumption of innocence under Article 48(1) of the Charter and Article 3 of that directive.\n\n102\\. Indeed, it follows from the written and oral observations of the interested parties, and in particular the Slovak and Hungarian Governments, that such a measure would run counter to the rule that the legal opinion of the higher court is, under national law, legally binding on the referring court. ( 51 )\n\n103\\. However, it is apparent from the Court’s case-law that the principle of primacy of EU law precludes a national court which is called upon to decide a case referred back to it by a higher court hearing an appeal from being bound, in accordance with national procedural law, by legal rulings of the higher court, if it considers, having regard to the interpretation which it has sought from the Court, that those rulings are inconsistent with EU law. ( 52 )\n\n104\\. It follows that, in so far as the referring court considers, having regard to the judgment of the Court to be delivered, that the statement of reasons of the appeal court infringed the presumption of innocence, it should, within the exercise of its jurisdiction to decide a case referred back to it by the appeal court, not consider itself bound by the *substantive rulings* of the appeal court which it considers to be vitiated by such a breach and disregard them.\n\n105\\. In that regard, I am not convinced by the Slovak Government’s argument regarding the ‘ancillary’ nature of a possible breach of the presumption of innocence, on the ground that it does not concern an ‘integral part of the legal ruling of the higher court’. Admittedly, the legal ruling of the higher court at that stage of the proceedings is not necessarily binding in subsequent proceedings concerning the accused person’s guilt or innocence in substance. Indeed, any breach of the presumption of innocence at that stage could be remedied at a later stage of the proceedings. ( 53 ) Nevertheless, it cannot be ruled out that the initial failure to comply with the requirement for a fair trial before a case is sent for trial may seriously prejudice the fairness of the trial. ( 54 ) As regards the failure to comply with the presumption of innocence, which is one of the components of a fair criminal trial, even though the terms of the appeal court’s decision are not binding on the court called upon to rule on the substance of the case, that preliminary decision could result in the accused person being sent for trial before the criminal court, which is a *decisive* step of the criminal proceedings. Furthermore, it is possible that the rulings of the appeal court at that preliminary stage, although contrary to the presumption of innocence, could carry some weight with the court adjudicating on the substance, which, moreover, is the lower court, with potential implications for the burden of proof.\n\n106\\. Accordingly, the principle of primacy of EU law requires, from that preliminary stage of the criminal proceedings, the referring court to disregard rulings of the appeal court that it considers contrary to the presumption of innocence.\n\n107\\. In the second place, the referring court is uncertain as to the procedural consequences it must infer from the finding of a breach of the presumption of innocence by the appeal court, and specifically whether, in the light of the principles of primacy and effectiveness, it must re-examine the evidence and reassess whether all the constituent elements of the offence are present.\n\n108\\. To understand the scope of that question, it should be recalled that, in that preliminary stage of the proceedings, the appeal court twice set aside the order of the court of first instance and referred the case back to it for a ‘new decision’. The order of the appeal court of 8 February 2022 stated that the court of first instance had to ‘re-examine the evidence obtained and, only then, individually assess whether all the constituent elements of the alleged offence were present’, before deciding the case in accordance with the applicable legal provisions. It should also be noted that even though the referring court took the view that the appeal court, in at least some of its substantive findings as set out in the statement of reasons, infringed the presumption of innocence, the fact remains that its annulment of the decision to take no further action has no bearing on whether the accused person is actually guilty and it is of a purely procedural nature.\n\n109\\. I would also point out that, in the context of the division of roles between the Court and the national courts which is the basis of Article 267 TFEU, it is not for the Court to interpret the provisions of national law, to rule on the compatibility of a national measure with EU law or to state which rule or remedy provided for by national law a referring court must follow. It is for the referring court to interpret its national law within the framework of its procedural autonomy, subject to compliance with the principle of equivalence, which requires that national rules are not less favourable than those governing similar domestic situations, and the principle of effectiveness, which requires that national procedural rules do not make it excessively difficult or impossible in practice to exercise the rights conferred by EU law. ( 55 ) However, in order to give the referring court some useful guidance, I consider it appropriate to specify the following.\n\n110\\. While, according to the principles of primacy and effectiveness, ( 56 ) any rulings of the appeal court that are contrary to the presumption of innocence must be disregarded, those principles do not seem necessarily to require the procedural effects of the appeal court’s decision to be disregarded. I draw a distinction between, on the one hand, an infringement of EU law that stems from the organic requirements of the court in question, which could lead to a declaration that a decision delivered by a body that does not constitute an independent and impartial tribunal previously established by law within the meaning of EU law ( 57 ) is null and void, and, on the other hand, a ‘circumstantial’ infringement – that is to say, in a given case – of the guarantee of a fair trial, the extent of which would be limited to the rulings given by the appeal court in that case. In other words, it is my view that the effectiveness of the presumption of innocence does not preclude the annulment, by the appeal court, of the decision to take no further action from having procedural effects. Nor does it preclude the referring court, after disregarding the rulings of the appeal court that it considers contrary to the presumption of innocence, from adopting a new decision, potentially finding that the case should be closed without further action, and if necessary re-examining the evidence in the file, as instructed by the appeal court in its decision.\n\n111\\. In addition, if the referring court were to adopt, at that stage, a new, duly reasoned decision, it could justify the fact that it disregarded the substantive rulings considered to be contrary to the presumption of innocence, thereby ensuring the fairness of the proceedings for all parties involved. Last, such a decision could also constitute a legal basis for the accused person to assert, if necessary, a right to compensation for that breach of the presumption of innocence.\n\n112\\. Accordingly, I consider that, at that preliminary stage of the criminal proceedings, the principles of primacy and effectiveness of the presumption of innocence do not preclude the referring court, while disregarding the rulings of the appeal court that it may consider to be contrary to the presumption of innocence, from complying with the order of the appeal court in so far as it requires it to hear and decide the case again.\n\n113\\. I am therefore of the opinion that the answer to the referring court must be that Article 4(2) of Directive 2016\u002F343, read in the light of Article 3 of that directive and Article 48(1) of the Charter, must be interpreted as meaning that it does not preclude the referring court, on the one hand, from disregarding the rulings of the appeal court that it may consider to be contrary to the presumption of innocence and, on the other hand, from hearing and deciding the case again in accordance with its national procedural law, as governed by the principles of equivalence and effectiveness.\n\n**IV.** **Conclusion** 1 ( i ) The name of the present case is a fictitious name. It does not correspond to the real name of any party to the proceedings.\n\n114\\. In the light of the foregoing considerations, I propose that the Court of Justice answer the questions referred for a preliminary ruling by the Mestský súd Bratislava I (Bratislava I City Court, Slovakia) as follows: (1) The right to respect for the presumption of innocence provided for in Article 3 and in Article 4(1) of Directive (EU) 2016\u002F343 of the European Parliament and of the Council of 9 March 2016 on the strengthening of certain aspects of the presumption of innocence and of the right to be present at the trial in criminal proceedings, read in the light of Article 48(1) of the Charter of Fundamental Rights of the European Union, must be interpreted as meaning that it does not preclude a criminal court, hearing an appeal against a preliminary decision of a procedural nature, from taking a position on the incriminating evidence by carrying out an assessment in fact and in law as to whether the constituent elements of the criminal offence at issue are present, when it was not required to do so under national law and not all the evidence was taken at that preliminary stage of the proceedings, provided that it does not refer to the accused person as being guilty, which it is for the referring court to assess. (2) Article 4(2) of Directive 2016\u002F343, read in the light of Article 3 of that directive and Article 48(1) of the Charter, must be interpreted as meaning that it does not preclude the referring court, on the one hand, from disregarding the rulings of the appeal court that it may consider to be contrary to the presumption of innocence and, on the other hand, from hearing and deciding the case again in accordance with its national procedural law, as governed by the principles of equivalence and effectiveness.\n\n* * *\n\n( 2 ) Badinter, R., ‘La présomption d’innocence, histoire et modernité’, in Catala, P., *Le droit privé français à la fin du XXe siècle: études offertes à Pierre Catala* , Litec, Paris, 2001, p. 134.\n\n( 3 ) See Vigouroux, C., ‘Présomption d’innocence’, in *La conscience des droits. Mélanges en l’honneur de Jean-Paul Costa* , Dalloz, Paris, 2011, p. 657 et seq., especially p. 665.\n\n( 4 ) Directive of the European Parliament and of the Council of 9 March 2016 on the strengthening of certain aspects of the presumption of innocence and of the right to be present at the trial in criminal proceedings ([OJ 2016 L 65, p. 1](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Foj\u002FJOL_2016_065_R_TOC)).\n\nI should point out that that directive is the result of a step-by-step process based on Article 82 TFEU, concerning judicial cooperation in criminal matters within the European Union, which provides for the principle of mutual recognition of judgments and judicial decisions, generally considered to be the cornerstone of judicial cooperation in criminal matters within the European Union since the Tampere European Council of 15 and 16 October 1999. The European Commission Green Paper of 26 April 2006 on the presumption of innocence (COM(2006) 174 final, 24 June 2006) was followed by the resolution of the Council of 30 November 2009 on a Roadmap for strengthening procedural rights of suspected or accused persons in criminal proceedings ([OJ 2009 C 295, p. 1](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Foj\u002FJOC_2009_295_R_TOC)), which became an integral part of the European Council’s Stockholm Programme on the area of freedom, security and justice (The Stockholm Programme – An open and secure Europe serving and protecting citizens) ([OJ 2010 C 115, p. 1](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Foj\u002FJOC_2010_115_R_TOC)), and led, in November 2013, to the Proposal for a Directive of the European Parliament and of the Council on the strengthening of certain aspects of the presumption of innocence and of the right to be present at trial in criminal proceedings (COM(2013)821 final), followed by Directive 2016\u002F343, which was adopted on 9 March 2016 (see Daminova, N., ‘The CJEU Faced with “Presumption of Innocence” Directive 2016\u002F343: Reshaping the ECHR Standards?’, in *European Yearbook of Human Rights* , Intersentia, 2022, pp. 137 to 175). Based on the competence conferred on the Union by Article 82(2)(b) TFEU to lay down minimum rules on the rights of individuals in criminal procedure, it is one of six directives adopted by the European Union to establish common minimum standards guaranteeing the right of access to an impartial court and the rights of suspects and accused persons under the Stockholm Programme (see Catteau, F., ‘Les droits procéduraux des personnes poursuivies et des victimes de criminalité devant la Cour de justice de l’Union. Vers un droit européen de la procédure pénale?’, *Cahiers de droit européen* , 2020\u002F2-3, pp. 485 to 547).\n\n( 5 ) See judgment of 8 April 2025, EPPO (Judicial review of procedural acts) ([C‑292\u002F23](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A255), [EU:C:2025:255](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A255), paragraph [36](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A255#point36) and the case-law cited).