LinkedIn Ireland v Data Protection Commission, 20th Apr

| | [Home ]
[Databases ]
[World Law ]
[Multidatabase Search ]
[Help ]
[Feedback ]
[DONATE ] | |
| # High Court of Ireland Decisions |
| You are here: BAILII >> Databases >> High Court of Ireland Decisions >>

LinkedIn Ireland UnLtd Company v Data Protection Commission (Approved) [2026] IEHC 235 (20 April 2026)
URL: https://www.bailii.org/ie/cases/IEHC/2026/2026IEHC235.html
Cite as:
[2026] IEHC 235 |
[New search ]

[Printable PDF version ]

[Help ]

harp graphic.

THE HIGH COURT

[2026] IEHC 235

Record Number 2024/580 MCA

IN THE MATTER OF SECTION 142 AND SECTION 150

OF THE DATA PROTECTION ACT 2018

Between

LINKEDIN IRELAND UNLIMITED COMPANY

**** Appellant

and

---

DATA PROTECTION COMMISSION

Respondent

JUDGMENT of Ms Justice Nessa Cahill delivered on 20 April 2026

INTRODUCTION

BACKGROUND

ISSUE 1: SCOPE OF SECTION 142

(a) Summary of the Parties' Arguments

(b) Relevant Provisions of the 2018 Act

(c) GDPR

(d) Conclusions on the interpretation of the statutory language

(e) Absurdity Arguments

(i) Overlap between infringement and fine

(ii) Orders under Section 143(3)

(iii) Consequences of separate appeals

(iv) Content of the LinkedIn Decision

ISSUE 2:  INTERPRETATION OF SECTION 142

(a) Preliminary

(b) Inconsistency in statutory appeals

(c) Categorising appeals

(d) Understanding the categories

(e) Interpretation of section 142

(i) Language of section 142

(ii) Positions of the parties

(iii) Comparison with other legislation

(iv) Conclusions on the interpretation of the statutory language

(f) GDPR, Charter and ECHR

(c) Curial Deference

(i)  Parties' positions

(ii) Interaction between standard of review and curial deference

(iii) Relevant authorities

(iv) Article 60

ISSUE 4: ADMISSIBILITY OF NEW EVIDENCE AND ARGUMENT

CONCLUSIONS

INTRODUCTION

1. This judgment arises in the context of an appeal by the Appellant (" LinkedIn") from a decision of the Respondent (" the DPC") made on 22 October 2024 (" the LinkedIn Decision"), which was made pursuant to section 113(2)(a) of the Data Protection Act 2018 (" the 2018 Act") and Article 60 of Regulation 2016/679/EU (" the GDPR").  By that Decision, the DPC found that LinkedIn had infringed Articles 5(1)a, 6(1), 13(1)(c) and 14(1)c of the GDPR and that corrective powers should be exercised in the form of a reprimand, an order to bring processing into compliance and the imposition of three administrative fines, totalling €310 million.

2. This judgment addresses the rules governing an appeal to this Court under the Data Protection Act 2018 (" the 2018 Act ").  Four issues were identified and agreed by the parties to be suitable and appropriate for preliminary trial.  They each concern the interpretation of provisions of that Act.

3. While the questions raised are complex, they do arise and have been presented, largely independently of the facts of these proceedings and in the absence of any relevant factual conflict.  For that reason, the background is briefly summarised to contextualise the issues raised but is not further examined.

4. The only truly salient facts are that the Data Protection Commission (" the DPC") conducted a six-year long investigation between 2018 and 2024, which culminated in a decision that LinkedIn had committed infringements of provisions of the General Data Protection Regulation (" the GDPR ") and a decision to exercise corrective powers, including the imposition of a fine in the amount of €310 million.

5. In addition to this statutory appeal, LinkedIn also issued judicial review proceedings challenging the decision of the DPC and certain provisions of the 2018 Act, invoking provisions of the Constitution, the Charter of Fundamental Rights of the European Union (" the Charter "), and the European Convention on Human Rights (" ECHR ").  There is a clear overlap between the two sets of proceedings.

6. The four preliminary issues are:

7. Is the Appellant entitled to appeal the decision of the Respondent dated 22 October 2024 (the "Decision") under section 142 of the Data Protection Act 2018 or are there aspects of the within appeal which can only proceed as an appeal under section 150 of the 2018 Act?

8. What type of appeal does section 142 of the Data Protection Act 2018 provide for and what is the standard of review/assessment to be applied by the Court in an appeal under section 142?

9. If there are aspects of the Appellant's appeal which can only proceed as an appeal under section 150 of the Data Protection Act 2018, is the standard of review/assessment to be applied by the Court the same under sections 142 and 150 of the 2018 Act or does a different standard of review/assessment apply in an appeal under section 150 of the 2018 Act?

10. Having regard to the answers to the foregoing questions, is the Appellant entitled to rely on evidence adduced and/or arguments made in the Appeal which were not already adduced or made to the Respondent during the inquiry process?

11. Each of the parties adopted the position that the same type of appeal and standard of review are applicable to an appeal under section 142 and section 150 of the Act.  Accordingly, the answers under Issue 2 will also answer the questions raised in Issue 3 and no submissions were made regarding Issue 3.  Consequently, the nature and the standard of review are addressed in this judgment by reference to section 142 only and Issue 3 is not addressed as a discrete topic.  Issue 4, by contrast, is focussed on section 150.

BACKGROUND

1. This judgment arises in the context of an appeal by the Appellant (" LinkedIn") from a decision of the Respondent (" the DPC") made on 22 October 2024 (" the LinkedIn Decision"), which was made pursuant to section 113(2)(a) of the Data Protection Act 2018 (" the 2018 Act") and Article 60 of Regulation 2016/679/EU (" the GDPR").  By that Decision, the DPC found that LinkedIn had infringed Articles 5(1)a, 6(1), 13(1)(c) and 14(1)c of the GDPR and that corrective powers should be exercised in the form of a reprimand, an order to bring processing into compliance and the imposition of three administrative fines, totalling €310 million.

2. The LinkedIn Decision was arrived at following a complaint-based inquiry conducted under section 110 of the 2018 Act (" the Inquiry") which spanned a period of more than six years. The stages of the Inquiry process will be summarised here.

The Complaint

1. The Inquiry was commenced following a complaint filed by La Quadrature du Net, a French non-profit organisation (" the Complaint"). The Complaint, which was asserted to have been made on behalf of 8,540 users of the LinkedIn service in France, was lodged on 28 May 2018 with the French supervisory authority for GDPR compliance, Commission Nationale de l'Informatique et des Liberts (" CNIL ") and primarily concerned LinkedIn's processing of the personal data of its members in the EEA and the UK for the purpose of " behavioural analysis and targeted advertising ", referred to in the LinkedIn Decision as " BA & TA ".  The Complaint alleged that such processing was not in compliance with the requirements of the GDPR and further alleged breaches of the transparency and fairness requirements in Articles 13 and 5(1)(a) GDPR.  The Complaint was referred by CNIL to the DPC as the supervisory authority (" SA ") with jurisdiction over LinkedIn in matters concerning cases of cross-border processing under the GDPR.

The Investigation

1. On 20 August 2018 LinkedIn was notified by the DPC ** of the commencement of the Inquiry (" the Notice of Commencement").  The letter further contained a request for information and a translated copy of the Complaint.  LinkedIn answered this request for information on 1 October 2018. On 17 January 2020, the DPC sent LinkedIn a second letter containing a further request for information. On 14 July 2021, the DPC provided LinkedIn with a draft statement of the issues in the investigation and invited submissions.  LinkedIn's submissions on the draft statement of issues were furnished on 4 August 2021.  On 4 October 2021, the DPC sent a further request for information.  LinkedIn provided the requested information on 29 October 2021. The DPC provided LinkedIn with its finalised statement of issues on 22 December 2021.

The Preliminary Draft Decision

1. By letter dated 9 March 2023, LinkedIn was informed that a preliminary draft decision was being prepared. This letter also noted the DPC's position that for the purposes of any administrative fine under Article 83 GDPR LinkedIn and Microsoft were considered to comprise a single undertaking. The DPC further requested certain information about the group structure, including a corporate structure chart and figures related to Microsoft's turnover.

2. LinkedIn complied with this request without prejudice to its assertion that LinkedIn and Microsoft should not be treated as a single undertaking for the purposes of considering the imposition of an administrative fine.  On 25 April 2023, the DPC provided LinkedIn with its Preliminary Draft Decision (" the PDD") and invited submissions in response. The DPC's preliminary findings, contained in the PDD, were that LinkedIn had infringed Article 5(1)(a), Article 6, and Articles 13(1 )(c) and 14(1 )c) of the GDPR.  The DPC indicated its intention to order the following corrective measures:

(i) to reprimand LinkedIn.

(ii) to make Corrective Orders requiring Linkedln to bring its processing into compliance with the GDPR within three months.

(iii) to impose fines of between €430 - €490 million on LinkedIn.

1. On 20 July 2023 LinkedIn delivered submissions in response to the PDD, supported by factual and expert evidence. In its response, LinkedIn disputed the provisional conclusions and the proposed fines, submitting that they were excessive and disproportionate.

The Draft Decision

1. On 11 July 2024, the DPC submitted a draft decision to the concerned SAs in other Member States (" the CSAs ") in accordance with Article 60(3) GDPR and section 113 of the 2018 Act.  On 11 July 2024 the DPC furnished LinkedIn with a copy of the draft decision (" the Draft Decision") along with notice that the Draft Decision had been sent to the CSAs in accordance with Article 60 GDPR.  The Draft Decision broadly reached the same conclusions as the PDD finding that Article 5(1)(a), Article 6, and Articles 13(1)(c) and 14(1)(c) GDPR had been breached.  The DPC did concede some of the objections raised by LinkedIn in its response to the PDD, including submissions that it was not appropriate to impose a separate fine for infringement of the fairness principle under Article 5(1)a) GDPR, as that conduct had already been taken into account in imposing the fines in respect of other infringements.  As a result of the changes made in the Draft Decision, the proposed fines (which had totalled between €430 and €490 million under the PDD) were reduced to between €290 and €320 million in the Draft Decision.

The Revised Draft Decision and Comments

1. Three CSAs raised comments on the Draft Decision but no objections were raised.  On 24 September 2024, the DPC informed LinkedIn that it proposed to make non-material amendments to the Draft Decision to take account of the views expressed by the three CSAs in these comments. The DPC further informed LinkedIn that it proposed to have regard to these comments in determining the amount of the fines to be imposed within the scale already determined by the DPC in the Draft Decision.  LinkedIn delivered submissions on 8 October 2024.

The Decision

1. In its final Decision, dated 22 October 2024, the DPC made the following findings:

a) LinkedIn did not validly rely on Article 6(1)(a) GDPR to process third-party data of its members for the purpose of " BA & TA ", excluding analytics, on the basis that the consent obtained by Linkedln was not freely given, sufficiently informed or specific, or unambiguous. Accordingly, LinkedIn contravened Article 6 and Article 5(1)(a) GDPR.

b) LinkedIn did not validly rely on Article 6(1)(f) GDPR for its processing of first-party personal data of its members for "BA & TA ", or third-party data for analytics. Accordingly, LinkedIn contravened Article 6 and Article 5(1)(a) GDPR.

c) LinkedIn did not validly rely on Article 6(1)(b) GDPR to process first-party data of its members for the purpose of " BA & TA ". Accordingly, LinkedIn contravened Article 6 and Article 5(1)(a) GDPR.

d) LinkedIn contravened Article 13(1)(c) and Article 14(l)(c) GDPR in respect of the information it provided to data subjects regarding its reliance on Article 6(1)(a), Article 6(1)(b) and Article 6(1)(f) GDPR as lawful bases.

e) LinkedIn did not contravene Article 13(1)(d) or Article 14(2)(b) GDPR.

f) LinkedIn contravened the principle of fairness in Article 5(l)(a) GDPR.

The Proceedings

1. On 18 November 2024 LinkedIn issued a statutory appeal of the Decision pursuant to sections 142 and 150 of the 2018 Act.  The appeal was grounded on the affidavit of Ms Sue Duke and supported by an affidavit sworn by Professor Andrew Stephen and Mr Abhishek Shrivastava.  On the same date, an ex parte application for leave to seek judicial review proceedings was filed.

2. By letter dated 18 November 2024 solicitors for LinkedIn wrote to the DPC enclosing copies of the statutory appeal and judicial review papers. In that correspondence, LinkedIn explained that the judicial review proceedings were confined to challenges to the validity and compatibility of aspects of the statutory scheme and that the statutory appeal was intended to address the merits of the Decision, with a view to avoiding duplication between the two sets of proceedings.  By letter dated 25 November 2024, solicitors for the DPC responded declining to confirm that all issues raised could be addressed within the statutory appeal and stating that the procedural route by which to challenge the Decision was a matter for the appellant to determine.  Further correspondence was exchanged between the parties in December 2024 in which LinkedIn reiterated its position that duplication should be avoided and sought confirmation that the issues raised could be addressed within the statutory appeal. The DPC maintained that it was for LinkedIn to determine how to proceed with its challenge.

3. On 16 December 2024 Gearty J. granted LinkedIn leave to seek judicial review on the basis of a draft amended statement of grounds (" the Judicial Review ").

4. On 25 February 2025, the DPC delivered its statement of opposition, verified by an affidavit sworn by Ms Fleur O'Shea.  Replying affidavits were sworn by Ms Duke on 8 April 2025 and on 26 May 2025 (a replying affidavit having been sworn by Mr Owen O'Donnell on behalf of the DPC on 6 May 2025).  The DPC's statement of opposition in the judicial review proceedings was delivered on 4 March 2025.  The State delivered its statement of opposition in the Judicial Review on 18 March 2025.

Identification of Preliminary Issues

1. The second affidavit of Ms. Duke addressed, among other matters, the position adopted by the DPC as to the scope and standard of appeal under sections 142 and 150 of the Data Protection Act 2018 and indicated that LinkedIn might seek the determination of those matters as preliminary issues.

2. In May 2025, further correspondence was exchanged between the parties concerning the possible determination of preliminary issues relating to the scope of sections 142 and 150, the applicable standard of review, and the admissibility of new evidence on appeal.  On 22 May 2025, a motion was issued by LinkedIn seeking an order pursuant to Order 25, rule 1 and/or Order 34, rule 2 of the Rules of the Superior Courts directing a trial of the four specified preliminary issues. The motion was grounded on an affidavit sworn by Mr Richard Willis.  The DPC agreed to the trial of these preliminary issues and, by order of the High Court (Gearty J.) made on consent on 25 June 2025, the Court directed a trial of these preliminary issues. The order further provided for the participation of the State at the hearing of the motion.

3. Written submissions were delivered by LinkedIn on 24 September 2025, by the DPC on 29 October 2025 and by the State on 18 November 2025.  The preliminary issues were heard over the course of three days, on 2, 3 and 4 December 2025.   Very helpful supplemental submissions were then delivered by LinkedIn on 19 and 24 March 2026; by the DPC on 16 March 2026 and on 26 March 2026; and by the State on 20 March 2026.

ISSUE 1: SCOPE OF SECTION 142

1. The question that arises under Issue 1 is whether an appeal under section 142 encompasses the entire Decision or whether it is confined to an appeal against the imposition of the fine. **** The decision as to the scope of section 142 hinges primarily on the interpretation of sections 142 and 150(5), an exercise which should be undertaken in accordance with the principles set out in the judgment of Murray J. in Heather Hill Management Company CLG v An Bord Pleanala [2022] IESC 43 (O'Donnell C. J., O'Malley, Woulfe, Hogan JJ. in agreement) (" Heather Hill ") which were summarised as follows:

" The words of the section are the first port of call in its interpretation, and while the court must construe those words having regard to the context of the section of the Act in which the section appears, the pre-existing relevant legal framework and the object of the legislation insofar as discernible, the onus is on those contending that a statutory provision does not have the effect suggested by the plain meaning of the words chosen by the legislature to establish this " (at [128]).

1. I will therefore address the correct interpretation of those statutory provisions (read in their immediate and broader context), before turning to the submissions of LinkedIn about the impact of that interpretation.

(a) Summary of the Parties' Arguments

1. LinkedIn contends that the correct interpretation of the 2018 Act is that, when a fine is imposed, the decision in its entirety, including the infringement findings and any other corrective powers that were exercised, may be appealed under section 142. This is said to be a recognition of the need for special protection when fines are imposed.  If the decision culminates only in other corrective measures and no fine, then the appeal falls only under section 150(5).

2. It is contended that, if it was intended that there must be a bifurcation of the appeal, this must be provided for in explicit terms, but this was not done. It is contended that there is no basis in the language of sections 142 and 150 to separate out the appeals as between infringement and fines or to require the issues to be decided according to different rules.