\n\n( 6 ) See judgment of 16 November 2021, Prokuratura Rejonowa w Mińsku Mazowieckim and Others ([C‑748\u002F19 to C‑754\u002F19](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2021%3A931), [EU:C:2021:931](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2021%3A931), paragraph [48](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2021%3A931#point48)). 7 ( 8 ) See judgment of 28 November 2019, Spetsializirana prokuratura ([C‑653\u002F19 PPU](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A1024), [EU:C:2019:1024](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A1024), paragraphs [32](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A1024#point32) and [33](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A1024#point33); ‘the judgment in *Spetsializirana prokuratura* ’).\n\n( 9 ) See the judgment in *Spetsializirana prokuratura* (paragraphs 35 and 37): ‘a judicial decision having as its sole purpose the potential continued detention on remand pending trial of an accused person seeks only to resolve the question whether that person must be released or not, in the light of all the relevant circumstances, without establishing whether that person is guilty of having committed the offence with which he [or she] is charged. … Thus, that decision cannot be considered [to be] a judicial decision on the guilt of the accused person for the purposes of that directive’.\n\n( 10 ) In that regard, Article 52(3) of the Charter states that, in so far as it contains rights which correspond to rights guaranteed by the ECHR, the meaning and scope of those rights must be the same as those laid down by that convention. As is apparent from the explanations relating to Article 48 of the Charter, which, in accordance with the third subparagraph of Article 6(1) TEU and Article 52(7) of the Charter, must be taken into consideration for the interpretation of the Charter, Article 48 corresponds to Article 6(2) and (3) ECHR. The Court must, accordingly, ensure that its interpretation of Article 48(1) of the Charter ensures a level of protection which does not disregard that guaranteed by Article 6(2) ECHR, as interpreted by the ECtHR (see, to that effect, judgment of 23 November 2021, IS (Illegality of the order for reference) ([C‑564\u002F19](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2021%3A949), [EU:C:2021:949](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2021%3A949), paragraph [101](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2021%3A949#point101))).\n\n( 11 ) See, to that effect, judgment of 5 September 2019, AH and Others(Presumption of innocence) ([C‑377\u002F18](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A670), [EU:C:2019:670](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A670), paragraph [42](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A670#point42)). For an application concerning the terms of a press release issued by the European Anti-Fraud Office (OLAF), see judgment of 1 October 2025, OC v Commission ([T‑384\u002F20 RENV](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AT%3A2025%3A925), [EU:T:2025:925](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AT%3A2025%3A925), paragraph [91](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AT%3A2025%3A925#point91) et seq.).\n\n( 12 ) See ECtHR, 25 March 1983, *Minelli v. Switzerland* (CE:ECHR:1983:0325JUD000866079, § 30).\n\n( 13 ) See Repík, B., ‘Réflexions sur la jurisprudence de la Cour européenne des droits de l’homme concernant la présomption d’innocence’, in *Liber amicorum Marc-André Eissen* , Brussels, Bruylant\u002FL.G.D.J., 1995, pp. 331 to 345, especially p. 333. See, also, Dold, B., ‘Artikel 6. Recht auf ein faires Verfahren’, in Frowein, J.A., and Peukert, W. *Europäische MenschenRechtsKonvention* , *EMRK-Kommentar* , 4th edition, N.P. Engel Verlag, Kehl am Rhein, Nos 67 to 69, 2024, pp. 323 to 324.\n\n( 14 ) Viewed as a procedural guarantee in the context of the criminal trial itself, the presumption of innocence imposes requirements in respect of, inter alia, the burden of proof, legal presumptions of fact and law, the privilege against self-incrimination, pre-trial publicity and premature expressions, by the trial court or by other public officials, of a defendant’s guilt (see ECtHR, 12 July 2013, *Allen v. the United Kingdom* (CE:ECHR:2013:0712JUD002542409, § 93), and ECtHR, 11 June 2024, *Nealon and Hallam v. the United Kingdom* (CE:ECHR:2024:0611JUD003248319, § 101)).\n\n( 15 ) In keeping with the need to ensure that the right guaranteed by Article 6(2) is practical and effective, the ECtHR has, over time, developed a ‘second aspect’ to the presumption of innocence, which comes into play after the criminal proceedings have concluded, either with an acquittal or a discontinuance. The ECtHR has made it clear that following the discontinuance of criminal proceedings the presumption of innocence required that the lack of a person’s criminal conviction be preserved in any other proceedings of whatever nature; and also that the operative part of an acquittal judgment must be respected by any authority referring directly or indirectly to the criminal responsibility of the interested party (ECtHR, 11 June 2024, *Nealon and Hallam v. the United Kingdom* (CE:ECHR:2024:0611JUD003248319, §§ 102 to 109 and the case-law cited)). See, in that regard, Beernaert, M.-A., ‘La présomption d’innocence et ses prolongements au-delà du procès pénal (obs. sous Cour eur. dr. h., Gde Ch, arrêt *Nealon et Hallam c. Royaume-Uni* , 11 juin 2024)’, *Revue trimestrielle des droits de l’homme* , Vol. 2, No 142, 1 April 2025, pp. 515 to 526.\n\nSee, also, on the distinction between the two aspects of the presumption of innocence, Jebens, S.E., ‘The Scope of the Presumption of Innocence in Article 6 § 2 of the Convention – Especially on its Reputation-Related Aspect’, in *Droits de l’homme – Regards de Strasbourg, Liber Amicorum Luzius Wildhaber* , N.P. Engel, Kehl, 2007, pp. 207 to 227. 16 ( 17 ) See, inter alia, ECtHR, 6 December 1988, *Barberà, Messegué and Jabardo v. Spain* (CE:ECHR:1988:1206JUD001059083, § 77).\n\nLet us also consider an example from judicial history. In the infamous Landru trial that began on 7 November 1921 before the cour d’assises de Seine-et-Oise (Assize Court of Seine-et-Oise, Versailles, France), the oral argument of Vincent de Moro-Giafferi, Landru’s lawyer, illustrated that principle by asking Robert Godefroy, the public prosecutor, whether ‘a man can be convicted for the murder of 10 women today, and yet the following day no finding of presumed death can be issued to the families, in the absence of any bodies?’. In a now famous scene, Landru’s lawyer went even further, claiming that one of the victims, presumed dead, had been found and was ready to appear. The entire courtroom, including the jurors, reportedly turned towards the door … which was proof, according to Landru’s lawyer, that no one was certain of Landru’s guilt. It should be added that the public prosecutor immediately responded with the argument that only Landru had not turned his head (for details of the Landru trial, see the website of the French Ministry of Justice available at: https:\u002F\u002Fwww.justice.gouv.fr\u002Factualites\u002Factualite\u002Fproces-landru).\n\n( 18 ) See ECtHR, 10 February 1995, *Allenet de Ribemont v. France* (CE:ECHR:1995:0210JUD001517589, § 35), *Revue trimestrielle des droits de l’homme* , No 24, 1995, pp. 661 to 672, note by Spielmann, D. See, also, ECtHR, 12 April 2012, *Lagardère v. France* (CE:ECHR:2012:0412JUD001885107, § 74), and ECtHR, 27 February 2014, *Karaman v. Germany* (CE:ECHR:2014:0227JUD001710310, § 41).\n\n( 19 ) See Hirsch, C., ‘La présomption d’innocence dans la jurisprudence de la Cour européenne des droits de l’homme (2006-2013)’, *Annales du droit luxembourgeois* , Vol. 23 (2013), Bruylant, Brussels, 2015, especially p. 160 et seq. See, also, ECtHR, 10 October 2000, *Daktaras v. Lithuania* (CE:ECHR:2000:1010JUD004209598, § 41), and ECtHR, 15 June 2021, *Vardan Martirosyan v. Armenia* (CE:ECHR:2021:0615JUD001361012, §§ 83 and 87) (citing the case-law on the choice of words in cases of continued detention on remand pending trial).\n\n( 20 ) See ECtHR, 28 November 2002, *Marziano v. Italy* (CE:ECHR:2002:1128JUD004531399, §§ 30 and 31), and ECtHR, 12 April 2012, *Lagardère v. France* (CE:ECHR:2012:0412JUD001885107, § 75).\n\n( 21 ) ECtHR, 28 November 2002, *Marziano v. Italy* (CE:ECHR:2002:1128JUD004531399, § 30).\n\n( 22 ) ECtHR, 28 November 2002, *Lavents v. Latvia* (CE:ECHR:2002:1128JUD005844200, §§ 126 and 127).\n\n( 23 ) ECtHR, 15 October 2013, *Gutsanovi v. Bulgaria* (CE:ECHR:2013:1015JUD003452910, §§ 202 and 203). 24 ( 25 ) See, inter alia, ECtHR, 26 March 1982, *Adolf v. Austria* (CE:ECHR:1982:0326JUD000826978, §§ 36 to 41), and ECtHR, 27 February 2014, *Karaman v. Germany* (CE:ECHR:2014:0227JUD001710310, §§ 63 to 65 and the case-law cited). See, also, ECtHR, 15 June 2021, *Vardan Martirosyan* *v. Armenia* (CE:ECHR:2021:0615JUD001361012, § 81). In paragraphs 83 to 90 of that judgment, the ECtHR carefully examined the wording of each decision at issue in their respective contexts. For example, the expression ‘the act committed’, although unfortunate, was to be understood in the context of the extension of pre-trial detention, taking into account the nature and the dangerousness of the act ‘committed’, further stating that the applicant ‘was charged with that offence’, the applicant’s guilt not being an issue to be determined at that stage, as the court of appeal noted (§ 84). See, conversely, §§ 87 and 88 for a violation of Article 6(2) ECHR concerning another decision, not corrected at a later stage of the proceedings.\n\n( 26 ) See ECtHR, 3 October 2019, *Fleischner v. Germany* (CE:ECHR:2019:1003JUD006198512, § 65), in which the ECtHR states that national law is part of that context.\n\n( 27 ) ECtHR, 10 October 2000, *Daktaras v. Lithuania* (CE:ECHR:2000:1010JUD004209598, §§ 44 and 45).\n\n( 28 ) In ECtHR, 21 September 2006, *Pandy v. Belgium* (CE:ECHR:2006:0921JUD001358302, § 19 and 43), comments made by an investigating judge had been picked up by the press (‘Pandy sees himself as Dreyfus, but I have other comparisons for him: Landru and Dr Petiot’). The comments were viewed as possibly unintentional and secondary in the context of a complex investigation that went smoothly, as the applicant himself acknowledged. However, the ECtHR held that they were still questionable from the point of view of the principle of the presumption of innocence, since they compared the applicant to known and recognised serial killers. Such comments were ‘not acceptable from an examining judge, responsible under Belgian law for investigating both inculpatory and exculpatory evidence, which justified a more thorough examination’ (§ 45).\n\n( 29 ) ECtHR, 19 June 2025, *Ravier v. France* (CE:ECHR:2025:0619JUD003232422, §§ 40 to 42).\n\n( 30 ) See judgments of 1 July 2008, MOTOE ([C‑49\u002F07](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2008%3A376), [EU:C:2008:376](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2008%3A376), paragraph [30](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2008%3A376#point30)), and of 30 May 2024, Amazon Services Europe ([C‑665\u002F22](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2024%3A435), [EU:C:2024:435](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2024%3A435), paragraph [40](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2024%3A435#point40)). 31 ( 32 ) See ECtHR, 25 March 1983, *Minelli v. Switzerland* (CE:ECHR:1983:0325JUD000866079, § 30), and ECtHR, 30 March 2010, *Poncelet v. Belgium* (CE:ECHR:2010:0330JUD004441807, § 50).\n\n( 33 ) See ECtHR, 21 September 2006, *Pandy v. Belgium* (CE:ECHR:2006:0921JUD001358302, § 43), and ECtHR, 21 June 2007, *Kampanellis v. Greece* (CE:ECHR:2007:0621JUD000902905, § 27) (concerning the words used in the context of a committal for trial by an investigating judge). See also Dold, B., op. cit., No 367, p. 459.\n\n( 34 ) The orders of the appeal court state that they are not amenable to ordinary appeal. See, in that regard, ECtHR, 15 June 2021, *Vardan Martirosyan* *v. Armenia* (CE:ECHR:2021:0615JUD001361012, § 88). The ECtHR held that it was prepared to entertain the possibility that the district court may have committed merely a technical error in poorly wording its decision, but notes that it subsequently never corrected it, the impugned decision not being amenable to appeal. 35 ( 36 ) For example, in ECtHR, 21 June 2007, *Kampanellis v. Greece* (CE:ECHR:2007:0621JUD000902905, § 28), one such phrase was found in the judges’ reasoned decision, adopted at a preliminary stage of the criminal proceedings brought against the applicant, by which his committal for trial was ordered. The ECtHR held that the use of the expression ‘to quieten down’ unequivocally expressed the idea that the applicant’s pre-trial detention could have had the effect of calming him down and curbing his criminal behaviour. It considered that, regardless of their position in the contested order, such remarks were not admissible from judges in an indictment division, responsible under Greek law for investigating both the inculpatory and exculpatory evidence, which required special care in the choice of words used.\n\n( 37 ) See, by analogy, judgment of 8 December 2022, HYA and Others(Impossibility of questioning prosecution witnesses) ([C‑348\u002F21](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2022%3A965), [EU:C:2022:965](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2022%3A965), paragraph [58](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2022%3A965#point58)), concerning the existence of counterbalancing factors capable of compensating for the difficulty caused to the defendant by the fact that the witness could not be directly cross-examined during the judicial stage of the criminal proceedings.\n\n( 38 ) On the continued detention on remand pending trial, see judgments of 19 September 2018, Milev ([C‑310\u002F18 PPU](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2018%3A732), [EU:C:2018:732](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2018%3A732), paragraphs [44](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2018%3A732#point44) and [48](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2018%3A732#point48)), and the judgment in *Spetsializirana prokuratura* (paragraph 29).\n\n( 39 ) See, to that effect, order of 12 February 2019, RH ([C‑8\u002F19 PPU](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A110), [EU:C:2019:110](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A110), paragraphs [57](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A110#point57) to [60](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A110#point60)).\n\n( 40 ) See judgment of 5 September 2019, AH and Others(Presumption of innocence) ([C‑377\u002F18](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A670), [EU:C:2019:670](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A670), paragraph [50](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A670#point50)).\n\n( 41 ) See order of 28 May 2020, UL and VM ([C‑709\u002F18](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2020%3A411), [EU:C:2020:411](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2020%3A411), paragraph [35](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2020%3A411#point35)).\n\n( 42 ) For example, under Luxembourg law, the role of the indictment division is to ‘determine whether indicia exist for each fact forming a key element of the pre-trial judicial investigation; it then assesses whether the indicia should be considered to be sufficient (Le Poittevin, *Code d’instruction criminelle annoté* , Article 221, Nos 27 and 28). … the investigating judge is only required to decide whether there are sufficient indicia to suspect that the accused committed the offence in circumstances that fall under the application of criminal law’ (see Thiry, R., *Précis d’Instruction Criminelle en Droit Luxembourgeois* , Vol. II, Lucien de Bourcy, Luxembourg, 1984, pp. 204 and 205). See also, on the prime facie evidence and sufficient grounds, the role of investigating judges in the Belgian judicial system (Franchimont, M., Jacobs, A. and Masset, A., *Manuel de procédure pénale* , 3rd edition, Larcier, Brussels, 2009, p. 541) and the French judicial system (pursuant to Article 184 of the French Code of Criminal Procedure).\n\n( 43 ) See ECtHR, 24 November 1993, *Imbrioscia v. Switzerland* (CE:ECHR:1993:1124JUD001397288, § 36), and ECtHR, 27 November 2008, *Salduz v. Türkiye* (CE:ECHR:2008:1127JUD003639102, § 50).\n\n( 44 ) See commentary by Dold, B., op. cit., No 67, especially p. 323. See, also, ECtHR, 9 November 2018, *Beuze v. Belgium* (CE:ECHR:2018:1109JUD007140910, § 147 et seq.), in which the ECtHR recalls that ‘the principle of placing the overall fairness of the proceedings at the heart of the assessment is not limited to the right of access to a lawyer under Article 6 § 3 (c) but is inherent in the broader case-law on defence rights enshrined in Article 6 § 1 [ECHR]’.\n\n( 45 ) See ECtHR, 12 May 2017, *Simeonovi v. Bulgaria* (CE:ECHR:2017:0512JUD002198004, § 114 and especially § 120), where the ECtHR provided a non-exhaustive list of factors to be taken into account, where appropriate, in order to assess the impact of procedural failings at the pre-trial stage on the overall fairness of the criminal proceedings. See, also, ECtHR, 9 November 2018, *Beuze v. Belgium* (CE:ECHR:2018:1109JUD007140910, § 148 et seq.), where the ECtHR states that such an approach is also consistent with the role of the ECtHR, which is not to adjudicate in the abstract or to harmonise the various legal systems, but to establish safeguards to ensure that the proceedings followed in each case comply with the requirements of a fair trial, having regard to the specific circumstances of each accused. 46 ( 47 ) See, by analogy, on the conditions for adopting decisions on pre-trial detention, judgment of 19 September 2018, Milev ([C‑310\u002F18 PPU](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2018%3A732), [EU:C:2018:732](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2018%3A732), paragraph [47](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2018%3A732#point47)), and order of 12 February 2019, RH ([C‑8\u002F19 PPU](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A110), [EU:C:2019:110](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A110), paragraph [59](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A110#point59)).\n\n( 48 ) See judgment of 5 September 2024, M.S. and Others(Procedural rights of minors) ([C‑603\u002F22](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2024%3A685), [EU:C:2024:685](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2024%3A685), paragraph [170](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2024%3A685#point170)). 49 ( 50 ) See, by analogy, judgment of 5 June 2018, Kolev and Others ([C‑612\u002F15](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2018%3A392), [EU:C:2018:392](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2018%3A392), paragraph [94](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2018%3A392#point94)), concerning, inter alia, the timing of the defence’s access to the case materials, and judgment of 21 October 2021, ZX (Regularisation of the indictment) ([C‑282\u002F20](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2021%3A874), [EU:C:2021:874](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2021%3A874), paragraphs [28](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2021%3A874#point28) and [32](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2021%3A874#point32)), concerning errors and omissions in the indictment which prejudice the right of the accused person to be provided with detailed information on the charges. 51 ( 52 ) See judgments of 5 October 2010, Elchinov ([C‑173\u002F09](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2010%3A581), [EU:C:2010:581](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2010%3A581), paragraphs [30](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2010%3A581#point30) to [32](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2010%3A581#point32)), and of 22 February 2022, RS (Effect of the decisions of a constitutional court) ([C‑430\u002F21](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2022%3A99), [EU:C:2022:99](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2022%3A99), paragraph [75](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2022%3A99#point75)).\n\n( 53 ) As stated by Kuty, F., *Justice pénale et procès équitable* , 1st edition, Larcier, Brussels, 2023, p. 1711: ‘the disregard of the presumption of innocence by the investigating judge does not constitute an irremediable defect. … any act of the investigating judge affected by a breach of the presumption of innocence does not, in itself, have the effect … of making it impossible to hold a fair trial before the trial court’.\n\n( 54 ) On the right to be assisted by a lawyer from the start of the proceedings, see ECtHR, 24 November 1993, *Imbrioscia v. Switzerland* (CE:ECHR:1993:1124JUD001397288, § 36), and ECtHR, 27 November 2008, *Salduz v. Türkiye* (CE:ECHR:2008:1127JUD003639102, § 54). In that regard, although the criminal justice system functions ‘as a machine for layering guilt’, the ‘“layer” of guilt should not be too thick at the start of the proceedings’ (see Roets, D., *La présomption d’innocence* , 2nd edition, Dalloz, Paris, 2023, pp. 59 and 60).\n\n( 55 ) See judgment of 21 October 2021, ZX (Regularisation of the indictment) ([C‑282\u002F20](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2021%3A874), [EU:C:2021:874](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2021%3A874), paragraph [35](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2021%3A874#point35)). I should point out in that respect that, according to recital 44 of Directive 2016\u002F343, the principle of effectiveness of EU law ‘requires that Member States put in place adequate and effective remedies in the event of a breach of a right conferred upon individuals by [EU] law. An effective remedy, which is available in the event of a breach of any of the rights laid down in this Directive, should, as far as possible, have the effect of placing the suspects or accused persons in the same position in which they would have found themselves had the breach not occurred, with a view to protecting the right to a fair trial and the rights of the defence’. 56 ( 57 ) See judgment of 4 September 2025, AW T ([C‑225\u002F22](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A649), [EU:C:2025:649](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A649), paragraph [68](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A649#point68)).","https:\u002F\u002Feur-lex.europa.eu\u002Flegal-content\u002FEN\u002FTXT\u002FHTML\u002F?uri=CELEX:62024CC0748","ecli-eu-c-2026-16",{"caseNumber":32},"62024CC0748",{"id":34,"ecli":35,"caseNumber":36,"courtId":5,"courtName":15,"decisionDate":16,"publicationDate":16,"fullText":37,"language":18,"source":19,"sourceUrl":38,"summary":15,"slug":39,"metadata":40},"267","ECLI:EU:C:2026:19","C-52\u002F25","OPINION OF ADVOCATE GENERAL\n\nBIONDI 1 Joined Cases C‑52\u002F25 and C‑53\u002F25 [Binanrier ( i )]\n\nRZ,\n\nGT (C‑52\u002F25)\n\nAX,\n\nUI (C‑53\u002F25)\n\nv\n\nRégion wallonne\n\n(Request for a preliminary ruling from the Cour d’appel de Mons  \n(Court of Appeal, Mons, Belgium)) \n\n(Reference for a preliminary ruling – State aid – Regulation (EU) No 702\u002F2014 – Block exemption for certain types of aid to the agriculture and forestry sectors – Aid to compensate for damage caused by an adverse climatic event that can be assimilated to a natural disaster – Reduction of the amount of aid in the event of failure to take out insurance – No possibility for the beneficiary of the aid to take out the required insurance)\n\n**I.** **Introduction**\n\n1\\. By its requests for a preliminary ruling in these two joined cases, the Cour d’appel de Mons (Court of Appeal, Mons, Belgium; ‘the referring court’) has referred three questions to the Court of Justice concerning the interpretation of Article 25 of Regulation (EU) No 702\u002F2014, ( 2 ) a provision concerning aid intended to compensate small and medium-sized enterprises (SMEs) active in primary agricultural production for damage caused by adverse climatic events that can be assimilated to natural disasters. Aid that complies with all the conditions set out in that article and in Chapter I of that regulation is exempt from the requirement to notify the European Commission laid down in Article 108(3) TFEU.\n\n2\\. More specifically, the three questions referred by the referring court concern paragraph 9 of Article 25 of Regulation No 702\u002F2014, which provides that compensation constituting such aid is to be ‘reduced by 50% unless it is granted to beneficiaries who have taken out insurance covering at least 50% of their average annual production or production-related income and the statistically most frequent climatic risks in the Member State or region concerned for which insurance coverage is provided’. ( 3 )\n\n3\\. The questions referred for a preliminary ruling arise in proceedings concerning claims for compensation brought by a number of farmers (RZ, GT, AX and UI; ‘the appellants in the main proceedings’) who suffered damage as a result of the drought that affected Belgium from August 2016 to June 2017. That drought has been recognised by the Belgian authorities as an agricultural disaster giving rise to entitlement to compensation. For the purpose of calculating the compensation, the Belgian authorities applied a national provision ( 4 ) implementing Article 25(9) of Regulation No 702\u002F2014, under which the amount of compensation is reduced by 50% ‘if the injured party has not taken out insurance covering at least 50% of his or her production against climatic risks’.\n\n4\\. In the present case, since the appellants in the main proceedings had not taken out any insurance against climatic risks, the competent Belgian authorities reduced their compensation by 50% in accordance with the abovementioned national provision.\n\n5\\. However, the appellants in the main proceedings lodged an appeal with the Belgian courts, arguing that they had been unable to take out insurance because, at the time the damage occurred, it was not possible in Belgium to insure their farms against certain climatic risks, such as drought.\n\n6\\. Following various proceedings, the Cour d’appel de Mons (Court of Appeal, Mons) was seised of the disputes referred to in the main proceedings. It emerges from the order for reference that, first, the risk of drought is not one of the most frequent climatic risks in Belgium, since the most frequent climatic risk in that Member State is hail; second, at the time of the events, there was no insurance against drought in Belgium, but there was insurance against hail; and, third, the appellants in the main proceedings asserted that their farms consist of permanent and temporary meadows and grassland used for feeding livestock, and that, as such, those meadows and that grassland are not insurable against the risk of hail.