3. Further, LinkedIn relies heavily on the overlap between a decision on infringement and a decision as to the imposition of fines and asserts that it is impracticable and absurd for appeals against such decisions to be subject to separate appeals under different statutory provisions and applying different standards and rules, including with respect toto the admission of new evidence and argument.

4. The position of both the DPC and the State is that section 142 governs only a decision on the imposition of a fine and that this is distinct and separate from the finding of infringement, which can only be appealed under section 150(5).  The DPC relies on the language of section 142 and the references to " fines " throughout that provision and its neighbouring provisions, particularly the incorporation of Article 83 GDPR in section 142(4).   The DPC also points to the language of sections 111, 112 and 113 which provide for a decision on infringement and " in addition " a decision as to the exercise of corrective measures, which points towards separate, distinct and segregable decisions.  Both the State and the DPC assert that there is no ambiguity in the language of these sections.

5. According to the State, the divisibility of the sanctions appeal arises from the language of the 2018 Act and the distinction drawn in that Act between sections 142 and 150(5) must be given effect to.

(b) Relevant Provisions of the 2018 Act

1. The first " port of call ", as confirmed in Heather Hill, is the language of the relevant provisions . Section 142(1) provides that,

" Without prejudice to section 150 , a controller or processor that is the subject of a decision under section 111 , 112 , 113 or 133 (9) to impose an administrative fine may, within 28 days from the date on which notice of the decision concerned was given to it under section 116 or, as the case may be, section 133 (9)(b) appeal to the court against the decision."

1. Section 142(4) then states, " The court shall, for the purposes of subsection (3), act in accordance with Article 83 ".

2. Section 150(5) provides a general right of appeal against a decision of the DPC (such as the decision at issue in these proceedings):

" A data subject or other person affected by a legally binding decision of the Commission under Chapter 2 or 3 may, within 28 days from the date on which notice of the decision is received by him or her, appeal against the decision. "

1. It is immediately apparent that these provisions cannot be construed in isolation, and must instead be read in conjunction with other proximate provisions of the 2018 Act and indeed of the GDPR (both appear in Part 6 of the 2018 Act which is concerned with the enforcement of the GDPR).

2. The correct interpretation of section 142 and section 150(5) and the positions advanced by the parties will be addressed, beginning with the construction of the statutory language itself.

Section 150(5)

1. Section 150(5) refers to a " legally binding decision " under Chapter 2 or 3 of Part 6.  Chapter 2 comprises sections 107 to 117 of the 2018 Act.  These sections concern, in general terms, investigations of suspected infringements of a provision or provisions of the GDPR or the 2018 Act.  Investigations may arise from complaints made to the DPC (section 108) or may be initiated by the DPC itself.  In either case, section 110 provides that the DPC may conduct an investigation and inquiry.   The relevant provision for the purposes of these proceedings is section 113 which addresses the outcome when a complaint has been investigated by the DPC as the lead Supervisory Authority (" LSA ") pursuant to the cooperation mechanism contained in Article 60 GDPR.

2. Chapter 3 concerns the enforcement of Directive (EU) 2016/680 (" the Law Enforcement Directive ") on the protection of natural persons with regard to the processing of personal data by competent authorities for the purposes of the prevention, investigation, detection or prosecution of criminal offences or the execution of criminal penalties.  It is not relevant here, but I would note that a similar approach to that taken in sections 111, 112 and 113 can be seen in Chapter 3.  For example, sections 124 and 125 provide for the DPC to make a decision as to whether there was an infringement and then state that the DPC " shall, in addition, make a decision " concerning whether and which corrective power to exercise.

Section 150(12)

1. Section 150(12) defines and limits the scope of the decisions to which section 150 applies.  It provides;

" For the purposes of this section, a "legally binding decision" means a decision —

(a) under paragraph (a),(b) or (da) of section 109(5) or paragraph (a),(b) or (da) of section 122(4),

(b) under section 111(1)(a), 112(1), 113(2)(b) or (6)(aa), 114, 124(1)(a) or 125(1), or

(c) to exercise a corrective power under Chapter 2 or 3."

1. Before addressing the specific provisions listed in section 150(12)(b), it is relevant to observe the breadth of section 150(12)(c).  By this provision, the exercise of any corrective power under Chapter 2 without distinction is within the scope of section 150. LinkedIn argued in oral submissions that a controller or processor could choose to appeal the decision to impose a fine under section 150(5) or under section 142.  The State seems to be of a different view and points out that fines are not expressly mentioned in section 150(12)(c).  The State also makes the uncontroversial point that it is Chapter 6 as opposed to Chapter 2 which contains the provisions of the 2018 Act which deal specifically and substantively with fines.  The DPC did not specifically address the question of whether section 150(12) includes administrative fines, although the tenor and logical inference of the submissions made by the DPC is that fines may only be appealed under section 142.

2. While there was limited focus on the terms of section 150(12)(c), I do think it is relevant to interpreting the intersection between sections 142 and 150 and aspects of the language of each.  The starting point is that it is inescapably the case that section 150(12) is broad in its formulation and would capture any decision on infringement under sections 111, 112 or 113 and the exercise of any corrective power under those provisions.  Indeed, all parties agree that section 150 is the default appeal provision under Part 6 of the 2018 Act.

3. As will be seen when it comes to section 115 and section 142(1) itself, a decision to impose an administrative fine under the 2018 Act is a decision taken under sections 111(2), 112(2) or 113(4).  Section 150(12)(c) does therefore on its terms include a decision to impose an administrative fine under each of those provisions.  This means that a decision to impose an administrative fine is a " legally binding decision " for the purposes of section 150(5). It may be recalled in this connection that section 150(5) is permissive in its formulation (a " person affected by a legally binding decision of the Commission under Chapter 2 or 3 may...appeal ").  The only relevant consequence of encompassing a fining decision within section 150 is that the controller or processor "may" appeal under section 150(5).  This does not detract from the undisputed fact that a decision to impose a fine may also be appealed under section 142.

4. There are some consequential points to note.  First, when section 150(12) is read this way, that makes sense of the proviso in section 142 that it is " without prejudice to section 150 " (to which I will return later).  On this interpretation, a processor or controller faced with a decision to impose an administrative fine may appeal under the default appeal route in section 150(5) or choose to appeal the decision to impose the fine under section 142.  This appears to be consistent with LinkedIn's interpretation of the scope of section 150 as regards the imposition of fines.  It does not however deal with the broader case made by LinkedIn about appealing findings of infringement.

5. Second, as the DPC points out, the language of " legally binding decision " in section 150 may be traced to Article 78(1) GDPR: " Without prejudice to any other administrative or non-judicial remedy, each natural or legal person shall have the right to an effective judicial remedy against a legally binding decision of a supervisory authority concerning them. "  If a controller or processor does elect to appeal a fining decision under section 150, this does not detract from or lessen the entitlement to an effective judicial remedy.  The extent of the protection available is further confirmed by the positions adopted by each of the parties to these proceedings (and by the State) that the same standard of review is applicable whether the appeal is pursued under section 150(5) or section 142.

6. As an aside (although one that is not irrelevant in light of the absurdity arguments made by LinkedIn), the option of appealing under section 150(5) would allow for an appeal against a decision on infringement and the exercise of non-fine corrective powers to proceed together with an appeal against a decision on the imposition of administrative fines.

Sections 111, 112, 113

1. The provisions referenced in section 142(1) are sections 111, 112, 113 and 133(9).  Section 133(9) is concerned with the imposition of administrative fines for failure to comply with an enforcement notice and is not relevant here.

2. Section 111 addresses the decisions that may be made by the DPC following an " own volition " investigation and provides (in relevant part),

" (1) Where an inquiry has been conducted of the Commission's own volition, the Commission,   having considered the information obtained in the inquiry, shall—

(a) if satisfied that an infringement by the controller or processor to which the inquiry relates has occurred or is occurring, make a decision to that effect, and

(b) if not so satisfied, make a decision to that effect.

(2) Where the Commission makes a decision under subsection (1)(a), it shall, in addition, make a decision—

(a) as to whether a corrective power should be exercised in respect of the controller or processor concerned, and

(b) where it decides to so exercise a corrective power, the corrective power that is to be exercised."

Section 107 of the 2018 Act defines " corrective power " as " a power conferred by Article 58(2) " of the GDPR.  Article 58(2) lists the corrective powers of the SAs which include the powers to:

i. issue warnings about intended processing operations which are likely to infringe the GDPR;

ii. issue reprimands where there has been infringement;

iii. order the controller or processor to bring processing operations into compliance with the GDPR;

iv. impose a ban on processing;

v. order the rectification or erasure of personal data.

1. Notably, one of the corrective powers referenced in Article 58 is the power " to impose an administrative fine pursuant to Article 83, in addition to, or instead of measures referred to in this paragraph, depending on the circumstances of each individual case " (Article 58(2)(i)).

2. Section 112 concerns inquiries on foot of complaints in respect of which the DPC is the competent SA.  Section 112(1) provides, in similar terms to section 111(1), that the DPC may either make a decision that there was infringement or a decision dismissing the complaint.  Section 112(2) (also mirroring section 111(2)) then provides:

"Where the Commission makes a decision under subsection (1)(a), it shall, in addition, make a decision—

(a) as to whether a corrective power should be exercised in respect of the controller or processor concerned, and

(b) where it decides to so exercise a corrective power, the corrective power that is to be exercised."

1. As already noted, section 113 of the 2018 Act governs the investigation and determination of complaints in respect of which the DPC is the lead SA for the purposes of Article 60 of the GDPR.  This is the applicable provision in this case.  It sets out the requirement that the DPC shall:

" make a draft decision in respect of the complaint (or, as the case may be, part of the complaint) and, where applicable, as to the envisaged action to be taken in relation to the controller or processor concerned " (section 113(2)(a)).

1. This is the provision pursuant to which the LinkedIn Decision was made.  The DPC must then " in accordance with Article 60 and, where appropriate, Article 65, adopt its decision in respect of the complaint or, as the case may be, part of the complaint " (section 113(2)(b)).

2. Section 113(4) then provides:

" Where the Commission adopts a decision under subsection (2)(b) to the effect that an infringement by the controller or processor concerned has occurred or is occurring, it shall, in addition, make a decision—

(a) where an inquiry has been conducted in respect of the complaint—

(i) as to whether a corrective power should be exercised in respect of the controller or processor concerned, and

(ii) where it decides to so exercise a corrective power, the corrective power that is to be exercised."

1. There is a notable element of repetition between sections 111(2), 112(2) and 113(4).  Each of these provisions refers to two decisions that the DPC is required to make or adopt: a decision as to whether there was (or is) an infringement and, if infringement is found, the DPC shall also (" in addition ") make a decision as to whether and which corrective power should be exercised. The DPC and the State contend that there are two discrete, segregable decisions provided for in these sections.

2. LinkedIn's position on the formulation of these provisions is that they do not provide for separate decisions, but rather a single decision dealing with both infringement and sanction (drawing an analogy with liability and quantum in civil cases).  There is however an acknowledgement that it is a two-stage decision making process.

3. I consider the requirement that the DPC must " in addition, make a decision " as to the corrective powers to be unambiguous.  It plainly indicates that there must be one decision on infringement and then a second additional decision as to the exercise of corrective powers (if infringement is found).   This is the interpretation urged by the DPC and the State and it does garner support from the language of sections 111(2), 112(2) and 113(4).

4. In each of section 111, 112 and 113, there is also a separate, final sub-section to the effect that, where the DPC " makes a decision " to exercise a corrective power, it " shall exercise the corrective power concerned " (section 111(3), section 112(3), section 113(6)(a)).

Section 115(1)

1. The next provision to note is section 115(1):

" For the purposes of exercising a corrective power under section 111, 112 or 113, the Commission may do either or both of the following:

(a) subject to Chapter 6, decide to impose an administrative fine on the controller or processor concerned;

(b) exercise any other corrective power specified in Article 58(2)."

1. The parties' position on section 115 may be briefly summarised as follows.  The DPC refers to section 115 as indicating a dichotomy between the imposition of a fine and other corrective powers and as evidencing the need for a decision as to the imposition of a fine.  LinkedIn argues that there is only one decision provided for in the 2018 Act and in the GDPR and that all aspects of the decision and all corrective measures imposed are addressed in the single indivisible decision, which may contain three findings (as to infringement, corrective measures and fines). LinkedIn does not however disagree that there is a difference between the imposition of a fine and other corrective powers.  Rather, it was contended on behalf of LinkedIn that section 115 highlights that the exercise of the corrective power to impose a fine brings the entire decision within the scope of the special section 142 appeal procedure.

2. I would immediately note that the separate treatment of fines in section 115 has a greater consequence than LinkedIn allows: it confirms that it is the decision to impose a fine that is " subject to Chapter 6 ".  That wording is solely applicable to the fining decision and is not referable to the exercise of other corrective powers.

3. One possible interpretation of the plain language of section 115 (read in light of the language of sections 111(2), 112(2) and 113(4)) is that there should be a decision on infringement, then a decision to exercise a corrective power, followed by the exercise of the corrective power, and " for the purposes of exercising " that power, there may then be a third decision to impose a fine.  None of the parties to these proceedings advanced that rather technical, literal interpretation and I am satisfied that it is not intended that there would be a third decision following a decision under sections 111(2), 112(2) or 113(4) as to which corrective power to exercise.  Instead, I regard the effect and purpose of section 115 as three-fold: first, it differentiates the specific corrective power to impose a fine from the other corrective powers that may be exercised; second, it confirms the necessity of a discrete, identifiable decision when the DPC is exercising the power to impose a fine; and, third, it provides that the decision to impose an administrative fine is " subject to Chapter 6 ".

4. This is consistent with the position (made manifest in section 142(1)) that the decision to impose a fine is a decision that itself is made under sections 111, 112 or 113 and not under section 115. This dichotomy between fines and other corrective powers can be seen in Article 58, which includes the imposition of fines among the list of possible corrective powers, but on the basis that it is a power to be exercised in accordance with Article 83.

5. When sections 111, 112, 113 and 115 are read together and in light of section 142(1), I am of the view that those provisions indicate that there are two separate decisions: (a) a decision as to whether there is infringement; (b) if there is a finding of infringement, a decision as to whether and which corrective power(s) to exercise.  If it is decided to exercise the power to impose a fine, there must be a specific decision to that effect (which is then subject to Chapter 6).

6. I pause here to note that the requirement of a separate " decision " is not to be equated with a requirement for a separate document containing each individual decision.   I did not detect any argument to the contrary by the parties to these proceedings and would simply note that there is nothing in the 2018 Act or the GDPR to suggest that several decisions cannot be contained in a broader composite " decision " (as indeed the DPC did in this case).   LinkedIn does however rely on the fact that there was a single " decision " issued here in support of its case as to the scope of section 142.

7. It appears to me that the question of whether findings of infringement and fine imposition are included in one document cannot itself be decisive as to the application of Article 60 or the scope of section 142.   There is nothing in section 142 to support such an interpretation of its scope.  The term " decision " is not defined in the GDPR or the DPA, it is plain that section 142 and Article 83 are directed to " a decision to impose an administrative fine " (section 142(1)) and a SA when " deciding whether to impose an administrative fine " (Article 83(2)).  Even if the decision regarding the imposition of a fine is included with other findings or decisions in an overall document or decision, that cannot detract from the specific treatment of fining decisions under Chapter 6.  The use of the term " decision " is not therefore necessarily an individually documented decision, but can rather be a " decision " within a broader document which also includes other statutory decisions.

8. The LinkedIn Decision at issue here is a single document which includes discrete sections dealing with decisions as to infringement, whether to exercise corrective powers and which powers to exercise.   The question of infringement is addressed under the headings of Issues A to E and occupies pages 22 to 107 of the document.  This is the decision adopted pursuant to section 113(2)(b).  The decision as to whether and which corrective powers to exercise is then addressed from pages 108 to 110, with the decision on the imposition of administrative fines commencing on page 111 and continuing to page 137.  The fact that each of these elements of the overall decision are in the same document does not alter the scope of section 142.

9. As the terminology may be apt to confuse, I use the term " decision " when addressing the " decision " as to whether there was infringement and the " decision " as to whether to exercise the corrective power to impose a fine.  LinkedIn in its submissions favours the designation of these as " findings " or stages in a decision-making process, rather than as separate " decisions ".   It is argued by LinkedIn that there is only one decision.  Based on the language of sections 111(2), 112(2), 113(4) and 115(1)(a) and the interpretation of that language in this judgment, I consider that the finding that there was infringement and the determination as to which corrective powers to exercise and, specifically, whether and in what amount to impose a fine, are properly to be termed " decisions " as that is the language of Part 6 of the DPA.  I would however emphasise that this is not to be confused with the references to " decision " in other contexts, particularly Article 60 GDPR.  As already noted, there is no rule or requirement of EU law as to how a decision of an SA is to be structured or composed and there is certainly no impediment to the DPC issuing a composite " decision " which includes several distinct decisions, as was done here.  This does not alter the analysis or interpretation of the relevant provisions of the DPA.  I do not attach any weight to the inclusion in a single composite document of the decision on infringement; the decision on the imposition of a fine; and the decision on the exercise of other corrective measures, or to the application by the DPC of the singular term " decision " to that document.  Rather, the material factor in the interpretation of the 2018 Act is the language of that Act and the relevant provisions (sections 111(2), 112(2) and 113(4)) unambiguously require distinct decisions and that is the correct framework for the analysis here (regardless of how the decisions are presented).

Section 116

1. A further provision to which LinkedIn attaches particular weight is section 116.  In the context of section 113 (the relevant provision here), section 116(2) provides:

" Subject to subsection (4), the Commission shall—

(a) as soon as practicable after it adopts a decision under section 113 (2)(b), give the controller or processor concerned a notice in writing setting out—

(i) the decision and the reasons for it, and

(ii) where applicable, the corrective power that the Commission has decided to exercise or, as the case may be, the action that it has decided to take, in respect of the controller or processor,

and

(b) in the case of a complaint lodged with the Commission, and as soon as practicable after the giving of the notice under paragraph (a), give the complainant concerned a notice in writing setting out—

(i) the decision and the reasons for it, and

(ii) where applicable, the corrective power that the Commission has decided to exercise or, as the case may be, the action that it has decided to take, in respect of the controller or processor. "

1. LinkedIn relies on the references in section 116 to " the decision " singular, as evidence of a unitary decision, the entirety of which must be appealable together, whereas the DPC contends that section 116 continues to evidence the dichotomy between the decision on infringement (and the reasons for it) and the corrective powers which the DPC has decided to exercise, each being addressed separately in that provision.  I am not persuaded that section 116 assists greatly in the interpretation of the scope of section 142.  It does refer to notice of a " decision " as a general, singular term, but then also refers - separately and disjunctively - to notice of " the corrective power that the Commission has decided to exercise ".  It does not clearly advance either interpretation although it is arguably more consistent with the view of the DPC as to the disjoint between infringement and corrective measures.

Chapter 6

1. The next question of interpretation of the 2018 Act concerns Chapter 6.  This Chapter contains sections 141, 142 and 143 and deals with the topic of the imposition of administrative fines.  Before addressing the text of those provisions, it is relevant to summarise LinkedIn's position.  LinkedIn embraces the proposition that it is the imposition of an administrative fine which warrants the particular appeal provisions contained in section 142.  However, it asserts that this does not result in or arise from a bifurcation of the decision on infringement and fines. The case made is that there is at all times a single " decision " and that, if a fine is imposed as part of that decision, the decision in its entirety is appealable under section 142. I will return to this question when I assess the alleged overlaps between infringement and fines and the absurdity arguments raised by LinkedIn.

2. There is however no dispute - nor could there be - but that the power to impose an administrative fine is of a different character to the other corrective powers conferred on an SA by the GDPR or that it is only when a fine is imposed that section 142 is available.  The dispute hinges on whether the other decisions may then also be appealed under section 142.

Section 141(1)

1. The starting point is section 141(1) which requires the DPC to act in accordance with section 141 and Article 83 GDPR in considering

" (a) whether to make a decision to impose an administrative fine, and

(b) where applicable, the amount of such a fine."

1. The plain language of " decision to impose an administrative fine " suggests that there is a distinct identifiable decision to impose a fine.  This also reflects section 115(1). The requirement to act in accordance with section 141 and Article 83 GDPR when considering whether and in what amount to impose a fine suggests a need for a deliberate and careful assessment of the requirements of those provisions when considering the imposition of an administrative fine.  This in turn points to a need for a clear segregation in the decision-making analysis, such that there is an identifiable decision-making stage during which those requirements are triggered.

Section 142(1)

1. Section 142(1) then provides that

" Without prejudice to section 150, a controller or processor that is the subject of a decision under section 111, 112, 113 or 133(9) to impose an administrative fine may, within 28 days from the date on which notice of the decision concerned was given to it under section 116 or, as the case may be, section 133(9)(b) appeal to the court against the decision."

1. The wording of this provision was addressed quite extensively by the parties and the discrete arguments - and my analysis of them - will next be addressed.

2. Beginning with the introductory wording that section 142 is " without prejudice to section 150 ", counsel for LinkedIn contends that this does not limit the scope of section 150, but is rather a statutory device to ensure there is no unintentional effect on that provision.  LinkedIn's position is that section 150 does allow for an appeal against decisions to impose corrective measures, including fines, but that section 142 is an alternative avenue of appeal when a fine is imposed and includes the important additional facility to introduce new arguments and evidence.

3. It is argued by the DPC that the starting point is that section 150 applies to all appeals and its breadth is acknowledged by the " without prejudice " language in section 142. This signals that section 142 is not intended to undermine or "read down" section 150.  The DPC contends that, if LinkedIn's interpretation of the scope of section 142 was correct, there would be no reference to section 150.  My understanding of the DPC's submissions on this point is that the words " without prejudice to section 150 " confirm and advance the DPC view that the decision on infringement remains a decision which is to be appealed under section 150.

4. I prefer the view of LinkedIn on this specific point (although not because of the need for a unitary appeal as LinkedIn suggests).  The starting point is that the parties agree that there is no possibility of a decision other than a decision "to impose a fine " being subject to section 142 (while sharply diverging as to what a decision "to impose a fine " encompasses). As I have already discussed in the context of section 150(12), I am satisfied and agree with LinkedIn that an appeal against a decision to exercise the power to impose a fine may be brought under section 150(5).

5. Accordingly, the words " without prejudice to section 150 " in section 142 serve a meaningful purpose: they confirm that the right of appeal under section 142 does not prejudice the right of a controller or processor to elect to challenge the imposition of a fine under section 150(5).  These words would not be necessary unless there was otherwise a risk that section 142 would prejudice the scope of section 150.

6. The broader proposition which appears to be advanced by the DPC - that the words " without prejudice to section 150 " confirm that section 150 is not to be generally " read down " but rather continue to apply to the infringement decision - involves a circularity of sorts.  If the "decision to impose a fine" is limited to the fining decision as the DPC contends, there is no risk that section 142 could apply to the decision more generally and the words " without prejudice to section 150 " would not be needed to confirm that the infringement decision falls under section 150.

7. The presumption against superfluous wording (Cork Co Council v. Whillock [1993] 1 IR 231) would lean against that interpretation. Somewhat ironically, it is only if LinkedIn was correct and both aspects of a DPC decision (infringement and fine) would otherwise come under section 142, that the carve-out for section 150 for which the DPC contends would make sense, but that cannot sensibly be the premise of the DPC's interpretation of section 142(1).

8. In any event, I do not favour that interpretation (if it is the interpretation of the DPC) and repeat my view that the reason for the reference to section 142 being " without prejudice to section 150 " is to confirm that a processor or controller retains the option to appeal a decision on the imposition of a fine under section 150(5), section 142(1) notwithstanding.

9. The next step is to interpret the words " adecision under section 111, 112, 113 or 133(9) to impose an administrative fine ". LinkedIn's position is that the " decision.. to impose an administrative fine " includes both findings of infringement and the corrective powers being exercised.   LinkedIn also contends that, if section 142 was to be confined to sections 111(2) or 112(2) or 113(4) (which relate to corrective powers), that would have been set out, as it was in section 115, but this did not happen.  As a matter of statutory language, I do not see that this point assists LinkedIn.  If there was a reference to sections 111(2), 112(2) or 113(4) in section 142, that would bring all corrective measures within the scope of section 142 but would point more firmly towards the exclusion of the infringement decisions under section 111(1), 112(1) and 113(2(b)

10. The DPC's view is that, if it was intended that section 142(1) would capture decisions on infringement as well as decisions on the imposition of fines, then the words " to impose an administrative fine " would be otiose.  The presumption against redundancy in legislation is referenced to counter that interpretation of section 142(1).  The DPC also contends - compellingly - that LinkedIn in fact reads section 142(1) as if it included the words " decision which includes the imposition of a fine" rather than decision "to impose an administrative fine ".

11. Weighing these contradictory positions, it seems to me that the plain language of section 142(1) does favour the view that it is the specific decision to impose a fine that is captured by that provision.  The reference to the decision " to impose an administrative fine " is specific and unambiguous.  If the Oireachtas intended the interpretation urged by LinkedIn, section 142(1) could have referenced " a controller or processor that is the subject of a decision under section 111, 112, 113 or 133(9) which includes the imposition of an administrative fine may, within 28 days from the date on which notice of the decision concerned was given to it under section 116 or, as the case may be, section 133(9)(b) appeal to the court against the decision."  That is not the wording of section 142(1).

12. Its interpretation of section 142(1) is supported, according to the DPC, by the focus on fines throughout Chapter 6.  I agree that this is the relevant immediate context of section 142(1) and points towards the focus of the appeal under that provision being on the imposition of the fine.  It must also be recalled that section 115(1) expressly subjects only the decision on the imposition of a fine to Chapter 6.

13. The primary answer by LinkedIn to these points of textual interpretation is that it does not dispute that the special appeal mechanism under section 142 is focussed on, and based on, the imposition of fines.  However, it contends that the decision to impose a fine is interwoven with and indistinguishable from, the findings of infringement.  I will return to these arguments below.

Section 142(3)

1. Section 142(3) provides:

"Subject to subsections (4) and (5), the court may, on the hearing of an appeal under subsection (1)—

(a) confirm the decision the subject of the appeal,

(b) replace the decision with such other decision as the court considers just and appropriate, including a decision to impose a different fine or no fine, or

(c) annul the decision."

1. LinkedIn relies on the references in section 142(3) to " the decision " of the DPC in the singular.  I do not regard this reliance on the singular form of " decision " as a compelling aid to the interpretation of the scope of section 142.  As already seen, sections 111, 112, and 113 envisage separate decisions on infringement and corrective powers and the position of the DPC and the State - with which I agree - is that the plain language of section 142 only addresses the decision on the imposition of fines.

2. LinkedIn also argues that an appeal against a fining decision alone cannot result in no fine, unless the findings of infringement are examined.  It is said that the power to annul " the decision " only makes sense if the decision includes findings of infringement and sanctions.  It is pointed out that section 142(3) is not limited and in particular does not say that an order annulling the DPC decision or substituting it with a decision imposing no fine are only possible if the circumstances of infringement so warrant.  I will return to these submissions and the response of the DPC and the State in respect of them, when I assess LinkedIn's points about the overlap that exists between findings of infringement and fines.

Section 142(4)

1. The DPC attaches particular weight to section 142(4): " The court shall, for the purposes of subsection (3), act in accordance with Article 83." Article 83 is concerned with the conditions that govern the imposition of administrative fines.  Accordingly, a court hearing an appeal under section 142 must - without qualification or carve-out - act in accordance with the rules of the GDPR governing the imposition of fines.  According to the DPC, section 142(4) is extremely important as it applies Article 83 to the entirety of section 142(3).  It is said that the interpretation urged by LinkedIn renders section 142(4) anomalous and that this is reinforced by sections 142(5) and (6) which deal with fines as opposed to infringement.  The DPC also contends that, if section 142 was as all-inclusive as LinkedIn asserts, section 142(4) should refer in general terms to the court acting in accordance with the GDPR.

2. The reply by LinkedIn is that the reason for including section 142(4) is that Article 83 is directed to the SAs and, given the court has a broad power to replace or annul the DPC decision (including the decision to annul or replace the decision), it must also be so bound.

3. While there is of course merit to the proposition that the court must be bound by Article 83 when reviewing a decision of an SA to which Article 83 applies, Article 83 only pertains to the imposition of an administrative fine and so it is only insofar as the decision being appealed also pertains to the imposition of an administrative fine that Article 83 is activated.

4. I prefer the position advocated by the DPC that section 142(4) only makes sense if the court is engaged in an exercise of reviewing the imposition of fines.  It does not properly sit with an appellate role of reviewing findings of infringement.  This provision therefore favours the overall proposition that section 142 is concerned with, and specifically and logically addressed to, a decision as to the imposition of a fine and not other decisions of the DPC.

Heading of section 142

1. Before leaving the language of section 142, the heading of that provision may be noted: " appeal against administrative fine ". The primary position of all parties is that there is no ambiguity in section 142 such as to warrant regard being had to that wording (applying sections 5 and 7 of the Interpretation Act 2005).  However, the DPC also contends that, if there was such ambiguity the heading of that section can be considered and supports its interpretation as to the scope of section 142.  I agree that there is no ambiguity in the language of section 142: it plainly only applies to a decision to impose an administrative fine and not a decision on infringement under sections 111(1), 112(1) or 113(2)(b).  There is no basis to have regard to the heading of that provision.  If there was ambiguity in section 142, the heading would however support the interpretation already arrived at.

Section 143(1)

1. Section 143(1) was referenced by both the DPC and LinkedIn.  It provides:

"Where a controller or processor does not appeal in accordance with section 142 (1) against a decision by the Commission to impose an administrative fine on the controller or processor, the Commission shall, as soon as is practicable after the expiration of the period referred to in that subsection, and on notice to the controller or processor concerned, make an application in a summary manner to the Circuit Court for confirmation of the decision...".

1. This provision does fall to be read in light of Zalewski v. Adjudication Office [2022] 1 IR 421 (" Zalewski "). In Zalewski, the Supreme Court found that the Workplace Relations Commission was engaged in the administration of justice under Article 34, with the majority of the Court holding that the powers and functions of a judicial nature which were exercised were of a limited nature and permitted as such by Article 37 of the Constitution.  I will address the reliance placed on Zalewski later in the judgment.

2. In the context of section 143(1), LinkedIn made the case in replying submissions that the confirmation function under that provision must be broader than a review of a decision to impose a fine, to ensure that the DPC is engaged in the exercise of limited functions and powers of a judicial nature within the meaning of Article 37 of the Constitution.

3. The interpretation of section 143(1), the impact of Zalewski and the question of whether the DPC is engaged in the exercise of limited powers and functions of a judicial nature within the meaning of Article 37 are all questions that will feature heavily in the separate judicial review proceedings which are pending. I am acutely conscious of the need not to trespass on that exercise and indeed it was emphasised repeatedly in oral submissions that I am not being asked to determine these important issues here. Moreover, the interpretation of section 143(1) is of somewhat tangential relevance to the issues before me and was not argued sufficiently to be determined.  For these reasons, I do not make any findings as to the scope of section 143(1).

(c) GDPR

Article 83

Article 83 applies to appeals under section 142.  It lists the factors to which regard must be had when determining whether and in what amount to impose an administrative fine.   These include factors such as the nature, gravity and duration of the infringement; the intentional or negligent character of the infringement; any actions taken to mitigate damage; the degree of responsibility of the controller or processor.  The overarching requirement is that fines must be " effective, proportionate and dissuasive " (Article 83(1)).

1. Article 83(8) mandates that, " [t]he exercise by the supervisory authority of its powers under this Article shall be subject to appropriate procedural safeguards in accordance with Union and Member State law, including effective judicial remedy and due process. "

2. This confirms that the requirements of procedural safeguards and an effective judicial remedy apply specifically to the exercise of the power to impose administrative sanctions and that it is for Member States to determine the appropriate procedures to achieve that end.

3. The State makes the point  –  with which I agree - that the GDPR does not itself dictate how a Member State should achieve the protections required by Article 83.  Recital 143 to the GDPR confirms the autonomy of Member States on such questions:

" Proceedings against a supervisory authority should be brought before the courts of the Member State where the supervisory authority is established and should be conducted in accordance with that Member State's procedural law."

1. The GDPR is silent as to whether there should be a single decision or two decisions or a composite decision, provided the decision to impose an administrative fine meets the criteria of Article 83.  The deference to Member States in this area is further confirmed by Article 83(9), a provision cited by both the DPC and LinkedIn.  That provision permits fines to be initiated by a SA and imposed by a court, provided the requirements of equivalence and effectiveness are satisfied (which is how the GDPR has been implemented in Estonia and Denmark).

2. The DPC in turn places reliance on the specific provisions governing fines and other corrective measures in Articles 83 and 58 GDPR respectively, although the position adopted by the DPC is that the GDPR does not mandate that there can only be one decision or preclude the Oireachtas from providing for separate appeals.  It is said that the GDPR does not dictate or influence how domestic legislation should be interpreted in this respect.  As already noted, I agree with this view of the impact of the GDPR on Issue 1.