\n\n7\\. The referring court therefore decided to stay the proceedings and to refer three questions for a preliminary ruling concerning the interpretation of Article 25(9) of Regulation No 702\u002F2014.\n\n8\\. At the request of the Court of Justice, the present Opinion focuses on the third question referred for a preliminary ruling, in which the referring court asks: ‘Does [Article 25(9) of Regulation No 702\u002F2014] allow the Member State concerned not to reduce the aid intended to compensate farmers for damage caused by a climatic event considered to be a natural disaster, where the aid beneficiary demonstrates that the type of production in which he or she engages (in this case, permanent and temporary meadows, as well as grassland for feeding livestock, and not crops) is not insurable in the Member State concerned (in this case, Belgium) against the statistically most frequent climatic risks, up to an amount of at least 50% of the average annual production or production-related income of his or her farm?’\n\n**II.** **Analysis**\n\n9\\. By its third question, the referring court asks, in essence, whether Article 25(9) of Regulation No 702\u002F2014 must be interpreted as meaning that a Member State may be authorised not to reduce by 50% aid intended to compensate for damage caused by adverse climatic events that can be assimilated to natural disasters, where the beneficiary of the aid demonstrates that it was not possible to obtain insurance cover for the type of production of his or her farm against the statistically most frequent climatic risks, as that was not insurable in that Member State.\n\n10\\. To answer the third question referred for a preliminary ruling by the referring court, it is necessary to interpret Article 25(9) of Regulation No 702\u002F2014.\n\n11\\. According to settled case-law, for the purpose of interpreting a provision of EU law, it is necessary to consider not only its wording but also the context in which it occurs and the objectives pursued by the rules of which it is part. ( 5 )\n\n12\\. First of all, with regard to Article 25(9) of Regulation No 702\u002F2014, referred to in point 2 of the present Opinion, that provision stipulates that the aid in question is to be reduced by 50% unless the beneficiary has taken out an insurance policy that must meet two requirements.\n\n13\\. In the first place, that insurance policy must cover at least 50% of the average annual production or corresponding income, and, in the second place, it must cover the climatic risks that are statistically most frequent in the Member State or region concerned for which insurance cover is provided.\n\n14\\. With regard to the second requirement, it is clear from the wording of the provision in question (in most of the language versions, as noted in point 17 of the present Opinion) that insurance cover for the most frequent climatic risks must be ‘provided’. In my opinion, it follows, conversely, that if insurance cover is not provided for those most frequent climatic risks, the condition relating to the need to take out an insurance policy cannot apply and, consequently, the reduction in aid cannot apply either. Furthermore, such a condition appears to be reasonable: it seems illogical to require, as a condition for obtaining part of the compensation, that an insurance contract be entered into where cover for the relevant risk is not available.\n\n15\\. A literal analysis of the provision in question would therefore seem to lead to an interpretation whereby the reduction in compensation does not apply if it can be demonstrated that it was not possible to take out the insurance policy on which the non-application of that reduction is conditional, since the beneficiary is not provided with insurance cover for the statistically most frequent climatic risks.\n\n16\\. Two further observations are relevant in that regard.\n\n17\\. First, the different language versions of the provision in question do not all coincide perfectly. However, the Italian-language version of the provision from which I have derived the abovementioned interpretation is confirmed by the text of several other language versions, including the English-language version, which refers to ‘risks … for which insurance coverage is provided’, the German-language version, which mentions ‘ *Risiken … für die Versicherungsschutz gegeben ist* ’, and the Spanish-language version, which refers to ‘ *riesgos … para los que se proporciona cobertura de seguros* ’. Several other language versions of the provision are also along the same lines. ( 6 ) The French-language version is less clear on that point, but its wording is not incompatible with the interpretation of the provision in question considered in point 15 above. ( 7 )\n\n18\\. Second, that interpretation is confirmed in the Communication from the Commission on guidelines for State aid in the agricultural and forestry sectors and in rural areas. ( 8 )\n\n19\\. Point 363 of the 2014 Guidelines, concerning aid to compensate for damage to agricultural production or the means of agricultural production and to prevent damage, expressly states that ‘derogation from this condition is only possible if a Member State can convincingly show that, despite all reasonable efforts, affordable insurance covering the statistically most frequent climatic risks in the Member State or region concerned was not available at the time the damage occurred.’ ( 9 )\n\n20\\. It can therefore be inferred from that point in the abovementioned Guidelines that the only possible derogation from the condition of taking out an insurance policy in order to avoid a 50% reduction in compensation is the lack of availability, on the date on which the damage occurred, of insurance covering the risks in question, despite the injured party having made all reasonable efforts to obtain such a policy.\n\n21\\. Although a document such as the Guidelines is binding only on the Commission ( 10 ) and therefore does not necessarily bind either the Court of Justice or the national courts in their interpretation of the provision in question, it nevertheless constitutes an important source of interpretation. That is especially true in a case such as the present one, where the question of interpretation relates to a regulation issued by the Commission itself. In such circumstances, a document such as the Guidelines constitutes the interpretation of the act by the institution that adopted that act.\n\n22\\. The interpretation of Article 25(9) of Regulation No 702\u002F2014, set out in point 15 of the present Opinion, appears to be confirmed by both contextual and teleological analysis.\n\n23\\. In contextual terms, that provision must be placed within the general context of the Common Agricultural Policy (CAP), in which risk management has become a crucial element, with the aim of increasing the resilience of farms facing growing risks, particularly those linked to increasingly frequent catastrophic events related to climate change, while promoting the spread of practices that increase risk prevention and adaptation. Those objectives have been considered fundamental both in the current programming period 2023 to 2027, which envisages an integrated risk management system, ( 11 ) and in the previous period (2014 to 2022). ( 12 ) Adaptation to climate change and increasing the resilience of farms in that context are also considered key priorities within the CAP, in the light of which risk management must be considered. ( 13 )\n\n24\\. Regulation No 702\u002F2014 and the new Regulation 2022\u002F2472 fit within that framework and constitute an implementation of those policy objectives. Under those regulations, aid to compensate for losses caused by adverse climatic events that can be assimilated to natural disasters, as referred to in Article 25 of Regulation No 702\u002F2014, falls within the framework of ‘risk and crisis management aid’, along with aid for the costs of the prevention and eradication of animal diseases and plant pests, and aid for insurance premiums. ( 14 )\n\n25\\. In that context, from a teleological point of view, as highlighted by the parties who have submitted observations to the Court, ( 15 ) the specific purpose of Article 25(9) of Regulation No 702\u002F2014 is to encourage farmers to take out insurance against the climatic risks that cause most of the damage eligible for compensation, in order to reduce the financial burden on the community through agricultural aid granted to compensate for damage related to adverse climatic events.\n\n26\\. That purpose of the provision in question is apparent from recital 54 of Regulation No 702\u002F2014, from which it can be inferred that, since ‘primary agricultural production is exposed to particular natural, climatic and health risks and crises … good risk and crisis management is a key tool for a sustainable and competitive agricultural sector’. In that context, therefore, ‘State aid for making good losses caused by adverse climatic events that can be assimilated to natural disasters … should be limited to helping beneficiaries facing particular difficulties despite having undertaken reasonable efforts to minimise such risks’.\n\n27\\. In the same vein, the abovementioned Commission Guidelines ( 16 ) emphasise that ‘in order to further improve risk management, beneficiaries must be encouraged to take out insurance wherever possible’.\n\n28\\. With that objective in mind, in order for the reduction not to be applied, aid beneficiaries must therefore have made all reasonable efforts to limit the risks associated with damage resulting from the adverse climatic event. However, once that has been demonstrated – as stated in the Commission’s own Guidelines – an undertaking that cannot be covered by insurance for such losses should be granted the maximum aid intensity. ( 17 )\n\n29\\. An analysis of the purpose of Article 25(9) of Regulation No 702\u002F2014 confirms the interpretation that the reduction in compensation does not apply where it is demonstrated that it was not possible to take out the insurance policy in question because insurance cover was not available for the statistically most frequent climatic risks. However, it adds the further condition that it must also be demonstrated that the beneficiaries of the aid have made all reasonable efforts to take out such a policy to limit the risks associated with damage resulting from the adverse climatic event.\n\n30\\. The arguments raised by the Commission in its observations do not preclude this interpretation of Article 25(9) of Regulation No 702\u002F2014.\n\n31\\. Indeed, in the first place, as can be seen from points 12 to 21 of the present Opinion, this interpretation is consistent with the wording of the provision, which, although it does not refer – as asserted by the Commission – to the reasons for the absence of insurance, nevertheless expressly stipulates that insurance cover must be provided.\n\n32\\. In the second place, my interpretation of Article 25(9) of Regulation No 702\u002F2014 is not contrary to the settled case-law of the Court, according to which the provisions of a regulation on State aid providing for a block exemption and the conditions laid down therein must be interpreted strictly. ( 18 ) Indeed, such an interpretation does not attribute to the regulatory provision and the conditions laid down therein a broader scope than that which can be inferred from its literal wording. On the contrary, the proposed interpretation, in accordance with the wording of the provision itself, avoids a possible interpretation that, as noted in point 14 of the present Opinion, would lead to an unreasonable application of the conditions laid down therein.\n\n33\\. In the third place, that interpretation does not conflict with the need – referred to by the Commission – to ensure uniform interpretation of the regulation at issue in the various Member States. Indeed, it guarantees that the provision can be applied on the basis of objective criteria (a demonstration that cover is impossible and that the beneficiary has made all reasonable efforts), which are applicable in the same way in all Member States, even though the circumstances (such as whether it is possible to take out an insurance policy in a given case) might vary.\n\n34\\. In the fourth place, the Commission refers in its observations to the requirement for Member States to notify the Commission of any aid that is more favourable to farmers so as to take account of the fact that they are unable to insure themselves against the statistically most frequent climatic risks.\n\n35\\. On that point, I note that the Court has observed that Regulation No 702\u002F2014, which was adopted pursuant to Article 108(4) TFEU, provides, in Article 3, that, notwithstanding the general obligation to notify each measure intended to grant or alter ‘new aid’ within the meaning of Article 108(3) TFEU, which constitutes one of the fundamental features of the monitoring system in the field of State aid, a Member State may, under that regulation, rely on the exemption from that requirement if an aid measure which it adopted or an aid project which it plans to adopt fulfils the conditions laid down therein. ( 19 )\n\n36\\. Therefore, an interpretation of Article 25 of Regulation No 702\u002F2014 that, in the absence of notification by the Member State, would deny the compensation provided for therein to a beneficiary who fulfils the conditions referred to in that provision for the grant of aid would be contrary to the scheme of the regulation itself.