Article 60

1. There are contrasting arguments made by reference to Article 60 GDPR, which sets out the process of cooperation that must take place between SAs when a lead supervisory authority(" LSA ") is investigating a complaint involving cross-border processing.  This is the so-called " one-stop-shop " mechanism. Under Article 60, the LSA here, the DPC- is required to cooperate with concerned supervisory authorities (" CSAs ") to endeavour to reach consensus; to exchange all relevant information with the CSAs; and to submit, without delay, a draft decision to the CSAs for their opinion. CSAs may raise reasoned and relevant objections to the draft decision, which must be resolved (if necessary by the dispute resolution procedure under Article 65(1)) before the decision can be adopted by the LSA.

2. LinkedIn emphasises that Article 60 (among other provisions of the GDPR) refers to a " draft decision" and " decision " as a singular term and asserts that the GDPR mandates or envisages that there be one decision only, which should be notified in its totality to other SAs and travel through the procedural steps of Article 60.   LinkedIn identified a series of provisions of the GDPR which refer to a singular decision on which other SAs may comment under Article 60(7). This is said to support LinkedIn's interpretation of the scope of section 142 as a single appeal provision.  The DPC disputes this reliance on Article 60 and the references to " decision " generally in the GDPR, pointing out that the GDPR does not dictate how decisions of the SAs or appeals against those decisions are to be structured.

3. I note that there is no definition of " draft decision " or " decision " in Article 60 or in the GDPR and no limitation or prescription as to what a " draft decision " or " decision " should contain, whether it can consist of different decisions or segregated findings, or how it should be constituted.  Recital 129 GDPR provides some guidance as to presentation of legally binding measures, while deferring to Member States' procedural law:

" Each legally binding measure of the supervisory authority should be in writing, be clear and unambiguous, indicate the supervisory authority which has issued the measure, the date of issue of the measure, bear the signature of the head, or a member of the supervisory authority authorised by him or her, give the reasons for the measure, and refer to the right of an effective remedy. This should not preclude additional requirements pursuant to Member State procedural law. "

1. I do not accept the contention by LinkedIn that the fact that Article 60 refers in bald terms to " the decision " or " draft decision ", translates into a prescription or requirement as to how a supervisory authority is to structure its findings or that it could supplant the procedural autonomy of Member States as to how they implement the GDPR procedures and requirements.

2. On the contrary, there is a range of different permutations in which decisions may be made and different ways in which a Member State may transpose the requirements of the GDPR.  As the DPC points out, a fine may be imposed by a Member State's court as opposed to its supervisory authority, as facilitated by Article 83(9), for example, with the result that the decision on infringement could be made by the SA but the decision on the fine could only be made by the court.  Article 60(8) addresses decisions rejecting complaints, which will clearly only involve findings on infringement.  It seems plain from Article 60 that, regardless of the structure or composition of the findings and whether they are presented as individual or separate decisions or not, the key requirement is that they must all be communicated under Article 60.

3. Consequently, I agree with the State that the GDPR does not answer or even inform the questions of how a Member State can or should structure appeals against decisions of SAs.  It is a matter within the " procedural autonomy " of Member States.  The appellate structure that is in place (including whether there is a separate appeal against the decision as to the imposition of a fine) is a matter of interpretation of domestic legislation, provided the requirements of effectiveness and equivalence and the specific requirements of Article 83 as regards the imposition of fines are satisfied.  I agree with the DPC and the State in this respect.

4. Accordingly, the GDPR does not cast light on the question of whether it is only a decision on the imposition of fines that is appealable under section 142 or whether that provision also applies to an appeal against a substantive infringement decision.  This is not to say that the context of the GDPR is not relevant.  Far from it: Article 83 of the GDPR is crucial to the exercise of the power to impose fines and evidences a solid, principle-based reason as a matter of EU law for the differentiation between the imposition of fines and other measures that may be imposed by SAs.

5. In that regard, I agree with the submission by the State that Article 83 underlines that a decision and appeal on the imposition of a fine has a distinct character from an anterior finding of infringement.

(d) Conclusions on the interpretation of the statutory language

1. According to the clear terms of sections 111(2), 112(2) and 113(4), there are separate decisions as to infringement and then as to the exercise of corrective powers and, insofar as the decision is to impose an administrative fine, that decision is distinguished from other corrective powers by section 115 and stated to be " subject to Chapter 6".

2. It appears to me that the language of these provisions points firmly towards the decision to impose an administrative fine as a discrete decision, distinct from findings of infringement and from the exercise of other corrective powers.

3. This interpretation is not only consistent with the statutory language of each provision read in isolation, but is also the interpretation of the scope of section 142 that is most consistent with the cumulative reading of the relevant provisions of Chapters 2 and 6 of Part 6 of the 2018 Act.   The parties to these proceedings all recognise the very particular context of administrative fines and the need - as a matter of EU, Charter, ECHR and Irish constitutional law - to ensure effective judicial remedies and oversight when administrative fines are imposed by the DPC.  This further points towards a rationale for a differentiated treatment and regime for fining decisions alone, which is consistent with section 142 having the more focussed remit for which the DPC contends.

4. These are the conclusions I have reached from interpreting the language of sections 142 and 150, in their immediate and broader legal contexts.  LinkedIn argues against the interpretation that section 142 addresses only the decision on the imposition of fines and points to considerations which it is asserted render that interpretation absurd, unworkable and unprecedented.  As noted in Heather Hill, the onus is on LinkedIn to establish that these provisions do " not have the effect suggested by the plain meaning of the words chosen by the legislature " (at [218]). **** I will therefore deal next with the arguments of impracticability and absurdity advanced by LinkedIn.

(e) Absurdity Arguments

1. There are several different dimensions to LinkedIn's case regarding the interaction between findings as to infringement and fining.  I will address the arguments made under the following headings:

i. Overlap between infringement and fine

ii. Orders under section 142(3)

iii. Consequences of separate appeals

iv. Content of the LinkedIn Decision

(i) Overlap between infringement and fine

1. The position of LinkedIn is that a decision on infringement is inextricably linked and intertwined with a decision on the imposition of a fine; that there is only one decision and section 142 must apply to the decision as a whole.  It is said to be unstateable that the court could have the power to review and substitute a decision imposing a fine - including by deciding to impose no fine - without any power to review the infringement findings and the facts that subtend the fine.

2. Emphasis is placed on Article 83(2) GDPR, which applies to any court hearing an appeal under section 142:

"When deciding whether to impose an administrative fine and deciding on the amount of the administrative fine in each individual case due regard shall be given to the following:

(a) the nature, gravity and duration of the infringement taking into account the nature scope or purpose of the processing concerned as well as the number of data subjects affected and the level of damage suffered by them;

(b) the intentional or negligent character of the infringement;

(c) any action taken by the controller or processor to mitigate the damage suffered by data subjects;

(d) the degree of responsibility of the controller or processor taking into account technical and organisational measures implemented by them pursuant to Articles 25 and 32;

(e) any relevant previous infringements by the controller or processor;

(f) the degree of cooperation with the supervisory authority, in order to remedy the infringement and mitigate the possible adverse effects of the infringement;

(g) the categories of personal data affected by the infringement;

(h) the manner in which the infringement became known to the supervisory authority, in particular whether, and if so to what extent, the controller or processor notified the infringement;

(i) where measures referred to in Article 58(2) have previously been ordered against the controller or processor concerned with regard to the same subject-matter, compliance with those measures;

(j) adherence to approved codes of conduct pursuant to Article 40 or approved certification mechanisms pursuant to Article 42; and

(k) any other aggravating or mitigating factor applicable to the circumstances of the case, such as financial benefits gained, or losses avoided, directly or indirectly, from the infringement."

1. The position advanced by LinkedIn is that an assessment of these factors will inevitably involve examining the entire substratum, factual and legal, of the findings of infringement.  It was said that the best example of this overlap is that negligence must be shown to justify the imposition of a fine, which (it is said) involves looking at the entire premise of the findings of infringement, including how the processing was carried out, to ascertain whether it was negligent. LinkedIn contends that it would be complex, tortuous, unworkable and unprecedented for the court to have to assess the decision as to the fine without any assessment of the underlying infringement findings.

2. The DPC's position is that LinkedIn overstates the linkage between the grounds of infringement and the factors to be weighed in relation to fines.  It is asserted that in an appeal, the findings of infringement would be assessed first by reference to the materials that were before the DPC and the analysis of the fines would then proceed on the basis of the findings of infringement (which would not be re-opened).  The court may receive new material directed to the question of the fine and the factors relevant to it.  In that second stage, the relevant factors are concerned with the scale, gravity and nature of the infringement and the damage caused, which are premised on there already having been a finding of infringement and are different to the issues which arise at the infringement stage.  Counsel for the DPC outlined how the appeal would be run in practice and contended that there would be no difficulty or complexity in the exercise that would need to be undertaken. The DPC asserts that there is no impediment to a controller challenging a fine without challenging infringement and that this situation is not uncommon in civil or criminal proceedings.

3. The State similarly emphasises that it is not unusual to have separate appeals against findings of infringement and fines in other legislation and points to examples such as the Central Bank Act 1942, section 33AW in that respect.  In the course of oral submissions, counsel for the State also opened Guidelines 04/2022 issued by the European Data Protection Board on the calculation of administrative fines under the GDPR (24 May 2023).  He referred to Chapter 4 of that document, which identifies the factors relevant to the calculation of fines, including " the nature, gravity and duration of the infringement " (at [51]).  The document states that SAs are required to " carry out a complete examination of all elements that constitute the infringement... " (at [53]).  " The intentional or negligent character of the infringement " also falls to be determined when fixing the fine (at [56]).  The point made by the State was that there is no impediment to, or difficulty with, a broad review of those factors in the context of an appeal against sanction, and that this is what is envisaged by the GDPR.  It is pointed out that issues of culpability may well be raised in an appeal against sanction and could be a basis for a controller to argue for the imposition of no fine in such an appeal.

Assessment

1. The fundamental question is whether and to what extent it is workable for a court to hear and determine an appeal concerning only a decision as to the imposition of fine, without also hearing and determining the appeal about the underlying infringement.

2. I consider that it is apparent from section 142(4), which incorporates Article 83 into the analysis, that any assessment of the fines imposed will involve some assessment of features of the infringing acts, such as their gravity and scale.  It may also involve detailed assessment of the specific knowledge and conduct of the controller and processor to assess whether and to what extent the infringement was negligent or intentional.   The Guidelines opened by the State highlight the depth of the analysis of the facts surrounding the infringement that may need to be carried out.

3. However, I do not share the concern of the LinkedIn that a court could not practicably weigh such factors for the purpose of deciding an appeal on fines without deciding infringement, or that this exercise would be so complex or unworkable as to favour LinkedIn's interpretation of the scope of section 142.

4. As the State points out, courts are quite accustomed to deciding liability and quantum separately in bifurcated hearings and I agree with the DPC that the assessment of the factors in Article 83 can proceed on the basis of, and without disturbing, the underlying finding that there was infringement.

5. I would also note that the right to bring appeals against the imposition of fines under section 142 is designed to enhance the protection of controllers and processors, which the 2018 Act envisages may result in a comprehensive, detailed appeal, with the introduction of new evidence and arguments on the topic of fines alone (quite apart from the right to appeal against the findings of infringement and pursue an effective judicial remedy in that respect).  This does not mean that the two appeals cannot run in tandem in the same proceedings.  Indeed, this is how the statutory appeal was prepared (by way of alternative) in this case.

(ii) Orders under Section 143(3)

1. LinkedIn relies on the breadth and extent of the powers conferred on the court by section 142(3) in support of its construction that section 142 covers an appeal against findings of infringement.  That provision allows a court to annul or replace a decision, including by replacing it with a decision that there be no fine.  LinkedIn asserts that a court could not replace the DPC's decision with an order that there be no fine without an examination of the findings of infringement.  The DPC, by contrast, contends that the annulment or variation of a fine under section 142(3) does not impugn findings of infringement and that the orders which may be made pertain to the fine (and factors relevant to the fine) alone.

---

Assessment

1. I have already rejected LinkedIn's arguments of impracticability and found that it is consistent with the language of the 2018 Act that a court can hear an appeal against a decision to impose a fine without necessarily trespassing on findings of infringement.  This conclusion unavoidably informs the interpretation of section 142(3) also.

2. It is undeniable that, if a court weighs the factors listed in Article 83(2), as it is bound by section 142(4) to do, it will assess many aspects of the infringement, including factors of gravity and culpability.  I see no impediment to the interpretation that, having undertaken that exercise, a court could decide that it is appropriate to replace the order of the DPC with an order that there be a different or no fine.  This could, for example, arise if the court disagreed with the view of the DPC as to the evidence that the infringement was intentional, or as to the sufficiency of the steps taken to mitigate the damage caused.  Such a decision could be reached without any challenge as to whether the infringement occurred.

3. I regard the power of the court to replace the decision with a decision imposing no fine as confirmation of the far-reaching remit of the court, but I do not share LinkedIn's view that this is an order that cannot sensibly be made without examination of, or interference with, the findings of infringement.

4. LinkedIn contends that the language of section 142(3) would have to be qualified if it was intended to limit the available orders to assessment of fines, and should, for example, refer to " a decision to impose a different fine or no fine (on any ground other than there was no infringement) ". I do not agree. The scope of section 142(3) must be read in light of section 142 as a whole and the content of Chapter 6, all of which is concerned with fines and factors related to fines. I consider the language of section 142(3) to be sufficiently clear.  It is also relevant that section 142(3) limits the orders provided for by the words " subject to subsections (4) and (5) ".  Section 142(4) incorporates a requirement that the court act in accordance with Article 83 and section 142(5) deals with a fine imposed on a public authority or body.  The entirety of section 142(3) is therefore qualified by reference to distinctly fine-related provisions.

5. The DPC points to the reference in section 142(3) to the available orders " including a decision to impose a different fine or no fine."  I do not regard the word " including " and the following words as broadening the scope of available orders so extensively as to include orders replacing or annulling findings of infringement.  Indeed LinkedIn confirmed in replying submissions that such a broadened interpretation was not being urged on the basis of these words.

6. Nonetheless, it is relevant to note the submissions of the DPC on this point and its reliance on NAMA v. Commissioner for Environmental Information [2015] 4 IR 626.  The judgment of O'Donnell J. (as he then was) (Denham C.J., Murray, Hardiman and Dunne JJ. in agreement) indicates that the word " including " (or synonyms for same) does not necessarily extend the category in question:

" [36] For my own part, I would have thought that the ordinary meaning of "include" is to shut in, enclose, confine, embrace, comprise or contain and comes from the Latin root claudere meaning to close or shut. It is thus a statement that the things included are within the term or definition. Where the term or definition is very clear, it may be natural to interpret the word "include" as somehow extending that meaning, since otherwise it would be superfluous. But there are many circumstances in ordinary language where the word "include" is used for clarity, to resolve any doubt, or for emphasis. "

1. O'Donnell J. concluded: " It is clear to me that in the context in which it is used, the word "include" here was not used to extend the meaning of subparas. (a), (b) and (c), but rather to illustrate the type of thing included within the core definition."

2. I consider a similar interpretation to be apposite here.  The language of section 142(3)(b) is an important illustration of the extent - and the nature - of the orders which may be made in an appeal under section 142, being orders which are focussed on the question of the fine and which extend to the possibility of imposing no fine. This interpretation that the orders remain limited to orders concerning the fine itself (and not matters of infringement or other corrective measures) is consistent with the context of that provision, which appears in Chapter 6 - the section directed to administrative fines - and with the fact that the appeal must be determined in accordance with Article 83 GDPR.   I would also observe that it is relevant to interpreting section 142(3) that there is no reference to replacing the DPC's decision that there was infringement, or to replacing the DPC's decision to exercise other corrective powers.

3. I would finally note that it is common case from the submissions of the parties that, if the finding of infringement is annulled, it would follow as a matter of course that the fine would also fall to be annulled.

(iii) Consequences of separate appeals

1. LinkedIn's case involves the argument that it would be absurd if there needed to be two separate appeals against a single decision, with different rules and standards and a different statutory structure, applicable to each appeal.  It is contended that this is not a sensible interpretation and that there is no reason why the Oireachtas would have designed so novel and complex an appeal procedure requiring parallel appeals to be brought under different provisions, rules and structures.  The point was made in oral submissions that the siphoning off of the appeal against fines is nonsensical and unfair as it deprives the appellant of the procedures required by the Constitution and the fairness required by the Charter and the ECHR.

2. A related argument by LinkedIn is that, according to the DPC's interpretation of section 142, a court could reject an appeal against a finding of infringement under section 150, without reviewing new evidence or material, and then, having admitted new evidence under section 142(2), form a different, contradictory view on the same questions when deciding the appeal against the fine.  This is something which the Oireachtas could not have intended, according to LinkedIn.  LinkedIn argued that, if there are separate appeals required and if the standards of review are different for each and no new evidence or material is admissible in an appeal against findings of infringement, there is then a " massive structural problem in the Act ".

3. The DPC disputes these contentions.  It asserts that there is no difficulty with both sections 142 and 150 being invoked in an appeal and that it is fundamentally wrong to contend that it would mean different regimes and rules applying to different aspects of the decision.  On the contrary, the DPC's position is that the nature and standard of review is the same under both section 142 and section 150, the powers of the court are the same under both provisions, and the only difference concerns the admission of new evidence.  The State agrees with the DPC and contends that not much of significance arises from this distinction between sections 142 and 150 in practical terms, as the same standard of review and form of appeal is applicable under both.

Assessment

1. Article 83 GDPR dictates that several features of infringement must be considered in the context of fines.  Even if there is no appeal on infringement, a court hearing an appeal against the fine imposed must consider the nature of the infringement and other factors.  This is part of the protections assured to those who face administrative fines.  It is accordingly quite inescapable that there will be an overlap between findings of infringement and fines.  This is an inherent part of the judicial protection afforded as part of EU law when fines for infringements of the GDPR are at issue.  This overlap cannot be a basis for dictating how Member States structure their appeal processes: the procedural autonomy of Member States remains intact.

2. As regards the prejudice which LinkedIn apprehends arising from the bifurcation of the appeals, the case pleaded in the judicial review proceedings challenging the adequacy of the appeal mechanism and asserting denial of a right to a fair hearing under the ECHR and the Charter is dependent on the scope of the appeal under section 150 being narrower than an appeal under section 142 (amended statement of grounds at [43], [51]).

3. This appears to be premised on the apprehension that the DPC was going to argue for a more narrow form of appeal under section 150 than section 142 (as pleaded in the amended statement of grounds in the judicial review proceedings at [34.2]). This did not occur: the position of all parties is that the same form and standard of review is applicable under both sections 142 and 150 (subject to the question of admission of evidence which is addressed as Issue 4).  In fact, no party to these proceedings envisages a different standard of review applying to an appeal under section 150 and section 142.  The only substantial difference envisaged is that the DPC would limit the possible admission of new evidence and material in an appeal under section 150.

4. This really does undermine the allegations of prejudice now being made.  The distinction between section 142 and section 150 which was feared and which formed the basis of the pleaded case simply did not materialise.  The two appeals can be included in the same statutory appeal proceedings (as pleaded by way of alternative in these proceedings) and the same standard and approach will apply to both.

5. There is the important qualification - on which significant emphasis was placed by LinkedIn - that section 142(2) is available in an appeal against a fining decision, but not an appeal under section 150(5).  The admission of new evidence or arguments in an infringement appeal is addressed under the heading of Issue 4 below, but the conclusion reached there is, in summary, that there can be no presumption or blanket view taken in the abstract about the admission of new evidence or arguments in an appeal under section 150(5) (or indeed section 142).  Nonetheless, it is plain that there is a specific statutory provision governing this question in section 142(2) resulting in a possibly lighter burden to justify the admission of new argument or evidence in an appeal under section 142.

6. A related point argued in support of LinkedIn's position as to the bifurcated appeals is that it would be a major structural problem with the Act if new evidence could only be admitted on the fining decision and not on the infringement aspect of the decision.  In light of the findings made in this judgment, this apprehended structural problem with the Act does not arise.

7. I do not see that the impact of the separation of appeals in the 2018 Act is such as to justify any deviation from the interpretation of sections 142 and 150 which I consider to flow from the plain language of those provisions read in context.

(iv) Content of the LinkedIn Decision

1. LinkedIn focussed on specific paragraphs of the Decision (paragraphs 276 and 435; paragraphs 163 and 437; paragraphs 381 and 438, among several examples) which are said to illustrate the same material and factors being weighed in the context of the decision on infringement and the fining decision.  According to LinkedIn, this overlap makes it impossible to assess the fine without assessing the underlying infringement.

2. The DPC by contrast argues that the question of interpretation cannot be determined by reference to how the DPC describes or has approached the Decision.  Further, it is said that in the Decision, the analysis of infringement is separate to the analysis of the factors that are relevant to the fine, which concern the seriousness of the breach and the factors relevant to the scale of the fine imposed, such as mitigation.

Assessment

1. As pointed out by LinkedIn, there is a degree of overlap and duplication as between findings of infringement and determinations on fines in the LinkedIn Decision. The same or similar facts do feed into several aspects of the decision on infringement and the assessment of the fines.

2. Paragraphs 433 to 446 of the LinkedIn Decision (the section most heavily emphasised by LinkedIn) reflect the DPC's assessment of the nature, gravity and duration of the infringement.  It is unavoidable that considerations of that nature will overlap with the findings of infringement and some overlap certainly does exist in the Decision in the paragraphs identified by LinkedIn.    For example, paragraph 163 is directed to whether members were given information that their data would be used for advertising analytics even if they did not provide consent to personalised advertising.  Paragraph 437 -  which is said to overlap with paragraph 163 - addresses the nature of the infringements and refers to insufficient and misleading communication to data subjects.  There is some overlap, although the focus is quite different: the decision on infringement is assessing whether the failures or breaches in question occurred, whereas the later decision on the fine is a summary of the " nature " of the infringements for the purpose of deciding the appropriate sanction.

3. I am not satisfied that the overlaps are as complete or problematic as LinkedIn perceives them to be, or that they provide a good basis to determine the interpretation of section 142 or section 150.  The Oireachtas has provided for a distinct, discrete route for appealing a decision to impose a fine and incorporates Article 83 into that process via section 142(4).  It is therefore part and parcel of an appeal under section 142 that there may be an assessment of the circumstances of infringement, as distinct from the fact of infringement which falls to be assessed under section 150.  The arguments by LinkedIn do not provide a valid basis to adopt an interpretation which alters that.  Indeed, during oral submissions LinkedIn itself identified flaws in the fining decision of the DPC in this case which are independent of the grounds of infringement, including an alleged lack of analysis and a failure to properly explain the basis of the fines and how they were deemed to be proportionate and dissuasive.  Various complaints of a breach of the right to fair procedures in determining the amount of the fine are also reflected among the grounds described by Ms Duke in her grounding affidavit in these proceedings.

4. Further, at a practical level, the existence of overlaps between grounds of infringement and the fining decision may well support an argument that the admission of new evidence and material related to these points is properly within the scope of section 142(2).  This is the type of question that can be properly ventilated and determined as and when necessary in light of specific proposed new evidence or arguments weighed in the context of a specific decision.  The more extensive the overlap between the grounds of infringement and fines - including if it is so pervasive as LinkedIn contends here - the stronger the case may be for new evidence and argument to be admitted under section 142(2).  I do not see this as a difficult or absurd matter or one which should colour the correct interpretation of the scope of section 142.  Rather, it may be viewed as a means by which the Oireachtas is seeking, as an overriding priority, to ensure the protection of the rights of the appellant in an appeal under that provision, having regard to the scale of possible fines and the constitutional considerations at play.

ISSUE 2:  INTERPRETATION OF SECTION 142

(a) Preliminary

1. LinkedIn and the State advance the interpretation that the appeal under section 142 is a type of " appeal on the record " which entails a full re-hearing of the issues by reference to the record of evidence at first instance.  LinkedIn contends that this is supplemented by the facility to seek to introduce new evidence and facts and its primary position is that there is no deference to the statutory decision-maker.  The DPC, by contrast, contends that the appeal is an appeal against error in which the applicable test is whether the decision is vitiated by a significant and serious error or series of errors as set out in Orange Ltd. v. Director of Telecoms (No. 2) [2000] 4 IR 159 (" Orange")(" the Orange test ").

2. The parties seem to amalgamate the nature of the appeal, the standard of review and curial deference as composite topics.  In this regard, I note the observation of McKechnie J. in FitzGibbon v. Law Society [2015] 1 IR 516 (" FitzGibbon ") that,

"[ 58] In my view there are three broad issues in this appeal or perhaps only one, depending on what is meant by the phrase 'the scope of the appeal'. I understand it in its full sense as reflecting the nature of the review, the test involved and the format of the appeal. It is helpful to keep these matters separate if possible, but frequently they have become blurred, though understandably so in this case, given the wording of the primary relief sought, which is for a full de novo hearing. If granted, such would inevitably have covered all three aspects. However difficult it may be, a consciousness of the distinction helps an understanding of the analysis."

1. I can see merit in the segregation of topics suggested by McKechnie J., but this is not how the topics have been presented in several of the judgments cited to me or in the submissions of the parties.  Indeed, in correspondence exchanged by the parties in May 2025 (and exhibited to the grounding affidavit of Richard Willis sworn on 21 May 2025 grounding the motion for the trial of these preliminary issues), the scope of the preliminary issues was discussed.  In a letter of 9 May 2025, the solicitors for the DPC queried the reference to " standard of appeal " in LinkedIn's proposal for the trial of preliminary issues, pointing to a lack of clarity as to what it encompassed.  LinkedIn's solicitors replied on 14 May 2025 that the term " standard of appeal " referred to both the type of appeal provided for under sections 142 and 150 and the standard of review/assessment to be applied by the court on appeal.

2. During the hearing, the DPC has presented the appeal against error for which it contends as being synonymous with the Orange test and the application of curial deference.  It is said in the DPC's written submissions that " The Orange test is, in effect, the "procedural law" under which certain statutory appeals are conducted in Ireland " (at [63]).  A similar composite approach was taken by LinkedIn, advocating for " an appeal on the record ".  These classifications are based on the judgment of Clarke J. in FitzGibbon.

3. Given the parties have intertwined the topics in this manner, I will also address the nature and standard of the review, as a composite topic (although I will address curial deference discretely).

(b) Inconsistency in statutory appeals

1. Before considering the specific categories formulated by Clarke J. in FitzGibbon and the arguments of the DPC that an appeal against error applying Orange is the default approach to statutory appeals, it is relevant to observe that there has been frequent comment on, and indeed criticism of, the diverse legislative approaches to statutory appeals.  Many of the comments addressed in this section were quite fairly opened by the DPC in its submissions.

2. In Nowak v. Data Protection Commissioner [2016] 2 I.R. 585 (" Nowak ") the Supreme Court was called on to interpret the Data Protection Act 1988 which provided no guidance as to the scope or extent of the review to be undertaken by the court or the orders that may be made on appeal. This was addressed by O'Donnell J. in his judgment (with which McKechnie, Clarke, MacMenamin, Laffoy, Dunne and Charleton JJ. were in agreement):

" [28] It is a remarkable feature of legislative drafting that Acts creating independent decision makers often provide for appeal to some court in the legal system as if that was an end in itself, and without specifying the nature of that appeal. There is a wide range of possible appeals, and the decision as to what form of appeal is appropriate in any case can have a very significant impact on the length, and therefore cost, of the proceedings in court. Failure to specify what is meant by an appeal to a court can also lead to preliminary issues and the possibility of appeals. It is, in theory, possible that the legislation which provides for an appeal to court may require any of the following: a full appeal on the merits to a court; a rehearing (normally restricted to the information that was before the decision maker); an appeal by reference to the test applied by this court or the Court of Appeal in relation to appeals set out in the well known case of Hay v. O'Grady [1992] 1 I.R. 210 ; an appeal limited to a point of law; an appeal where the court is empowered to annul a decision, but not to substitute its own decision; an appeal by way of case stated; an appeal where a decision may be set aside if it was vitiated by a serious error or a series of errors; or, finally, a statutory appeal which is indistinguishable from the standard applied on judicial review. It appears that little attention is paid at the legislative level to these different types of appeals and their consequences, and courts are often left to deduce the nature of the appeal from the limited information that can be gleaned from the language used, the structure of the Act, and, sometimes, the subject matter of the decision."

1. The comments of O'Donnell J. that a lack of legislative specificity in this area can result in preliminary issues resonate here. ** This question is also addressed by the Law Reform Commission in its important consultation paper on Reform of Non-Court Adjudicative Bodies and Appeals to Courts (LRC. CP 70-2025) (" the LRC Paper "), in which it is observed that " the multiplicity of appeal mechanisms creates inefficiencies and adds significant complexity to the process, burdening not only applicants without specialist knowledge, but also legal professionals and the judiciary " (at [7-132]).

2. Different approaches can be seen to this issue of consistency in statutory appeals.  Finnegan P. in Ulster Bank Investment Funds Limited v. Financial Services Ombudsman [2006] IEHC 323 (" Ulster Bank v. FSO ") considered the interpretation of section 57CL of the Central Bank Act 1942, the relevant wording being:

" (1) If dissatisfied with a finding of the Financial Services Ombudsman, the complainant or the regulated financial service provider concerned may appeal to the High Court against the finding.

(2) The Financial Services Ombudsman can be made a party to an appeal under this section..."

1. Section 57CM then confirmed the powers of the High Court, being the broad power to " make such orders as it thinks appropriate in light of its determination ", which includes but is not limited to an order affirming, modifying, setting aside or remitting the finding or directions of the Ombudsman.  These were undoubtedly broad powers.

2. Having addressed Orange, M&J Gleeson & Co v. The Competition Authority [1999] 1 ILRM 401(" M&J Gleeson "), Carrickdale Hotel Limited v Controller of Patents [2004] 3 I.R. 41 (" Carrickdale ") and Dunne v. Minister for Fisheries [1984] I.R. 230 (" Dunne "), Finnegan P. concluded that the test in Orange should be applied, noting:

" It is desirable that there should be consistency in the Courts in the standard of review on statutory appeals. Accordingly unless the words of the statute mandate otherwise it is appropriate that the standard of review in this case be that enunciated by Keane C.J., Kearns J. and Laffoy J. I see nothing in the wording of the statute with which I am concerned to mandate a different approach to the statutory appeal under the Central Bank Act 1942 section 57CL. . "

1. This was a decision that favoured consistency and leaned towards a form of presumption that the standard of review should be Orange unless the words of the statute mandate otherwise.

2. In FitzGibbon, McKechnie J. summarised various authorities, culminating the passage from the judgment of Finnegan P. in Ulster Bank v FSO just quoted and continued:

" [70] With the utmost respect I am not at all sure that 'consistency' should be the yardstick in determining what the relevant test should be, as deserving in all as that end result obviously is. Essentially what the court is involved in is an exercise in statutory interpretation which must be assessed in accordance with the relevant and well established canons of construction. If the results should differ as between one appeal provision and another, the approach cannot be altered simply to avoid this. If inconsistency, as to the nature and format of appeals should arise, it will do so only because of the variable statutory framework in place. This is not caused by the judicial approach but is a direct consequence of legislative terminology, which undoubtedly causes difficulties for the judiciary."

1. McKechnie J. concluded with a clear caution against excessive reliance on the need for consistency between statutory appeal codes, " even where similarity exists between statutory provisions of different codes":

"As Hogan and Morgan point out in Administrative Law in Ireland (4th ed., Round Hall, Dublin, 2010): '[f]actors [other than similarity] such as the purpose of the appeal and the statutory context are more relevant' (para. 11–119, at p. 549). Consequently one should resist the temptation of achieving uniformity, as desirable as this may be, unless, inter alia, the context so permits and clearly so provides" (at [74]).

1. The Supreme Court in Kelly v. Information Commissioner [2017] IESC 64 again expressed concerns about the uncertainty and instability surrounding statutory appeals.  McKechnie J (with O'Donnell, Clarke, Laffoy and Dunne J.J. in agreement) addressed this as follows:

" [54] .. It is no exaggeration to say that the resulting instability has often led to complex and costly litigation simply to scope out the remit of the provision in question, this without ever embarking on the substantive analysis. Not surprisingly, this situation has attracted a good deal of judicial comment, if not criticism, for its lack of uniformity and coherency, thus giving rise to much uncertainty as to the extent and reach of individual provisions."

1. He continued by calling for "greater urgency" in the introduction of reforms to introduce structure and reduce the level of current "preliminary scrutiny" of appeal provisions, noting that Kelly itself was " a good example of why this remains a major and indeed pressing issue of concern." The Court discusses " several variable forms of 'appeal' ", noting that in every case the statute must be construed and that each legislative framework must be construed " in its own right " (at [59]).

2. The appeal under section 42(1) of the Freedom of Information Act 1997 that was at issue in Kelly v. Information Commissioner itself was an appeal on a point of law only, and the debate centred on aspects of the appeal mechanism that are not relevant here.  Nonetheless the observations by McKechnie J. (with which the other four members of the Supreme Court agreed) that the key question is the interpretation of the words in the statute in the context in which they are used are universally relevant.