\n\n37\\. On that point, I note that the condition relating to the conclusion of the insurance contract provided for in paragraph 9 of that article is not a condition relating to the grant of the aid but concerns the amount of the aid granted.\n\n38\\. In the same vein, an interpretation of that provision to the effect that, in the absence of notification by the Member State, a beneficiary who satisfies all the conditions under that provision cannot receive the full amount of the aid even if it is established by a court that insurance covering the most frequent climatic risks was not objectively available to that beneficiary at the time the damage occurred, would not be merely contrary to the wording of that provision, as follows from points 12 to 21 above. It would also be contrary to the scheme of Regulation No 702\u002F2014 inasmuch as it would deny full compensation to a beneficiary who meets the requirements laid down in Article 25(1) of that regulation for receiving the aid. Such an interpretation would also go beyond the objectives pursued by that regulation, referred to in points 25 to 29 above, and would be contrary to the principle of proportionality referred to in recital 8 of Regulation No 702\u002F2014.\n\n39\\. In conclusion, it follows from the foregoing considerations that Article 25(9) of Regulation No 702\u002F2014 must, in my view, be interpreted as meaning that the 50% reduction in compensation in respect of aid intended to make good damage caused by adverse climatic events that can be assimilated to natural disasters, laid down in that provision, does not apply where the beneficiary of the aid demonstrates that, despite having made all reasonable efforts to take out an insurance policy to limit the risks associated with damage resulting from the adverse climatic event, it was not possible to take out an insurance policy because no insurance cover was available for the type of production of his or her farm for the statistically most frequent climatic risks.\n\n40\\. In the present case, the appellants in the main proceedings have argued that although, in general, at the time the adverse event occurred in Belgium, it was possible to obtain insurance cover for the most frequent climatic risks, namely hail, such cover was not available for the specific type of production carried out by the appellants in the main proceedings.\n\n41\\. It is for the referring court to verify those factual circumstances and to determine that the appellants in the main proceedings made all reasonable efforts to take out an insurance policy to limit the risks associated with damage resulting from adverse climatic events and that the type of production of the appellants in the main proceedings was indeed not insurable in the Member State concerned against the statistically most frequent climatic risks.\n\n**III.** **Conclusion** 1 ( i ) The name of the present case is a fictitious name. It does not correspond to the real name of any party to the proceedings.\n\n42\\. In the light of all the foregoing considerations, I propose that the Court should answer the third question referred for a preliminary ruling by the Cour d’appel de Mons (Court of Appeal, Mons, Belgium) as follows: Article 25(9) of Commission Regulation (EU) No 702\u002F2014 of 25 June 2014 declaring certain categories of aid in the agricultural and forestry sectors and in rural areas compatible with the internal market in application of Articles 107 and 108 of the Treaty on the Functioning of the European Union must be interpreted as meaning that the 50% reduction in compensation in respect of aid intended to make good damage caused by adverse climatic events that can be assimilated to natural disasters, laid down in that provision, does not apply where the beneficiary of the aid demonstrates that, despite having made all reasonable efforts to take out an insurance policy to limit the risks associated with damage resulting from the adverse climatic event, it was not possible to take out an insurance policy because no insurance cover was available for the type of production of his or her farm for the statistically most frequent climatic risks. It is for the referring court to ascertain those specific factual circumstances.\n\n* * *\n\n( 2 ) Commission Regulation of 25 June 2014 declaring certain categories of aid in the agricultural and forestry sectors and in rural areas compatible with the internal market in application of Articles 107 and 108 of the Treaty on the Functioning of the European Union ([OJ 2014 L 193, p. 1](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Foj\u002FJOL_2014_193_R_TOC)). That regulation was in force until 31 December 2022 and was replaced, as of 1 January 2023, by Commission Regulation (EU) 2022\u002F2472 of 14 December 2022 declaring certain categories of aid in the agricultural and forestry sectors and in rural areas compatible with the internal market in application of Articles 107 and 108 [TFEU] ([OJ 2022 L 327, p. 1](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Foj\u002FJOL_2022_327_R_TOC)). 3 ( 4 ) Article 6 of the arrêté du Gouvernement wallon considérant comme une calamité agricole la sécheresse d'août 2016 à juin 2017, délimitant l’étendue géographique de cette calamité et déterminant l'indemnisation des dommages (Order of the Walloon Government considering the drought from August 2016 to June 2017 to be an agricultural disaster, delimiting the geographical extent of that disaster and determining compensation for the damage).\n\n( 5 ) See judgment of 19 September 2024, Agrarmarkt Austria ([C‑350\u002F23](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2024%3A771), [EU:C:2024:771](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2024%3A771), paragraph [57](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2024%3A771#point57) and the case-law cited). 6 ( 7 ) The French-language version of Article 25(10) of Regulation 2022\u002F2472 has undergone a change in its last sentence compared to the text of Article 25(9) of Regulation No 702\u002F2014. Such a change cannot be found in the other language versions. Whereas the version of the text in Regulation No 702\u002F2014 referred to the ‘ *risques climatiques statistiquement les plus fréquents dans l’État membre ou la région concernés* , couverte par une assurance’, the new version refers to the ‘ *risques climatiques statistiquement les plus fréquents dans l’État membre ou la région concernés* que couvre l’assurance’. That amendment is in line with my interpretation of the provision.\n\n( 8 ) Communication from the Commission, European Union Guidelines for State aid in the agricultural and forestry sectors and in rural areas 2014 to 2020 ([OJ 2014 C 204, p. 1](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Foj\u002FJOC_2014_204_R_TOC); ‘the 2014 Guidelines’); see also Communication from the Commission, Guidelines for State aid in the agricultural and forestry sectors and in rural areas ([OJ 2022 C 485, p. 1](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Foj\u002FJOC_2022_485_R_TOC); ‘the 2022 Guidelines’). 9 ( 10 ) See judgment of 8 March 2016, Greece v Commission ([C‑431\u002F14 P](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2016%3A145), [EU:C:2016:145](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2016%3A145), paragraphs [69](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2016%3A145#point69) and [70](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2016%3A145#point70)). 11 ( 12 ) See, inter alia, Regulation (EU) No 1305\u002F2013 of the European Parliament and of the Council of 17 December 2013 on support for rural development by the European Agricultural Fund for Rural Development (EAFRD) and repealing Council Regulation (EC) No 1698\u002F2005 ([OJ 2013 L 347, p. 487](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Foj\u002FJOL_2013_347_R_TOC)), in particular Article 36, entitled ‘Risk management’, and Annex VI.\n\n( 13 ) On that point, see recitals 29 and 30 of Regulation (EU) 2021\u002F2115 of the European Parliament and of the Council of 2 December 2021 establishing rules on support for strategic plans to be drawn up by Member States under the common agricultural policy (CAP Strategic Plans) and financed by the European Agricultural Guarantee Fund (EAGF) and by the European Agricultural Fund for Rural Development (EAFRD) and repealing Regulations (EU) No 1305\u002F2013 and (EU) No 1307\u002F2013 ([OJ 2021 L 435, p. 1](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Foj\u002FJOL_2021_435_R_TOC)). 14 ( 15 ) Namely, the Commission and the Kingdom of Belgium. 16 ( 17 ) The points of the Guidelines referred to in the previous footnote state that, ‘regarding aid to compensate for losses caused by adverse climatic events which can be assimilated to a natural disaster, in order to avoid the risk of distorting competition, aid at the maximum aid intensity should be granted only to an undertaking that cannot be covered for such losses by insurance.’\n\n( 18 ) See judgment of 20 May 2021, Azienda Sanitaria Provinciale di Catania ([C‑128\u002F19](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2021%3A401), [EU:C:2021:401](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2021%3A401), paragraph [42](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2021%3A401#point42) and the case-law cited).\n\n( 19 ) See judgment of 20 May 2021, Azienda Sanitaria Provinciale di Catania ([C‑128\u002F19](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2021%3A401), [EU:C:2021:401](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2021%3A401), paragraph [42](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2021%3A401#point42)), and, by analogy, judgment of 5 March 2019, Eesti Pagar ([C‑349\u002F17](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A172), [EU:C:2019:172](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A172), paragraph [59](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2019%3A172#point59) and the case-law cited).","https:\u002F\u002Feur-lex.europa.eu\u002Flegal-content\u002FEN\u002FTXT\u002FHTML\u002F?uri=CELEX:62025CC0052","ecli-eu-c-2026-19",{"caseNumber":41},"62025CC0052",{"id":43,"ecli":44,"caseNumber":45,"courtId":5,"courtName":15,"decisionDate":16,"publicationDate":16,"fullText":46,"language":18,"source":19,"sourceUrl":47,"summary":15,"slug":48,"metadata":49},"264","ECLI:EU:C:2026:21","C-603\u002F24","OPINION OF ADVOCATE GENERAL\n\nKOKOTT 1 **Case C‑603\u002F24**\n\nStellantis Portugal, S.A.\n\nv\n\nAutoridade Tributária e Aduaneira\n\n(Request for a preliminary ruling from the Supremo Tribunal Administrativo (Supreme Administrative Court, Portugal)) \n\n(Reference for a preliminary ruling – Common system of value added tax – Directive 2006\u002F112\u002FEC and Directive 77\u002F388\u002FEEC – Supplies between affiliated undertakings – Adjustment of the intra-group sale price based on the warranty costs and operating costs incurred by the purchaser – Adjustment of the price of a supply as a separate supply of services effected for consideration – Relevance, for the purposes of value added tax law, of adjustments of a transfer price)\n\n**I.** **Introduction**\n\n1\\. Income tax law has many different facets. One of them is transfer pricing, which plays an important role between affiliated undertakings in the appropriate allocation of profits, taking into account the arm’s length principle. Unfortunately, determining the correct transfer price is less a matter of law and more a ‘matter of faith’. In that respect, there is no such thing as one correct transfer price, but rather various methods for determining it and thus a whole range of ‘correct’ transfer prices.\n\n2\\. That is probably the reason why, to date, it has always been attempted to keep VAT law, which generally does not involve such ‘matters of faith’, separate from it. The present request for a preliminary ruling shows that that may no longer be possible in the future. ( 2 )\n\n3\\. Unlike the case in *Arcomet Towercranes* , ( 3 ) already decided by the Court, the present case concerns a subsequent adjustment of the price of a supply intended to implement an intra-group allocation of profits. In that regard, the proceedings allow the Court to review the scope of adjustments of transfer pricing made for reasons of income tax law and the consequences of those adjustments under VAT law, a field that has received little attention to date, and, if possible, to give an answer that provides guidance beyond this specific case.\n\n**II.** **Legal framework**\n\n**A.** **European Union law**\n\n4\\. The legal framework for 2006, the year at issue in the present case, is constituted by the Sixth VAT Directive, ( 4 ) which did not yet provide for the transfer of the tax liability in the case of supplies of services to a taxable person established abroad. However, the provisions to be interpreted are identical to the corresponding provisions of Directive 2006\u002F112\u002FEC on the common system of value added tax ( 5 ) (‘the VAT Directive’).\n\n5\\. Article 2(1)(a) and (c) of the VAT Directive (corresponding to point 1 of Article 2 of the Sixth VAT Directive) provides: ‘The following transactions shall be subject to VAT: (a) the supply of goods for consideration within the territory of a Member State by a taxable person acting as such; … (c) the supply of services for consideration within the territory of a Member State by a taxable person acting as such’.\n\n6\\. Article 73 of the VAT Directive (corresponding to Article 11A(1)(a) of the Sixth VAT Directive) defines the taxable amount, and provides: ‘In respect of the supply of goods or services, other than as referred to in Articles 74 to 77, the taxable amount shall include everything which constitutes consideration obtained or to be obtained by the supplier, in return for the supply, from the customer or a third party, including subsidies directly linked to the price of the supply.’\n\n7\\. Article 90(1) of the VAT Directive (corresponding to Article 11C(1) of the Sixth VAT Directive) provides: ‘1. In the case of cancellation, refusal or total or partial non-payment, or where the price is reduced after the supply takes place, the taxable amount shall be reduced accordingly under conditions which shall be determined by the Member States.’\n\n**B.** **Portuguese law**\n\n8\\. It is true that the exact wording of the corresponding provisions of Portuguese law has not been communicated to the Court. However, it is to be assumed that the VAT Directive was transposed accordingly in Portugal by the Código do IVA (‘the VAT Code’).\n\n**III.** **The facts of the case**\n\n9\\. Stellantis Portugal, S.A. (‘the applicant’) is a company operating in the motor vehicle trade and, in 2006, the year at issue, was part (more precisely: its legal predecessor was part) of the General Motors Group, which manufactures and distributes vehicles and parts and accessories.\n\n10\\. That group was made up of original equipment manufacturers (‘OEMs’) which manufacture the products and\u002For supply them to the national sales companies (‘NSCs’) or national sales organisations (‘NSOs’), which distribute those products in a market that has been predefined in geographical terms.\n\n11\\. In the production chain, the applicant operated as an NSC\u002FNSO, purchasing vehicles from the European manufacturers (OEMs) of the General Motors Group. The vehicles purchased were resold to independent Portuguese dealers, which, in turn, sold those vehicles to the final customers.\n\n12\\. In the event of manufacturing defects, the final customer went to the dealer to have them repaired by the dealer in its own facilities. The Portuguese dealers then charged the applicant for the costs they incurred in repairing the vehicles, charging the relevant amount of VAT in respect of that supply of services. Such vehicle repairs could be carried out in the event of anomalies resulting from defects in the manufacturing process for the motor vehicles and the parts and accessories, anomalies related to the vehicles’ warranties and procedures related to roadside assistance.\n\n13\\. The applicant informed the European manufacturers of the General Motors Group (the respective OEMs) of the costs in distributing the motor vehicles and the parts and accessories manufactured by the group. Those costs included the vehicle repair costs referred to above and the applicant’s operating costs, in particular staff, electricity and marketing costs.\n\n14\\. Depending on the costs declared, an adjustment was then made in respect of the price of the vehicles sold to the applicant by the European manufacturers of the General Motors Group. That adjustment was made on the basis of a contract concluded between the companies of the General Motors Group to determine the prices of the vehicles transferred.\n\n15\\. Under that contract, ‘… General Motors Corporation (GM) and its GME subsidiaries are to apply a market-based pricing process to determine the transfer prices [of products] sold by the manufacturing units of GM (the OEMs) to the distribution units. The prices … shall be determined by deducting from the external sales prices the total amount of the relevant distribution costs and the amount of the operating profit aimed to be made …. In accordance with the procedure followed by GME, the initial transfer prices shall be determined in each reference period applying a gross margin discount to the expected external sale prices. At the end of each period, the transfer price shall be adjusted according to the operating profit previously agreed for the distribution unit, such that the actual amount of the operating profit for that unit is the same as the operating profit target amount ….’\n\n16\\. That contract also provided as follows: ‘… the initial transfer prices shall be adjusted at the end of each reference period in order to ensure that the actual financial results of the distribution unit tally with the agreed operating profit. … That adjustment shall be made … in the relevant period, with the relevant increase or reduction being recorded in the accounts of the selling unit (OEM).’\n\n17\\. The adjustment of the sale price of the vehicles was documented by means of a credit or debit note issued to the applicant by the European manufacturers of the General Motors Group.\n\n18\\. The applicant was the subject of an inspection, at its offices, carried out by the Serviços de Inspeção Tributária da Direcção de Finanças de Lisboa (Tax Inspection Services of the Lisbon Finance Directorate, Portugal). The inspection was general in scope and concerned accounting and tax matters relating to the 2006 financial year. The final inspection report was drawn up on 10 December 2009. In it, the tax authority found that the OEMs are responsible for repairs, given that they are the entities responsible for the manufacture and\u002For assembly of vehicles and parts and accessories. So, initially, the applicant bears all of the after-sales costs and then passes on the amount of those costs to the relevant OEMs.\n\n19\\. In that regard, the tax authority proceeds on the basis that the applicant supplied OEMs with services, within the national territory, which are subject to VAT under Article 1(1)(a) of the VAT Code. The costs of repairs, warranties and roadside assistance indicated by the applicant in that regard amounted to a total of EUR 6 352 516.47, which gives rise to value added tax in the amount of EUR 1 334 028.47. Thus, on 23 December 2009, the applicant was issued with supplementary VAT assessments for 2006 totalling EUR 1 504 215.49.\n\n20\\. It is apparent that those assessments were challenged by the applicant, but that they were upheld by the judgment of the Tribunal Central Administrativo Sul (Southern Central Administrative Court, Portugal). The applicant lodged an extraordinary appeal (*recurso de revista*) against that judgment on 16 November 2023.\n\n**IV.** **The preliminary ruling procedure**\n\n21\\. The Supremo Tribunal Administrativo (Supreme Administrative Court, Portugal) stayed the proceedings and referred the following question to the Court of Justice for a preliminary ruling: ‘Must Article 2 of the Sixth VAT Directive …, as worded in the version in force at the time of the facts, be interpreted as meaning that the concept of the supply of services effected for consideration contained in that provision includes an adjustment of the sale price of vehicles which is duly provided for and determined in a contract concluded between the parties, in order to achieve a minimum profit margin, and which is documented by means of a credit or debit note issued to the applicant\u002Fappellant by the European manufacturers of the General Motors Group?’\n\n22\\. In the proceedings before the Court, Stellantis Portugal, the Portuguese Republic and the European Commission submitted written observations. The Court decided not to hold a hearing in accordance with Article 76(2) of the Rules of Procedure of the Court of Justice.\n\n**V.** **Legal assessment**\n\n**A.** **Understanding the question referred for a preliminary ruling**\n\n23\\. The question posed by the referring court is, at first glance, surprising and can be understood, at second glance, only on the basis of the position taken by the Portuguese tax authority in the course of the proceedings.\n\n*1.*  *Service at both positive and negative prices?*\n\n24\\. On the basis of the intra-group rules, the price of the products sold to the applicant (cars and parts) aims to ensure that they achieve only or at least a certain result (that is relevant for the purposes of income tax law) and that they are taxed in Portugal. Therefore, the prices of the products sold are adjusted periodically by the seller. According to the request for a preliminary ruling, that adjustment is made in both directions. The consequence is that, in the event of a price increase, the purchaser subsequently still has to pay something to the seller, whereas, in the event of a price reduction, the purchaser has made an overpayment and is therefore reimbursed. For that reason, the seller issues corresponding credit or debit notes.\n\n25\\. During the year in question, the price was reduced and a reimbursement was made to the purchaser. The tax authority now apparently wishes to treat that reimbursement as consideration for a service supplied to the seller by the purchaser in Portugal. That is why the referring court asks whether the reduction in the purchase price resulting from a supply made by the seller can constitute consideration for a service supplied by the purchaser to the seller.\n\n26\\. I fail to see what service is supplied by a purchaser who, for whatever reason, subsequently pays a lower or higher purchase price. ( 6 ) Nor does Portugal explain this in any more detail in its observations. Since, in the present case, the purchase price could also have increased (for example, if fewer warranty cases occur or if the purchaser’s\u002Fapplicant’s own operating costs decrease), the assumption of a separate supply of services by the purchaser becomes even more daring. According to the Portuguese tax authority, the purchaser would have supplied the seller with a service, and what is more, according to the logic of those authorities, the purchaser would even have paid for that service itself.\n\n27\\. In the light of the foregoing, that would be a supply of services in return for negative consideration, which is not provided for by VAT law and which has no consequences, for the purposes of VAT law, for the ‘supplier of services’. That also makes sense because VAT, as a general tax on consumption, is intended to tax the expense of a consumer good. ( 7 ) However, if the recipient of a service does not have any expense but, on the contrary, receives money, the reason for the VAT burden does not apply. Consequently, in so far as the purchaser of goods, depending on the circumstances provided for in the contract, must subsequently pay more or less for a supply, that quite clearly militates against the existence of a supply of services by the purchaser and, on the contrary, militates in favour of an adjustment of the consideration for the supply already made.\n\n*2.*  *Bearing of costs as a service?*\n\n28\\. The Portuguese tax authority appears to consider that the bearing of the warranty costs of ‘defective’ vehicles purchased and resold by the applicant constitutes a service supplied to its seller. Ultimately, the defective cars are repaired by the last seller (the car dealer at the end of the supply chain) at its own expense, but those costs are then charged to the applicant. The latter bears the warranty costs, but those costs are, to a certain extent, already taken into account in its purchase price. It is only when they are higher than expected (and consequently the result provided for under income tax law is no longer achieved) that the purchase price of the goods is reduced.\n\n29\\. However, it is quite clear from Article 2(1) of the VAT Directive that merely bearing costs cannot constitute a supply of services. VAT law does not tax payments as such. A payment (and thus also a refund of costs incurred) is relevant for the purposes of VAT law only when it is made ‘in return for’ a supply of services. That is why Article 2(1) of the VAT Directive states that a taxable transaction exists only where a supply of services is effected *for* consideration. It is also apparent from Article 73 of the VAT Directive that only payments obtained by the supplier *in return for* the supply are relevant.\n\n30\\. The fact that the applicant initially bears the cost of repairing the products it sells and that those costs (in addition to other costs – as the Commission rightly points out) are taken into account in the pricing of the product purchased does not therefore mean that it thus supplies any service to a third party. Rather, it is fulfilling its own obligations as a seller to its purchaser, which is in turn faced with the warranty claims of its customers (the final customers).\n\n*3.*  *The relevance, for the purposes of VAT law, of a distribution of profits?*\n\n31\\. Ultimately, the purchase price adjustment serves here ‘only’ to allocate the profits of one intra-group company to another. That is, in principle, irrelevant for the purposes of VAT law. Profit distributions are not made for an economic activity (supply of goods or services), but only for participation in another’s economic activity. If even a dominant holding company is claimed not to be a taxable person in spite of it receiving millions in dividends, ( 8 ) profit adjustments can also hardly be relevant in the context of a subsequent adjustment of transfer pricing. The adjustment of profits for the purposes of income tax law is in principle alien to VAT law and therefore does not constitute consideration for a supply of services, regardless of who supplies them.\n\n*4.*  *Interim summary*\n\n32\\. The adjustment of the consideration for a supply, expressly referred to in the question, can therefore never itself constitute a supply of services.