3. Returning to the topic of achieving consistency, Faherty J. observed in Xerico Ltd. v Residential Tenancies Board [2025] IECA 202 (" Xerico "),

" ...the fact that an appeal against error is the norm in the regulatory context cannot of course exclude the requisite statutory interpretation exercise by this Court. As put by O'Malley J. in Petecel v. The Minister for Social Protection [2020] IESC 25 , at para. 106, "[q]uestions as to the scope of the jurisdiction of the High Court in a statutory appeal must always be answered by reference to the terms of the statute creating that jurisdiction.  In Fitzgibbon, McKechnie J warned against the temptation to seek to achieve uniformity with other statutes, unless "the context so permits and clearly so provides". Similarly, in Pecetel. at para. 106, O'Malley J. opined that "[t]he fact that one Act uses similar terminology to another may be helpful but is not necessarily determinative." " ** (at [56]).

1. As will be seen, Orange was not adopted as the applicable test in Xerico.

2. There is some apparent divergence between the decided cases as to whether diverse legislative appeal mechanisms can or should be interpreted so as to achieve consistency among them.   I note that the DPC does rely in its supplemental written submissions on statements that a full rehearing is " exceedingly rare " (LRC Paper at [7.95]) and on the statement by Hogan & Morgan that a statutory appeal is an appeal against error " on most occasions ", and posits the case that it is relevant to interpreting the 2018 Act.  However, I think some circumspection is required.  The reference to full rehearings being " exceedingly rare " in the LRC Paper, for example, is in the context of a discussion about appeals de novo from the District Court to the Circuit Court.  When Hogan & Morgan discuss what occurs " on most occasions " this is a qualified statement:  " One might venture to suggest that on most occasions in which the Oireachtas provides for an "appeal" from a specialised first-instance body to a generalist appellate decisionmaker, the natural inference is that an "appeal against error" is intended..."  (at [11-59]).

3. This is not to dispute that an appeal against error has often been found to be the form of statutory appeal intended by the Oireachtas or that this is part of the relevant context for the assessment of the 2018 Act.  The point is that it cannot detract from the central task of interpreting the words of the  specific statute, a point well made by Hogan & Morgan: " Determining the scope of an appeal will invariably require a close analysis of the text, surrounding context and animating purpose of the relevant statutory scheme. Similarity of statutory wording cannot be determinative of the scope of the appeal " (at [11-64).

4. In short, regardless of the most common or typical approach in other statutory appeal regimes, the key question in each case must be the correct interpretation of the specific statutory provisions being applied.  Accordingly, while some guidance can be drawn from the approach taken to the interpretation of legislation addressing other statutory appeal mechanisms, I do not favour the view expressed by the DPC that there is a presumption that section 142 provides for an appeal against error based on the Orange test.  I prefer the position of LinkedIn and the State that there is no such presumption.  Accordingly, the question to ask is not that posited by the DPC in written submissions -  " whether there is any reason why Orange should not apply "– but rather how sections 142 and 150 should properly be interpreted in light of the guidance of the Supreme Court in Heather Hill.

(c) Categorising appeals

1. As the parties have presented the task of interpretation on the basis of the categories identified in FitzGibbon, it is necessary to understand what the parties mean by the terms " appeal on the record " and " appeal against error ".

2. In FitzGibbon, a solicitor appealed against a decision of the Law Society to sanction him.  The High Court (Kearns P.) held as a preliminary issue that the appeal should not be a de novo hearing but rather a review in which oral evidence would be called only if necessary.  On appeal to the Supreme Court, the solicitor argued against curial deference being afforded to the Law Society and contended that the appeal should be in the form of a de novo hearing with oral evidence.  The Law Society's position was that the legislature could have, but did not, provide for a hearing de novo in section 11 of the Solicitors (Amendment) Act 1994 which governed the procedure, and that oral evidence should be the exception.  Section 11 provided:

"A solicitor in respect of whom a determination or direction has been made or given by the Society under section 8 (1), 9 (1) or 12 (1) of this Act or who has received a notice for production or delivery of documents from the Society under section 10 (1) of this Act may, within a period of 21 days of the notification of such determination or direction to him, or the receipt of such notice by him, apply to the High Court for an order directing the Society to rescind or to vary such determination or direction, or to vary or withdraw such notice, and on hearing such application the Court may make such order as it thinks fit."

1. The three judgments of Denham C.J., McKechnie and Clarke J.J. all agreed that the appeal should be dismissed, with Clarke J. agreeing with the reasoning of Denham C.J.  In his judgment, Clarke J. addressed the different categories of statutory appeals, explaining - significantly - that **

" t hese comments are offered in the hope that they may be taken on board by those who are charged, whether in the public or private spheres, with drafting rules or legislation (whether primary or secondary) which provide for the possibility of an appeal " (at [93]).

1. In the interests of guiding those charged with drafting legislation, Clarke J. then addresses four categories of statutory appeals, noting that hybrid versions are also possible.  The categories described are de novo appeals; appeals on the record; appeals against error and appeals on a point of law.  The descriptions of these categories are addressed further in the next section of this judgment.

2. Hogan, Morgan & Daly observed that the approach by Clarke J. was an " analytically rigorous attempt to set out a typology of appeals is a useful framework for discussion " (at 11-50).  The authors also observed, significantly, that " it should be borne in mind that the Oireachtas will not necessarily legislate by reference to these different types, and there may thus be variations within these types " (at [11-51]).

3. There is a wide variety of additional and overlapping ways a statutory appeal may be characterised.  O'Donnell J. (as he then was) noted in Nowak,

" It is, in theory, possible that the legislation which provides for an appeal to court may require any of the following: a full appeal on the merits to a court; a rehearing (normally restricted to the information that was before the decision maker); an appeal by reference to the test applied by this court or the Court of Appeal in relation to appeals set out in the well known case of Hay v. O'Grady [1992] 1 I.R. 210 ; an appeal limited to a point of law; an appeal where the court is empowered to annul a decision, but not to substitute its own decision; an appeal by way of case stated; an appeal where a decision may be set aside if it was vitiated by a serious error or a series of errors; or, finally, a statutory appeal which is indistinguishable from the standard applied on judicial review " (at [28]).

1. McKechnie J. made similar observations in FitzGibbon as to the breadth of the range of possible forms of statutory appeals.

2. While the categories devised in FitzGibbon are a useful framework and could form a basis for statutory appeals to be enacted by the Oireachtas, this does not alter the fact that there are other ways appeals can be categorised and, as Faherty J. observed in Xerico, " the appellations "appeal on the record", "appeal against error" and "hybrid appeal" are adjectival descriptions by the courts and not found in any statute " (at [19]).

3. In weighing different appeal regimes and their interpretation, it is also important to bear in mind the clear admonition in Heather Hill that it is not the role of a court to select between competing policy objectives in interpreting legislation, without an objective justification for the decision (at [220]).  The court's view as to the form of appeal that is most efficient, practicable, preferred or even typical cannot be the basis for interpreting the scope of appeal under section 142.

4. In short, many different forms of existing appeal mechanisms and further diverse forms of appeal could yet be introduced by the Oireachtas.  There have been some recent examples of the enactment of legislation that does closely emulate the categories in FitzGibbon.  A recent judgment in Central Bank of Ireland v. CD [2026] IEHC 203 addressed a provision of the Central Bank Reform Act 2010 (introduced in 2023) which incorporated the Orange test, as confirmed by Barniville P: " [t]The formulation contained in s. 45(6)(a) ('any significant and serious error or series of such errors'), is taken from Orange Limited v Director of Telecoms (No. 2) [2000] 4 IR 159 (per Keane CJ at pp. 184 - 185) " (at [259]).  The LRC Paper highlights the Competition (Amendment) Act 2022 and the Communications Regulation and Digital Hub Development Agency (Amendment) Act 2023 which appear to enshrine in statute the form of appeal for which the DPC here contends.  That LRC Paper was the subject of helpful supplemental written submissions by the parties in late March 2026.  The DPC contended that the express incorporation of the Orange test in the legislation cited by the LRC could be seen as an endorsement of that test, and that the silence in the 2018 Act on this point does not mean that Orange does not apply.

5. It is of course the case that the Orange test has been found to apply to appeals under legislation which does not include such express language, but the legislation just cited is relevant here, as it demonstrates that the Oireachtas can and does legislate in such clear terms when Orange is the intended standard of review.  It did not do so here.

(d) Understanding the categories

1. There is no consensus between the parties as to whether or how the categorisation in FitzGibbon applies to the 2018 Act.  The position of LinkedIn is that section 142 does not (and is not required to) neatly fit into any of the categories identified in FitzGibbon.  The State regards the appeals under the 2018 Act as a form of hybrid of those categories.  The DPC maintains that the appeals are squarely within the ambit of appeals against error applying Orange.

2. My overriding view is that it is not apparent that the drafters of the 2018 Act did intend to, or did, legislate on the basis of the taxonomy in FitzGibbon. Nonetheless, as the parties have presented their positions in these proceedings by reference to those categories, it is very important to establish what they understand those categories to mean.

3. The position of the DPC is relatively clear: the appeal should be an appeal against error in which the Orange standard is applied and curial deference is afforded to the DPC (although there are nuances to the position on curial deference, which are addressed under that heading).  There is however a marked divergence between the parties as to what an " appeal on the record " entails. This has become particularly striking during the exchange of supplemental written submissions.

4. The " appeal on the record " is described as follows by Clarke J. in FitzGibbon:

" the appellate body must come to its own conclusions as to the proper result of the issues before it without regard to the decision made by the first instance body. However, and in contrast to a de novo appeal, the default position in respect of an appeal on the record is that the evidence and materials which are properly relied on by the appellate body are the same as those which were before the first instance body " (at [111]).

193. Clarke J. then proceeded to note that,

" there may be some limitations on the proper availability, at least in its pure form, of an appeal on the record. Those limitations stem from the possibility that, in respect of certain types of issues, it may not be legitimate for an appeal body to reach an independent conclusion in relation to such issues solely by considering the first instance record and without having regard to the findings of the first instance body " (at [113]).

1. This passage introduces the concept of an appeal on the record " in its pure form " and, in it, Clarke J. acknowledges that on certain types of issues, it may be necessary to have regard to the decision reached by the statutory body.  He then identified an additional circumstance in which there may be a difficulty in conducting an appeal purely on the record:

" The rules of constitutional justice require that, where a decision on contested facts is necessary to enable a decision maker to adjudicate on the rights and obligations which require to be determined, a party potentially affected by a finding of fact on such contested matters, is, amongst other things, entitled to test (both by cross-examination and by the presentation of competing evidence) the evidence which might lead to such an adverse finding " (at [114]).

1. While this is not applicable to every appeal, Clarke J. observed that when such rights are engaged, " an appeal purely on the record " may not be available (at [116]).  These passages from the judgment are notable here, as they show that, in devising the category of "appeal on the record", Clarke J. contemplated a possible distinction between " an appeal on the record " and " an appeal purely on the record ".

2. I n its written submissions, LinkedIn addresses its view on Issue 2 as follows:

" For the reasons outlined above, the right of appeal under section 142 of the 2018 Act 2018 must be interpreted such that the Court hearing the appeal be entitled to reach its own conclusions on all issues of law and fact. The said right of appeal does not (and is not required to) fit neatly within the broad categories identified by the Supreme Court in FitzGibbon. However, when section 142 of the 2018 Act 2018 is interpreted in light of its wording, the ECHR, the Charter and the Constitution, the type of appeal envisaged thereby can be described as a type of "appeal on the record" wherein:

85.1 The Court has full jurisdiction to quash in all respects, on questions of fact and law, the decision of the DPC.

85.2 The Court is not obliged to defer to the decision of the DPC and has full jurisdiction to examine all questions of fact and law relevant to the dispute before it.

85.3 Evidence and arguments which were not before the DPC in the Inquiry can be advanced in support of the appeal (section 142(2) of the 2018 Act 2018).

85.4 The Court is conferred with jurisdiction to replace the Decision with such other decision as the Court considers just and appropriate, including a decision to impose a different fine or no fine (section 142(3) of the 2018 Act 2018)."

1. LinkedIn therefore signalled that the form of " appeal on the record " contended for did not fit neatly into that category as described in FitzGibbon, but was rather a variation which met four specific criteria.  The only real point of contention from among those criteria concerns deference (there being no dispute but that an appeal under section 142 will have the first, third and fourth features identified by LinkedIn).

2. In its post-hearing supplemental submissions (at [7]), the DPC emphasises the comment by Clarke J. that "[i]n an appeal on the record, the appellate body must come to its own conclusions as to the proper result of the issues before it without regard to the decision made by the first instance body " (FitzGibbon at [111])  The DPC also contends (at [11]) " Under an 'appeal on the record' standard, the Courts would have to undertake assessments against the "full factual backdrop", including conducting any balancing exercise themselves. This would be so even if there was no evident error in the DPC's assessment."  This characterises an appeal on the record as one in which no regard is had to the DPC decision and a full balancing against the full facts is necessary even if there are " evident " or " clear errors ".

3. LinkedIn disputes this characterisation of an appeal on the record, submitting that the DPC's submissions " significantly exaggerate " the burden on the court in an appeal on the record.  LinkedIn further contends that, in an appeal on the record, the issues in the appeal are necessarily defined by the terms of the decision under appeal and the pleadings.  Reliance is placed on FitzGibbon where the court referenced the potential relevance of the underlying decision in an appeal de novo (at [103]).  LinkedIn asserts " That analysis applies mutatis mutandis to an appeal on the record ".

4. What emerges is a different understanding between the parties as to what an " appeal on the record " entails.  This is not only a divergence that is apparent in this case.  I would note the description of an " appeal on the record " in Hogan & Morgan, Administrative Law in Ireland (5 th Ed., 2019) at [11-55] to 11-56:

"11-55 An appeal on the record also seems to be what the Electronic Communications Appeal Panel had in mind in Hutchinson 3G Ireland Ltd v Commission for Communications Regulation:

'It seems therefore that what is envisaged by the Regulations (as interpreted in the light of Article 4.1 [of the Framework Directive]) 169 is an examination of the decision of the Regulator as opposed to a reassessment de novo by the Panel of whether the Appellant is a 'significant market power in the wholesale voice call termination market on individual mobile networks.' This will mean that the Panel can focus on evidence and materials upon which the Regulator based its decision and look at the inferences and conclusions it drew from those materials.'

11-56 The Panel was of the view, therefore, that the standard as envisaged by the Regulations was broader than the more deferential standard associated (as we shall see) with appeals against error.

'The Regulations envisage that the Panel can annul, or annul in part, the decision of the Regulator if, on an examination of the Regulator's decision and the decision-making process, there are errors of fact or law (which includes erroneous inferences of fact, or errors as to jurisdiction and procedure) such as would vitiate the decision. In practical terms, therefore, the error (or errors) would need to be of significance before the Panel could annul the decision, in whole or in part. It should not be a trivial error. It should be one which, when objectively assessed, had a bearing on the decision reached by the Regulator. However, the error(s) need not go to the root of the decision either. Rather, the error(s) should be material in the sense that they are objectively relevant to, and have a bearing on, the conclusion the Regulator came to. In coming to this conclusion, the Panel will take into account the view of the Regulator given its expertise on certain technical matters, but ultimately can substitute its own opinion if it takes a different view in respect of these matters'. "

1. It is important that the appeal described in Hutchinson 3G Ireland v. ComRe g (ECAP Decision No. 01/05, Appeal No. 2004/01, 10 February 2005) was regarded by Hogan & Morgan as an appeal on the record.  The LRC Paper also noted a view that Hutchinson 3G Ireland v. ComReg seemed to be an appeal on the record (at [7.21]).

2. When this view of appeals on the record is weighed, such an appeal includes an assessment of the underlying decision and the inferences and conclusions drawn from the materials before the decision maker, to test whether there were errors of fact or law, which were material, more than trivial errors, without necessarily going to the root of the decision.  An appeal on the record, on this interpretation, provides for the expert view of the decision maker on technical matters to be taken into account, but a different view can be taken and the opinions on technical matters may be substituted if erroneous in a material respect.  The key difference with such an appeal on the record is that the standard is broader than the more deferential standard in an appeal against error (according to Hogan & Morgan).

3. These accounts of appeals on the record underline the range of different views that are available on what that form of appeal entails.  This in turn highlights the danger of adopting general labels to seek to convey a multitude of different facets of a statutory appeal process, without breaking down, testing or clarifying what is intended, and without any real basis to believe that these various facets are to be found in the statutory scheme in fact enacted by the Oireachtas.