\n\n33\\. If the question is taken literally, it can therefore be answered very quickly. The mere adjustment (upwards or downwards) of a sale price for a supply in principle never itself constitutes a supply of services within the meaning of Article 2(1) of the VAT Directive.\n\n**B.** **The actual question referred for a preliminary ruling: Consequences of a price adjustment made for reasons of income tax law**\n\n34\\. If the request for a preliminary ruling is examined in more detail, however, it raises the dogmatically interesting question of the consequences, under VAT law, of an adjustment of transfer pricing made for reasons of income tax law. On close inspection, that question has even been referred to the Court in another case (the case in *Arcomet Towercranes* ( 9 )) although it was not answered exhaustively in that case.\n\n35\\. In that judgment, the Court reformulated the question, which had been appropriately referred and which was not straightforward to answer, making it rather straightforward to answer. It is doubtful whether that rather straightforward answer was also a useful answer. I will therefore take the opportunity here and reformulate a question that is rather straightforward to answer so that the underlying problem becomes clear and can be resolved by the Court. If the Court does not resolve the problem here, the resolution of the problem would only be delayed. The uncertainties already created in practice by the judgment referred to above would then continue to exist.\n\n36\\. In essence, the referring court seeks to ascertain how adjustments to a sale price agreed between two intra-group companies (also referred to as the transfer price for the purposes of income tax law) are to be treated for the purposes of VAT law where that price adjustment serves primarily to allocate the profit between the two affiliated undertakings in an appropriate manner (within the meaning of income tax law).\n\n37\\. As a general rule, the aim of contractually agreed profit allocation is to prevent, for the purposes of income tax law, the tax authority from subsequently adjusting transfer prices because the arm’s length principle may not have been observed, so that one company (usually that of a low-tax jurisdiction) generated, from the perspective of the tax authority (usually that of a high-tax jurisdiction), excessive profits that were actually generated and taxable in the other jurisdiction (in the example, the high-tax jurisdiction).\n\n38\\. Advocate General Richard de la Tour, who, unlike the Court, addressed the difficult question in the case referred to above, agrees that, for the purposes of VAT law, rules or criteria relating to income tax cannot be decisive. ( 10 ) What matters is whether the conditions of a transaction that is taxable (and liable to tax) (Article 2(1) of the VAT Directive) are satisfied where a transfer price is subsequently adjusted.\n\n39\\. Unlike Advocate General Richard de la Tour, ( 11 ) however, I consider it to be entirely possible and even necessary to give a principled answer to the question of the consequences under VAT law of an adjustment of transfer pricing made for reasons of income tax law. Despite the complex reality in a systematic area of law such as VAT law, this does not necessarily have to be decided afresh, on a case-by-case basis, each time.\n\n40\\. Indeed, there are certain situations in which an adjustment of transfer pricing may, in some circumstances, have consequences for the purposes of VAT law, and others in which it does not. Those situations can also be distinguished from each other, without being exhaustive.\n\n**C.** **The reformulated question**\n\n41\\. In that regard, the reformulated question and the one that actually requires an answer is as follows: ‘Must an adjustment of the consideration for an intra-group supply that is made for income tax reasons be regarded as a significant change in the taxable amount for the purposes of VAT law in accordance with Article 90 of the VAT Directive or as the taxable amount for a separate service supplied by the person who benefits from the change in the transfer price, or is it irrelevant for the purposes of VAT law as a mere adjustment of profits between two intra-group undertakings?’\n\n42\\. A number of situations are relevant in that regard. Accordingly, it is conceivable for the taxable person to make separate services or supplies with a view to generating input and output, which thus influences the profits as desired. The Court appears to have assumed such a situation in the case in *Arcomet Towercranes*. ( 12 ) However, it would also be possible for the parties to invoice something only on paper in order to steer profits accordingly. These are usually agreements that allow adjustments in both directions, whereby a supply of services, at least in one direction, is difficult to imagine (see, in that regard, section C.1.).\n\n43\\. It is also conceivable, however, for the tax authority (for example, ten years later in the context of a tax audit) to make an adjustment of transfer pricing with a view to an appropriate allocation of profits, which – usually without any real payment flows – ultimately leads to a (fictitious) attribution of profits on one side (and, in fact, a fictitious increase in expenses and thus a decrease in the profits on the other side) (see section C.2.). However, it is also conceivable for the existing purchase price subsequently to be adjusted on account of certain variables which were still open when the initial purchase price was set (see section C.3.).\n\n*1.*  *Separate supply of service for the creation of input and output vs. fictitious invoicing of fictitious supplies intended to correct profits*\n\n44\\. Under Article 2(1)(c) of the VAT Directive, ‘the supply of services for consideration within the territory of a Member State by a taxable person acting as such’ is subject to VAT.\n\n45\\. According to settled case-law, a supply of services is effected ‘for consideration’ within the meaning of Article 2(1)(c) of the VAT Directive and is therefore subject to VAT only if there is a legal relationship between the provider of the service and the recipient pursuant to which there is reciprocal performance, the remuneration received by the provider of the service constituting the actual consideration for the service supplied to the recipient. ( 13 ) That criterion relating to the existence of a legal relationship – as is also rightly apparent from the recent case-law of the Court of Justice – is to be given a broad meaning. ( 14 )\n\n46\\. Where the service to be supplied and the consideration to be paid for it arise from an underlying contract, such a legal relationship can, in principle, be assumed to exist. Moreover, this appears to have been the correct starting point for the Court ( 15 ) and the Advocate General ( 16 ) in the case in *Arcomet Towercranes*. If consideration has actually been agreed for a supply of services, provided that a corresponding positive result is achieved or not achieved, there is (at the time when the relevant condition is fulfilled) a supply of services for consideration that is relevant for the purposes of VAT law.\n\n47\\. In the present case, there is not even a contract under which a service supplied by the purchaser (the applicant) to the seller (OEM) could somehow be justified. First, the applicant purchases only the cars and resells them. That resale also entails bearing the warranty costs. Second, the tax authority’s assumption ‑ that the applicant supplies services in the form of the distribution and management of the warranty in favour of its seller and therefore in return for the consideration resulting from the price adjustment ‑ would make no sense if the applicant’s profit exceeded the prescribed range. In the event of lower distribution costs or lower warranty costs, the applicant would still have to pay the seller something for its own supply of services. That is far from the reality.\n\n48\\. If something of that nature had been contractually agreed, that would rather militate in favour of a fictitious supply of services, so that such a contract would rightly ( 17 ) be irrelevant to the taxation of the expense of a consumer good under VAT law. However, in the present case, it was not even the subject of a contractual agreement, but a supply of services is merely alleged by the tax authority.\n\n*2.*  *Subsequent adjustment of the transfer price by the tax authority for the purposes of profit adjustment*\n\n49\\. However, the situation is different where an adjustment of the transfer price, which has been ordered by the tax authority (and is thus unilateral), is made with a view to an appropriate allocation of profits between companies (usually between different States).\n\n50\\. So long as that occurs only in connection with the taxation of company X’s income in Member State A, it is a pure fiction. For income tax reasons, the profit is increased accordingly. Whether company Y’s profit is reduced in State B (which would logically be the case) is an entirely different question. Indeed, given that the other State may have other views about the correct transfer price and that, as I explained above, there is no such thing as one correct transfer price, that is not necessarily the case. That can also be described as a dilemma for the purposes of income tax law, for which, inter alia, dispute resolution procedures ( 18 ) have been developed.\n\n51\\. By contrast, VAT law is based on the principle that an adjustment of the price (consideration) always has the same effect on both sides. Thus, the supplier’s tax liability is reduced or increased (Articles 90 and 73 of the VAT Directive), while the recipient’s input tax deduction increases or decreases proportionally (Article 186 of the VAT Directive). In that regard – this also distinguishes VAT law from income tax law – it is not the objective value of a supply which is decisive, but the value taken as the basis by the parties, as the Court only recently reiterated. ( 19 )\n\n52\\. For that reason, Working paper No 923 of the VAT Committee also quite rightly states: ‘There is a tension between the transfer pricing rules set out for the purposes of direct taxation[,] which, based on the arm’s length principle[,] seek to arrive at the arm’s length valuation of a transaction (i.e. the open market value), and the VAT rules, generally based on the existence of a supply for consideration, where consideration is seen as a subjective value (i.e. the price actually paid)’. ( 20 )\n\n53\\. The tax authority’s view of the correct transfer pricing may therefore have consequences under income tax law; it does not, in principle, have any consequences under VAT law. In the present case, the parties’ views as to the amount of the consideration remain decisive.\n\n54\\. The VAT Expert Group’s entirely accurate statements also refer to this. According to those statements, such transfer pricing adjustments are outside the scope of the VAT Directive. ( 21 ) The VAT Expert Group states, however, that a distinction should be drawn between cases where the adjustment is provided for by contract and that, if there is an adjustment for previous supplies, that must lead to an adjustment of the taxable amount. ( 22 )\n\n55\\. On the other hand, if, for reasons of income tax law, *the tax authority* unilaterally (and ultimately fictitiously) increases (or decreases) the profit subsequently, that does not constitute any change in the consideration agreed between the undertakings concerned. Therefore, in principle, that does not alter the services previously supplied in return for the agreed consideration (see Article 73 of the VAT Directive) and thus also does not alter the amount of the VAT due and already paid. However, that does not appear to be the situation in the present case.\n\n*3.*  *Subsequent adjustment of an undetermined but determinable (variable) price*\n\n56\\. The present case concerns, rather, the subsequent adjustment of a price for the vehicles supplied, which, at the time of the agreement, was yet to be determined and was therefore designated by a reference price, but which could vary upwards or downwards depending on changes in various parameters. In that regard, it is entirely possible to speak of an undetermined but determinable (variable) price. According to the facts presented by the referring court, the decisive parameters are the costs borne by the purchaser (the applicant) (the applicant’s own distribution costs and the warranty costs borne by it) in connection with the distribution of the vehicles produced by the General Motors Group and purchased by the applicant.\n\n57\\. At the end of each period, the transfer price was adjusted according to the operating profit previously agreed for the distribution unit (in this case, the applicant), such that the actual amount of the applicant’s operating profit was the same as the target amount of operating profit.\n\n58\\. That method of calculating the sale price between the OEMs and the applicant ensured that the applicant retained the profit that had been allocated to it within the group, but no more or less. That profit allocated depends on the variables (its own distribution costs and the warranty costs borne by it), which were calculated subsequently for each asset purchased. Depending on the evolution of the variables, the purchase price increased or decreased subsequently.\n\n59\\. For such situations, there are corresponding provisions in the VAT Directive governing the consequences under VAT law. There is therefore no need to construct fictitious supplies of services beyond the supplies made, particularly where those could lead to a negative price.\n\n60\\. Article 90 of the VAT Directive expressly states that, where the price is reduced after the supply takes place, the taxable amount is to be reduced accordingly under conditions which are to be determined by the Member States. In this case, that would be the taxable amount of the supply to the applicant. At the same time, Article 73 of the VAT Directive covers subsequent price increases because, under that provision, the taxable amount of the seller’s supply is ultimately all it receives for its supply. That naturally includes subsequent increases of the purchase price.\n\n61\\. Those two provisions clearly show that adjustments to the actual price of a supply change the taxable amount of that supply. When that is the case, such a change in the taxable amount cannot at the same time constitute a supply of services. A subsequent price adjustment therefore has an impact only on the taxable amount of the underlying transaction.\n\n62\\. In other situations, the Court has already had to consider the effects of subsequent price changes and has always done so without assuming the existence of a (fictitious) service ‘by adjusting the contract’. Thus, comparable questions arise in customs law when the customs value is adjusted subsequently. While in the case in *Hamamatsu* ( 23 ) the Court held that a general adjustment of transfer prices without any specific reference to adjustments concerning the individual goods imported constituted an unforeseeable change with no effects under customs law, in the case in *Tauritus* ( 24 ) it considered the *contractually agreed subsequent change of an undetermined but determinable (variable) purchase price* of individual goods to be significant under customs law. That corresponds to the distinction proposed here according to the arrangements for adjusting a transfer price.\n\n63\\. In my view, the Court’s ruling in the case in *Arcomet Towercranes* ( 25 ) does not preclude that conclusion. First, that case did not concern the adjustment of a transfer price, but the existence of a supply of services for consideration if the profit was too high or too low. Irrespective of whether the result is actually convincing in that specific case or whether fictitious services were in fact invoiced, the Court ultimately held only that there is a supply of services for consideration where the parties have contractually agreed on a supply of services for consideration, presumably on the assumption that the service was actually supplied.\n\n**VI.** **Conclusion** 1 ( 2 ) See also, on this topic, Echevarria, G., *The interplay between transfer pricing and Value Added tax: Recent Case Law* , in EC Tax Review 2025\u002F5, p. 201 (p. 207 et seq.).\n\n64\\. I therefore propose that the question referred for a preliminary ruling by the Supremo Tribunal Administrativo (Supreme Administrative Court, Portugal) should be answered as follows: Article 2(1)(c) and Articles 73 and 90 of Council Directive 2006\u002F112\u002FEC of 28 November 2006 on the common system of value added tax must be interpreted as meaning that the relevance, for the purposes of VAT law, of an adjustment of profits made for reasons of income tax law depends on what it relates to and how it is made. Where the adjustment of profits is made by means of separate supplies of services for consideration (creation of input and output) and there are not only fictitious supplies of services, those separate supplies of services for consideration constitute taxable transactions for the purposes of Article 2(1)(c) of Directive 2006\u002F112. Where the adjustment of profits is made unilaterally and subsequently by the tax authority solely for the purposes of an appropriate allocation of profits between two tax-levying States, that is not, in principle, relevant for the purposes of VAT law. On the other hand, where, as in the present case, the adjustment of profits is made by means of a sale price which has been provided for precisely for that purpose and agreed to be variable and which relates to a specific supply of goods, that constitutes a reduction in the taxable amount under Article 90 of Directive 2006\u002F112 or a further part of the taxable amount under Article 73 thereof in respect of the supply made. Since the change in the taxable amount of a supply relates solely to the consideration, it cannot itself constitute a ‘supply of services for consideration’ within the meaning of Article 2(1)(c) of Directive 2006\u002F112.\n\n* * *\n\n( 3 ) Judgment of 4 September 2025, Arcomet Towercranes ([C‑726\u002F23](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A646), [EU:C:2025:646](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A646)).\n\n( 4 ) Council Directive 77\u002F388\u002FEEC of 17 May 1977 on the harmonisation of the laws of the Member States relating to turnover taxes – Common system of value added tax: uniform basis of assessment ([OJ 1977 L 145, p. 1](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Foj\u002FJOL_1977_145_R_TOC)).\n\n( 5 ) Council Directive of 28 November 2006 ([OJ 2006 L 347, p. 1](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Foj\u002FJOL_2006_347_R_TOC)). 6 ( 7 ) See my Opinion in Zlakov ([C‑744\u002F23](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A332), [EU:C:2025:332](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A332), point [39](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A332#point39) et seq.) for further details.\n\n( 8 ) At least according to the Court of Justice – see fundamental judgment of 20 June 1991, Polysar Investments Netherlands ([C‑60\u002F90](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A1991%3A268), [EU:C:1991:268](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A1991%3A268), paragraph [13](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A1991%3A268#point13)); see also judgments of 8 September 2022, Finanzamt R(Deduction of VAT linked to a shareholder contribution) ([C‑98\u002F21](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2022%3A645), [EU:C:2022:645](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2022%3A645), paragraph [41](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2022%3A645#point41) et seq.); of 8 November 2018, C&D Foods Acquisition ([C‑502\u002F17](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2018%3A888), [EU:C:2018:888](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2018%3A888), paragraph [30](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2018%3A888#point30) et seq.); and of 17 October 2018, Ryanair ([C‑249\u002F17](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2018%3A834), [EU:C:2018:834](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2018%3A834), paragraph [16](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2018%3A834#point16) et seq.). However, that is not entirely convincing and is therefore not without controversy – see my Opinion in Högkullen ([C‑808\u002F23](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A149), [EU:C:2025:149](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A149), point [32](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A149#point32) et seq.).\n\n( 9 ) Judgment of 4 September 2025, Arcomet Towercranes ([C‑726\u002F23](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A646), [EU:C:2025:646](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A646)).\n\n( 10 ) Opinion in Arcomet Towercranes ([C‑726\u002F23](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A244), [EU:C:2025:244](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A244), point [38](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A244#point38)).\n\n( 11 ) Opinion in Arcomet Towercranes ([C‑726\u002F23](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A244), [EU:C:2025:244](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A244), point [32](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A244#point32)).\n\n( 12 ) Judgment of 4 September 2025, Arcomet Towercranes ([C‑726\u002F23](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A646), [EU:C:2025:646](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A646)).\n\n( 13 ) Judgments of 4 July 2024, Credidam ([C‑179\u002F23](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2024%3A571), [EU:C:2024:571](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2024%3A571), paragraph [36](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2024%3A571#point36)); of 15 April 2021, Administration de l’Enregistrement, des Domaines et de la TVA ([C‑846\u002F19](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2021%3A277), [EU:C:2021:277](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2021%3A277), paragraph [36](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2021%3A277#point36)); and of 3 March 1994, Tolsma ([C‑16\u002F93](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A1994%3A80), [EU:C:1994:80](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A1994%3A80), paragraph [14](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A1994%3A80#point14)).\n\n( 14 ) Judgment of 27 April 2023, Fluvius Antwerpen ([C‑677\u002F21](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2023%3A348), [EU:C:2023:348](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2023%3A348), paragraph [31](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2023%3A348#point31)).\n\n( 15 ) Judgment of 4 September 2025, Arcomet Towercranes ([C‑726\u002F23](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A646), [EU:C:2025:646](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A646), paragraphs [35](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A646#point35) and [36](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A646#point36)).\n\n( 16 ) Opinion of Advocate General Richard de la Tour in Arcomet Towercranes ([C‑726\u002F23](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A244), [EU:C:2025:244](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A244), point [41](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A244#point41) et seq.).\n\n( 17 ) A possibly different conclusion is reached in the judgment of 4 September 2025, Arcomet Towercranes ([C‑726\u002F23](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A646), [EU:C:2025:646](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A646), paragraph [48](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A646#point48)), and in the Opinion of Advocate General Richard de la Tour in Arcomet Towercranes ([C‑726\u002F23](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A244), [EU:C:2025:244](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A244), point [48](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A244#point48)), which, however, did not address the argument to the contrary due to the specific facts of the case.\n\n( 18 ) Within the European Union, there is now even Council Directive (EU) 2017\u002F1852 of 10 October 2017 on tax dispute resolution mechanisms in the European Union ([OJ 2017 L 265, p. 1](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Foj\u002FJOL_2017_265_R_TOC)) for that purpose. However, that directive covers only disputes arising from the interpretation and application of treaties and conventions which provide for the elimination of double taxation on income and, where appropriate, on capital.\n\n( 19 ) Judgment of 3 July 2025, Högkullen ([C‑808\u002F23](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A516), [EU:C:2025:516](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A516), paragraph [21](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A516#point21)), with reference to judgments of 25 November 2021, Amper Metal ([C‑334\u002F20](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2021%3A961), [EU:C:2021:961](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2021%3A961), paragraph [28](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2021%3A961#point28)), and of 7 November 2013, Tulică and Plavoşin ([C‑249\u002F12 and C‑250\u002F12](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2013%3A722), [EU:C:2013:722](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2013%3A722), paragraph [33](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2013%3A722#point33)). 20 ( 21 ) See Working paper No 071 REV2 of the VAT Expert Group, taxud.c.1(2018)2326098 – EN, p 2. 22 ( 23 ) Judgment of 20 December 2017, Hamamatsu Photonics Deutschland ([C‑529\u002F16](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2017%3A984), [EU:C:2017:984](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2017%3A984), paragraphs [18](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2017%3A984#point18) and [31](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2017%3A984#point31) et seq.).\n\n( 24 ) Judgment of 15 May 2025, Tauritus ([C‑782\u002F23](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A353), [EU:C:2025:353](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A353), paragraphs [60](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A353#point60) and [61](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A353#point61)).\n\n( 25 ) Judgment of 4 September 2025, Arcomet Towercranes ([C‑726\u002F23](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A646), [EU:C:2025:646](http:\u002F\u002Fpublications.europa.eu\u002Fresource\u002Fecli\u002FECLI%3AEU%3AC%3A2025%3A646)).","https:\u002F\u002Feur-lex.europa.eu\u002Flegal-content\u002FEN\u002FTXT\u002FHTML\u002F?uri=CELEX:62024CC0603","ecli-eu-c-2026-21",{"caseNumber":50},"62024CC0603"]