4. In this case, it seems relatively clear from the submissions made by LinkedIn that the form of " appeal on the record " which is in fact argued for is not squarely within the category described in FitzGibbon and is not an appeal on the record " in its pure form ".  In addition to the question of admitting new evidence, it was presented (by way of an alternative, secondary proposition) that deference may be available on certain issues in an appeal under the 2018 Act.  The form of " an appeal on the record " contended for appears to resemble the form of appeal addressed in Hutchinson 3G Ireland v. ComReg.

5. Indeed, further categorisation challenges can be seen in the DPC's contention that LinkedIn collapses the distinction between a de novo appeal and an appeal on the record (owing to the argument for admission of new evidence on appeal).   To assess this argument, it is necessary to understand what an appeal de novo means.  What Clarke J. said in describing this category of appeal was the following:

" It seems to me that the critical characteristics of a de novo appeal are two fold. First, the decision taken by the first instance body against whose decision an appeal is brought is wholly irrelevant. Second, the appeal body is required to come to its own conclusions on the evidence and materials properly available to it. The evidence and materials which were properly before the first instance body are not automatically properly before the appeal body. It seems to me that, by defining an appeal as a de novo appeal, any legally effective instrument necessarily carries with it those two requirements " (at [102]).

1. He further explained the scope of the evidence in a de novo appeal as follows:

"... the default position, in the absence of any specific rule to the contrary, must be that, in the case of a de novo appeal, it remains for the parties to again present to the appellate body whatever evidence or materials may be considered necessary for their case. Likewise, if, and to the extent that, the process may be inquisitorial, then, again in the absence of rules to the contrary, the inquisitorial process must start afresh before the appellate body " (at [104]).

1. He explained that, even in an appeal de novo, the proceedings that went before are " not ...necessarily entirely irrelevant " (at [103]).

2. I reject the contention that LinkedIn is collapsing the distinction between appeals on the record and de novo appeals for three reasons.  First, it appears to me that, even if new evidence is admitted, that is not to say that the primary evidence being canvassed in the appeal would not remain the evidence that was before the DPC.  I have seen no suggestion that there would be a fresh exercise of collation of all relevant evidence and the pleadings and affidavits to date in these proceedings are certainly not prepared that way.  Second, unlike an appeal de novo (such as from the Circuit Court to the High Court or that considered in Barry v. Medical Council [1997] IEHC 204), there has been no suggestion here that the burden will not rest on the controller or processor appealing the DPC's decision to identify and demonstrate the grounds of appeal.   On the contrary, the DPC points out in its written submissions that the appellant bears the burden in the appeal.  The pleadings in these proceedings make plain how an appeal of this nature will run - regardless of which of the parties prevails in this judgment - and this is by way of an appeal in which the moving party bears the burden of justifying a departure from the decision of the DPC.   Third, as LinkedIn points out, the LinkedIn Decision sets the parameters of the appeal.  It seems clear that the appeal will be run on the basis of, and by way of challenge to, the LinkedIn Decision.  Indeed, the right of appeal in section 142 is wholly premised on " the decision " and the orders that may be made are orders to confirm, replace or annul " the decision ".  No form of appeal under that provision could be heard or determined wholly without regard to the underlying decision.  A similar point was made in an authority cited by LinkedIn, Competition and Markets Authority v. Flynn Pharma [2020] 4 All ER 934 (" Flynn "). There the UK Court of Appeal considered the scope of the appeal from a decision of the UK Competition and Markets Authority to the Competition Appeals Tribunal.  The Court noted that, while Article 6(1) ECHR was applicable,

"... the jurisdiction of the Tribunal is not unfettered.This flows primarily from the fact that the appeal is not a de novo hearing but takes the decision as its starting, middle and end point. Under s 46 CA 1998 the appeal is 'against, or with respect to,' the decision and includes 'whether' there has been an infringement. That focus upon the impugned decision is reflected in the procedural rules of the Tribunal. The appellant must identify the decision under appeal and set out why it is in error " (at [141]).

1. The same analysis is appropriate here.  It confirms there can be no question of the appeal under section 142 being an appeal de novo, or an appeal in which no regard is had to the LinkedIn Decision.  It does not however, assist in deciding whether the appeal is otherwise in the form preferred by LinkedIn or that advocated by the DPC.

2. Insofar as the DPC contends that LinkedIn is effectively pursuing an appeal de novo, I do not believe that point is well made and certainly do not regard section 142 as intending that form of appeal.  In assessing the features of the forms of appeal devised in FitzGibbon and invoked by the parties here, it is also necessary to understand the " appeal against error " for which the DPC contends.  This is described by Clarke J. as an appeal in which " the appellate body does have regard to the determination of the first instance body and must, in order for the appeal to be allowed, be satisfied that the first instance body was in some way in error " (at [119]).

3. Clarke J. assessed the form of " error " which must be shown in " an appeal against error ", noting the Orange test before continuing as follows:

" Relevant rules may also provide for the precise basis on which the appellate body is to assess the question of error. However, in the absence of any specific rules, the default position will be that an appeal body should not interfere with findings of fact at first instance unless there was no sustainable basis for the finding in question or where the finding was clearly in error. Likewise, in the absence of a specific rule defining the nature of the appeal, an appellate body should be free to exercise its own judgment as to whether other findings of the first instance body are in error. In the absence of express rules qualifying the extent to which the appellate body should come to its own independent view on such matters, there may well be cases where a limitation can properly be implied because of the nature of the issues to be determined and the expertise of the respective first instance and appellate bodies " (at [123]).

1. Based on this account of an appeal against error, the default setting is that a court is free to exercise its own judgment as to whether there was an error by the statutory body, subject only to the qualifications that it should not interfere with findings of fact unless there is no sustainable basis for those findings or they are " clearly in error " and there may be some " limitation " when warranted by the nature of the issues and the expertise of the statutory body.  Although he referred to it, Clarke J. did not adopt Orange as the applicable test or standard in an " appeal against error ".  I deal with deference separately below, but pause here to note that the question of deference on findings of fact must now be read in light of Stanberry Investments Limited v. Commissioner of Valuation [2020] IECA 33 ** (at [49]).   The form of " appeal against error " presented by the DPC is squarely and entirely based on Orange as the applicable test to the entirety of the appeal and all grounds.  It is therefore a somewhat hybrid version of that proposed by Clarke J. in FitzGibbon.

2. The parties did present the preliminary issues here on the basis of a choice between an appeal against error (applying Orange) and an appeal on the record.  However, having assessed the formulation of those categories in FitzGibbon and the arguments advanced here, it seems to me that the real choice presented here is between a hybrid form of appeal against error in which new evidence may be admissible and Orange is the test for all grounds of appeal, and a hybrid form of appeal on the record in which new evidence may be admissible and there is no scope for curial deference.

(e) Interpretation of section 142

1. I will address the interpretation of section 142 under the following headings:

(i) Language of section 142

(ii) Positions of the parties

(iii) Comparison with other legislation

(iv) GDPR, Charter and ECHR

(v) Conclusions

(i) Language of section 142

1. The starting point must be the language of the section itself.  As Murray J. (O'Donnell C.J., O'Malley, Woulfe, Hogan J.J. agreeing) observed in Heather Hill,

" [218] The words of the section are the first port of call in its interpretation, and while the court must construe those words having regard to the context of the section of the Act in which the section appears, the pre-existing relevant legal framework and the object of the legislation insofar as discernible, the onus is on those contending that a statutory provision does not have the effect suggested by the plain meaning of the words chosen by the legislature to establish this."

**

**

1. Section 142 provides,

"(1) Without prejudice to section 150 , a controller or processor that is the subject of a decision under section 111 , 112 , 113 or 133 (9) to impose an administrative fine may, within 28 days from the date on which notice of the decision concerned was given to it under section 116 or, as the case may be, section 133 (9)(b) appeal to the court against the decision.

(2) The court, on hearing an appeal under subsection (1), may consider any evidence adduced or argument made by the controller or processor concerned, whether or not already adduced or made to an authorised officer or the Commission.

(3) Subject to subsections (4) and (5), the court may, on the hearing of an appeal under subsection (1)—

(a) confirm the decision the subject of the appeal,

(b) replace the decision with such other decision as the court considers just and appropriate, including a decision to impose a different fine or no fine, or

(c) annul the decision."

1. While the power to appeal to the court is sparsely stated, the powers conferred on the court are important.  In particular, LinkedIn and the State attach some weight to the formulation of the power in section 142(3)(b) to " replace the decision with such other decision as the court considers just and appropriate, including a decision to impose a different fine or no fine..."

2. On its terms, this provision confers a very broad and specific power on a court to determine what is a " just and appropriate " decision and to replace that for the DPC's decision. If the Oireachtas intended to legislate for a highly deferential form of appeal, this could have been said here, but the formulation is notably different.

3. There is an additional provision in section 142(2) that the court may consider evidence or arguments, whether or not they were presented to the DPC.  This is an undoubtedly important provision and indicates that the appeal may involve a hearing of additional evidence and argument beyond that which was before the DPC.   I will return to discuss the question of deference, but observe here that the decision of the Oireachtas to expressly provide for the facility for a court to hear evidence that was not heard by the DPC and on which the views of the DPC are not known, is not naturally reconcilable with curial deference being afforded to the DPC's decision.

4. Section 142 should be read in its immediate statutory context.  The section of the 2018 Act in which section 142 appears is Part 6.  This provides for the conduct of an investigation in respect of a complaint of breaches of the GDPR.  The rights specifically conferred on a controller or processor during that process are section 137, which confers a right to respond within 7 days to a notice issued by an investigating authorised officer setting out particulars of the complaint and section 139, which confers a right to make submissions within 28 days of receipt of a draft investigation report.

5. There are also extensive provisions governing the powers of the DPC or an authorised officer to obtain information.  This includes the power to conduct an oral hearing.  Section 138(10) 2018 Act specifically provides, "[f[or the purposes of an investigation, an authorised officer may, if he or she thinks it proper to do so, of his or her own volition conduct an oral hearing.  A power to conduct an oral hearing following receipt of an investigation report was also conferred on the DPC by section 140 of the 2018 Act.  This power may be exercised if the DPC is of the view that further information is required (section 140(1)).  These provisions of Part 6 comprise the immediate statutory context within which section 142 must be assessed.

(ii) Positions of the parties

1. LinkedIn contends that section 142 provides for an appeal on the record entailing a full review of all facts and law and with the court being entitled to reach its own conclusions on all such issues without deferring to the DPC.  It is said that the wording of section 142 is not compatible with an appeal against error. In support of this interpretation, emphasis is placed on the fact that there is no provision for remittal of the matter to the DPC and on the fact that the court may substitute the decision of the DPC with its own decision, which is characterised as a complete invocation of the court's powers.  It is said that the wording of the Act is crucial and particular reliance is placed on Xerico. The position of LinkedIn is that - like in Xerico - the investigative and decision-making process here is truncated, with no oral hearing, and the processor not having sight of the draft decision before it is sent to the supervisory authorities in other states and only limited opportunities to make submissions on the fining amount in the final draft decision.  It is said that the power to substitute or replace the decision of the DPC has no analogue in other cases, such as Nowak.

2. The position of the DPC is that the appeal provided for is an appeal against error to the Orange standard of a serious and significant error.  The DPC contends that all issues of fact and law may be ventilated, and the appeal is not limited to points of law.  It is said that an appeal against error does provide for such an examination of all questions of law and fact and that a narrow or strict view of the appeal provisions is not being taken.  The DPC contends that the authority on which LinkedIn relies heavily - Xerico – can be distinguished, the key difference being the opportunity under the 2018 Act to be heard and participate in the process preceding the impugned decision.

3. The position of the State is that the law is lacking clarity in this respect, but that the better interpretation of the 2018 Act is that an appeal under s 142 is an appeal on the record, allowing for a full re-hearing of the issues by reference to the record of evidence at first instance, with the facility to introduce new evidence and facts.  According to the State, this is primarily a question of domestic statutory interpretation and important indicators of the nature of the appeal envisaged by section 142 are the orders a court can make (pointing to Rye v. Competition Authority [2012] IESC 52 (" Rye "); the fact that there is no reference in the 2018 Act to an appeal on a point of law or an error on a point of law; the absence of any possibility of remittal; the power of the court to substitute its own decision for the DPC; and the unlimited jurisdiction to admit new evidence.  It is contended that these features of the appeal under section 142 indicate something more than appeal against error with Orange as a threshold test. Xerico and Barry v. Medical Council are both relied on as close comparators.  It is also said there is nothing in the 2018 Act to suggest the High Court does not have full appellate jurisdiction.

(iii) Comparison with other legislation

1. In order to interpret section 142, it is helpful to consider comparisons and contrasts with other statutory appeal regimes.  As already observed, this should neither be taken too far nor allowed to detract from the primary task, which is the interpretation of the specific statutory language of the 2018 Act in context.

2. These cautions notwithstanding, it is inevitable that comparisons with other legislation and their interpretation are relied upon.  In this section of the judgment, I will address the most relevant Irish judgments, all of which were helpfully opened by the parties, with a focus on judgments concerning appeal provisions that are asserted to bear some resemblance to sections 142 and 150.  I do not consider every judgment included in the books of authorities and exclude from my analysis, for example, cases interpreting legislation which expressly provides for appeals on a point of law (such as **** Rotunda Hospital v. Information Commissioner **** [2013] 1 **** I.R. 1; Minister for Communications, Energy and Natural Resources v Information Commissioner **** [2022] 1 **** I.R. 1) and cases interpreting legislation that provides for a de novo appeal (Barry v. Medical Council [2007] IEHC 74).

3. Approaching the cases cited by the parties chronologically, the first is M&J Gleeson.  That case centred on section 9 of the Competition Act 1991:

"(1) Any undertaking or association of undertakings concerned, or any other person aggrieved by a licence or a certificate of the Authority under section 4 (2) or (4), or the Minister, may appeal to the High Court within 28 days of publication pursuant to this Act of the licence or certificate and on the hearing of any such appeal the Court may confirm, amend or revoke the licence so appealed against, or, in the case of such a certificate, may cancel or refuse to cancel the certificate....

(4) Any undertaking or association of undertakings concerned or the Minister may within 28 days of the notification or publication pursuant to this Act of a revocation or amendment by the Authority, pursuant to section 8 (3), of a licence under section 4 (2), or of a revocation by the Authority, pursuant to section 8 (6), of a certificate under section 4 (4), appeal to the High Court.

1. The plaintiffs, drinks wholesalers, appealed a decision of the Competition Authority to grant a licence to Guinness Ireland Group Ltd. in relation to an agreement to acquire a 100% stake in a drinks distribution company on condition that it reduce its stake in, and rights in relation to, another drinks distribution company.

2. Kearns J. emphasised the specialist and qualified personnel at the Authority and the lengthy process of consultation and consideration over many months that preceded the publication of the challenged decision.  Notably, that process did include oral submissions and meetings.  Kearns J. immediately noted that the appeal provided for in section 9 is not limited to an appeal on a point of law but also does not specify that it is a re-hearing.  The parties agreed that the standard for judicial review proceedings was not appropriate.  Instead, the focus - based on the submissions made by the applicant - was on whether the word " appeal " should be given a wide interpretation and amount to an appeal de novo, involving a full re-hearing of the entire matter, embarking on exactly the same exercise that had been undertaken by the Authority afresh.  It was however agreed by the parties that the evidence received by the Authority should be treated as evidence in the appeal and the findings of primary fact by the Authority were " inviolate " and would not be challenged in that appeal.  Kearns J. observed that what the applicants were contending for was in fact " something less than a de novo hearing and something more than an appeal on a point of law " (at page 409).

3. As regards the standard of review, Kearns J. relied on and adopted the decision of the Canadian Supreme Court in Canada (Director of Investigation and Research) v. Southam Inc. [1997] 1 SCR 748 which points to " a posture more deferential than exacting ".  The threshold applied by the Canadian Supreme Court in that case was one of " reasonableness " or what the Court regarded as the similar test of " clearly wrong ".   The Canadian Supreme Court held as follows:

"... an appeal from a decision of an expert tribunal is not exactly like an appeal from a decision of a trial court. Presumably if Parliament entrusts a certain matter to a tribunal and not (initially at least) to the courts, it is because the tribunal enjoys some advantage that judges do not. For that reason alone, review of the decision of a tribunal should often be of a standard more deferential than correctness ...

I conclude that the ... standard should be whether the decision of the Tribunal is unreasonable. This is to be distinguished from the most deferential standard of review, which requires courts to consider whether a tribunal's decision is patently unreasonable. An unreasonable decision is one that, in the main, is not supported by any reasons that can stand up to a somewhat probing examination. Accordingly, a court reviewing a conclusion on the reasonableness standard must look to see whether any reasons support it..."

Kearns J. concluded " I think that is the correct standard in so far as it makes reasonableness simpliciter the critical standard. In essence this is the test which the Supreme Court in Ireland adopts by way of standard, with the added dimension of curial deference which suggests a standard in this context more deferential than exacting."  He concluded on this point that, "[t]he court must be satisfied that the authority's decision lacks a reasonable basis. " M&J Gleeson therefore endorses a " reasonableness " standard.

The point which the DPC seeks to derive from M&J Gleeson is that new evidence can be admitted in an appeal in which the Orange standard is being applied. M&J Gleeson in fact pre-dates Orange and applies a predominantly " reasonableness " standard overlain with considerations of deference.   It is not quite " Orange ", although, as will be seen, the two have been treated as the same consistent test since (see Carrickdale at [30]).  It may be noted that the legislation applied in M&J Gleeson conferred on the court the power to amend a licence (albeit not a certificate) and there was no power of remittal.  However, I do accept the proposition that the possibility of admitting new evidence does not preclude Orange being the applicable standard of review.

Turning then to Orange itself, this was an appeal under s. 111(2B)(i) of the Postal and Telecommunications Services Act 1983 (as amended) from a decision of the Director of Telecommunications Regulation awarding a licence for mobile telephony. Th e issue in the appeal was a decision by the Director to refuse to grant the third mobile telephone licence in the State to the plaintiff, having ranked another applicant for the licence, Meteor Mobile Communications Ltd., first in the competition for the licence. The statutory provision governing the appeal gave effect to a requirement of Commission Directive 90/388/E.E.C. that there be an appeal from a decision of the national regulatory authority (i.e. the Director) to an institution independent of that authority.

Section 111(2B)(i) provided:

"A person may, within 28 days of the receipt by him or her of a notification under paragraph (h), appeal to the High Court against the decision concerned (not being a decision to amend a term or condition of the licence concerned) and the High Court may confirm the decision or direct the Minister, as may be appropriate, to refrain from granting, revoking or suspending the licence concerned, and the Minister shall comply with a direction under this subparagraph and shall not implement the decision unless and until it is appropriate to do so having regard to the outcome of the appeal."

An important point about this provision is that the court does not have the power to substitute the Minister's decision. There is no other indication of the nature of the appeal and no rules of court or other legislative guidance as to its scope.

In Orange the parties had agreed that it was not necessary for the plaintiff to demonstrate that the decision was so manifestly unreasonable as to be contrary to common sense, so Keane C.J. approached the matter " on that basis " (page 184), subject to the qualification that curial deference was necessary, given the lesser knowledge and expertise available to the court than to the expert statutory body.

Keane C.J. concluded that the appeal was not intended to be a " re-examination from the beginning of the merits of the decision appealed from culminating, it may be, in the substitution by the High Court of its adjudication for that of the first defendant " (at page 184).  This is an important confirmation of the need to weigh the powers of the court and specifically points to the significance of a court having - or not having - the power to substitute its decision for that of the statutory body.  This has an obvious relevance when I turn later to the interpretation of the wording of section 142 of the 2018 Act.  Keane C.J. proceeded to observe that the appeal was not solely to judicial review grounds. Rather, based on the specific provisions of the relevant act, the decision could be set aside if was shown

"... as a matter of probability that, taking the adjudicative process as a whole, the decision reached was vitiated by a serious and significant error or a series of such errors.  In arriving at a conclusion on that issue, the High Court will necessarily have regard to the degree of expertise and specialised knowledge available to the first defendant " (at page 185).

1. This test therefore incorporates deference to the decision maker, specifically envisaging that the judge hearing the appeal will assess the degree of specialised expertise and knowledge held when " arriving at a conclusion " as to whether there was " a serious and significant error ".  In the case before him, Keane C.J. assessed whether the High Court judge was correct in finding the first defendant did not act reasonably in various respects and found that none of the alleged grounds of unreasonableness were made out.

2. Proceeding chronologically, the next judgment on which the DPC relies is Carrickdale. That arose from a dispute between certain owners of nightclubs and discos and a collective organisation of record companies (the second defendant).  An appeal was brought by the plaintiff (a nightclub owner) against an award made by an arbitrator who was appointed by the Controller of Patents to fix the amounts payable to the second defendant by way of equitable remuneration, pursuant to section 17(4)(b) of the Copyright Act 1963.  There was a hearing before the arbitrator and a transcript of that hearing was available to the court.  The appeal was governed by section 41(3), which provides,

" An appeal shall lie to the High Court from any award made by an arbitrator in pursuance of a reference under this section to which the parties to the dispute did not consent and the High Court may make such order confirming, annulling or varying the award of the arbitrator as it thinks fit."

1. During the hearing of the appeal before Laffoy J (after much of the appeal had been opened by the plaintiff), an issue arose as regards the standard of review and the parties addressed the court on that topic.  Neither party contended that the appeal should be a full de novo appeal or that the applicable standard was that applicable in a judicial review and there was no suggestion that the test was whether the award was so manifestly unreasonable as to be contrary to common sense.  Instead, the position of the second defendant was that the standard should be " reasonableness simpliciter " (applying M. & J. Gleeson), whereas the plaintiff contended that it should be an appeal by way of " rehearing " akin to appeals to the Supreme Court under Order 58, Rule 1 of the RSC (as it then was).

2. Rejecting the case for an appeal by way " rehearing ", Laffoy J. reasoned,

" It seems to me that what the modern jurisprudence of the Supreme Court enjoins this court to do is, first, to look at the legislative code and, secondly, to have regard to the policy adumbrated by Hamilton C.J. in Henry Denny & Sons (Ireland) Limited v. Minister for Social Welfare [1998] 1 IR 34 " (at [29]).

1. Laffoy J. went on to conclude that there was " very little difference in substance " between the provisions governing the appeal in Orange and section 41 of the Copyright Act 1963 and applied the Orange test to the appeal before her, noting that she saw " no essential difference between the test as so expressed and the test as posited by Kearns J. in M. & J. Gleeson v. Competition Authority [1999] 1 I.L.R.M. 401 " (at [30]).

2. It is a point of some note that the legislation in both Carrickdale and in Orange contained concise appeal provisions which confer a bare right to appeal and vest the court with the power to confirm, annul or vary the award (Carrickdale) or to confirm the decision or direct the Minister to refrain from implementing it (Orange). The DPC contends that this is an example of a wide appeal jurisdiction, similar to the 2018 Act, in which Orange was found to apply . However, it is undeniable that some of the important features of section 142 are not present in Carrickdale, particularly the power to replace or substitute the decision of the statutory body.

3. Rye v. Competition Authority is an interesting case on which reliance was placed by the DPC.  The relevant appeal mechanism was contained in section 24 of the Competition Act 2002, which provides in relevant part:

"(1) An appeal may be made to the High Court against a determination of the Authority under paragraph (b) or (c) of section 22 (3)....

(4) Any issue of fact or law concerning the determination concerned may be the subject of an appeal under this section but, with respect to an issue of fact, the High Court, on the hearing of the appeal, may not receive evidence by way of testimony of any witness and shall presume, unless it considers it unreasonable to do so, that any matters accepted or found to be fact by the Authority in exercising the relevant powers under section 22 were correctly so accepted or found.

(5) Notwithstanding subsection (4), the High Court, on the hearing of an appeal under this section, may receive evidence by way of the testimony of one or more witnesses if it considers it was unreasonable for the Authority to have accepted or found as a fact any matter concerned.

(6) Without limiting the exercise of the judicial function with respect to a particular case, it shall be the duty of the High Court, in so far as it is practicable, to hear and determine an appeal under this section within 2 months after the date on which the appeal is made to it.

(7) On the hearing of an appeal under this section, the High Court may, as it thinks fit—

a. annul the determination concerned,

b. confirm the determination concerned, or

c. confirm the determination concerned subject to such modifications of it as the court determines and specifies in its decision...

(9) An appeal to the Supreme Court against a decision of the High Court under any of the foregoing provisions of this section shall lie only on a question of law."

1. There are some quite distinctive features of that provision, including the bar on the High Court receiving new oral evidence and the mandatory presumption that any facts accepted or found by the Authority were correctly accepted or found (unless this is found to be unreasonable).  The DPC emphasises that, like here, there was no power of remittal under the Competition Act 2002 and Orange was nonetheless applied and new evidence was admissible on appeal.

2. In the course of his judgment, Cooke J noted that an appeal under section 24 of the 2002 Act was not a judicial review or an appeal on a point of law, but rather an appeal in which " any issue of law or fact " concerning the determination may be raised.  Adapting a finding by Costello J. in Dunne v. Minister for Fisheries, Cooke J. noted that " the appeal may raise both the question 'is it lawful or unlawful?' and the question 'is it right or wrong?' "

3. However, he also emphasised that it was "not expressly an appeal by way of rehearing of the original notification in which the decision of the court fully replaces that of the Authority ".  This was underscored in his judgment by the duty to hear and determine the appeal within 2 months (section 24(6)), which he regarded as "a considerable restraint upon the scope of an appeal."

4. Further, according to Cooke J., the most significant indicator of the limited form of the appeal was section 24(4), which, " requires the court to presume, unless it is considered unreasonable to do so, that matters of fact have been correctly accepted or found by the Authority in its determination."  Cooke J. explained, " the court is required not to re-open or interfere with findings of fact so long as that presumption stands, thereby limiting in a significant degree the scope of the procedure as an appeal directed at the correctness of the determination on its merits, as opposed to one confined to its substantive and procedural legality. " ** Cooke J. went on to contrast this aspect of section 24 from other appeal mechanisms under the Competition Act 2002 which allow a court to " confirm, amend or annul " a declaration without a presumption in favour of the Authority equivalent to section 24(4) (specifically section 15).  He concluded, "[t]hus, while clearly wider in scope than a judicial review as to legality, the appeal under s. 24 is clearly narrower than that available under s. 15. In addition, of course, the decision on the appeal under s. 24 is subject to the narrow, although not absolute, time constraint of ss. (6). "

5. I would pause to observe that the 2018 Act contains no equivalent to section 24(4) (the presumption that ** " matters accepted or found to be fact by the Authority " were correctly found) or section 24(6) (two month limit for hearing and determination of appeal), both of which were central to the assessment of the nature of the appeal in Rye.

6. Cooke J. then analysed the standard of review, focussing on the question of how wrong the Authority's decision must be, for it to be interfered with. ** The question he identified was,

" if it is open to the notifying parties to seek to prove that errors have been made which render the determination wrong, how wrong must it be if it is to be annulled by the court? What test of error is to be applied having regard, in particular, to the deference which is due to the specialist expertise of the Authority inherent in the presumption in its favour on matters of fact in s. 24(4)?"

1. It is plain from the judgment that heavy weight was attached to the presumption in section 24(4) in determining the standard and scope of the appeal.  The appellant argued for " manifest error " as applied by the European Courts in reviewing equivalent decisions of the European Commission under the Competition Rules of the Treaty, whereas the Authority advocated for Orange to be applied.  Cooke J. favoured the latter, finding as follows:

"5.14 While, as indicated above, the primary consideration in all of these cases is the correct construction of the relevant statutory provisions, the Court agrees that the standard of deference indicated by Keane C.J. in the Orange case is appropriate to be applied to the present statutory appeal. Although the terms of that statutory appeal were different and the administrative context from which the appeal arose was quite distinct (an appeal against a refusal by the Director of Telecommunications Regulation to grant a third mobile telephone licence in the State), and which (sic) the appeal was not expressly limited by a presumption equivalent to that contained in s. 24(4) in the present case, the judicial task required does not appear to be materially different to that required here."

1. He further reasoned,

" 5.15 Subject to giving correct effect to the precise terms in which the statutory appeal in each case is expressed, it is obviously desirable as a matter of policy to ensure that the criteria upon which an appeal is based remain consistent with other closely analogous statutory appeals under domestic legislation and also, in the area of competition law, with the Authority's obligation to ensure that its decisions are consistent with Community law where they have a Community dimension. "

1. The firm emphasis on the precise terms of the statutory appeal as expressed in legislation is also significant.  While applying Orange, Cooke J ultimately concluded that there were material errors in the Authority's decision including in the assessment of the relevant product market and that these errors vitiated the finding of substantial lessening of competition in each of the three product markets upon which the negative determination was based.  The determination was annulled.  Given the particular legislative provisions at issue in Rye and the findings made, I do not regard Rye as an authority that points towards the application of Orange as the correct test here.

2. Nowak is another case on which particular reliance is placed by the DPC. Mr Nowak was a trainee accountant who had failed accountancy exams set by the Institute of Chartered Accountants in Ireland. And was refused access to his examination script. The DPC refused to investigate Mr Nowak's complaint on the basis the script was not " personal data " and the complaint was frivolous and vexatious. The Circuit Court ruled that no appeal lay against that decision, a ruling that was upheld by the High Court and Court of Appeal.  Leave was granted to appeal to the Supreme Court in respect of two issues, namely whether the Court of Appeal erred in law in holding that Mr Nowak was not entitled to appeal to the Circuit Court and in holding that the DPC was entitled to conclude that the script was not personal data. The Supreme Court ultimately adjourning the proceedings, making a preliminary reference to the Court of Justice of the European Union on the question of whether an examination script constituted " personal data ".  The Court did find that the conclusion that the complaint was frivolous or vexatious was a decision which could be appealed to the Circuit Court under section 26 of the Data Protection Act 1988.

3. Section 26(1)(d) governed the appeal and provided simply that,

ii. An appeal may be made to and heard and determined by the Court against—

iii. ... a decision of the Commissioner in relation to a complaint under section 10 (1) (a) of this Act,

iv. and such an appeal shall be brought within 21 days from the service on the person concerned of the relevant notice or, as the case may be, the receipt by such person of the notification of the relevant refusal or decision."

1. Section 26(3)(b) of the Data Protection Act 1988 further provided,

"An appeal may be brought to the High Court on a point of law against such a decision; and references in this Act to the determination of an appeal shall be construed as including references to the determination of any such appeal to the High Court and of any appeal from the decision of that Court."

1. In the course of his judgment in that case, O'Donnell J. (at [30]) identifies several possible forms of statutory appeals ranging from a full de novo hearing to an appeal on a point of law only.  Interestingly, one option referenced by him is " an appeal where the court is empowered to annul a decision, but not to substitute its own decision. "  This does point to the potential importance of a court being conferred with the power to substitute its decision for that of the decision-maker in a statutory appeal when characterising the type of appeal intended by the Oireachtas.

2. In Nowak, the DPC contended for the application of Orange, whereas the appellant contended for a full rehearing of his appeal, relying on Dunne v. Minister for Fisheries. O'Donnell J. noted that in Dunne the court held that, while it was an appeal on the merits, the appeal court would normally be limited to the information before the decision maker and that " [a]ccordingly, even this option is something less than a full rehearing as occurs, for example, in the High Court on a Circuit Court appeal." (at [29]). **

3. O'Donnell J. observed:

" Both of these cases have their place. Dunne v. Minister for Fisheries is perhaps an example of an older style of appeal from a decision maker, in that case the Minister for Fisheries, who might have no greater knowledge or expertise than the High Court judge. Since the decision was one to be made by a minister who, as distinct from his or her department, might have no particular expertise in the area, it was not, perhaps, implausible that a High Court judge would be empowered to make a decision on the merits on the material presented to the Minister. The important feature supplied by appeal to a court is the prospect of review by a third party with guaranteed independence." (at [29]).

1. The points made about Dunne are important: the form of appeal in that case could be said to be justified by both the absence of particular specialist expertise on the part of the Minister and the objective of ensuring independence in the appeal process. Dunne was then contrasted with Orange:

"Orange Ltd. v. Director of Telecoms (No. 2) is an example of the more modern trend where a significant number of decisions on areas of some complexity are now made by statutorily independent decision makers who, moreover, may be selected for appointment because of, and in any event may have developed, very considerable technical expertise. The purpose of court review in such a case may be different " (at [29]).

1. Considerable emphasis was placed on paragraph 30 of the judgment of O'Donnell J., particularly by the DPC:

"... courts, while perhaps having no expertise in the underlying area, do have considerable experience both in decision making and in review of decision making and reasoning processes. On the other hand, even t he greatest admirer of courts might think it unlikely that individual courts could, in the course of a single case, develop the type of technical expertise acquired by, and available to, specialist bodies in a complex area, and in any event, might reasonably doubt that adversarial litigation is the most effective or cost efficient way of educating a judge on technical issues to the point where he or she could, with confidence, substitute his or her decision on a technical issue for that of the original decision maker. This functional analysis perhaps supports the test identified in Orange Ltd. v. Director of Telecoms (No. 2) [2000] 4 IR 159 : a court can be expected to detect errors of law, and may identify serious errors in reasoning or approach. It can be said that if an error is sufficiently clear and serious to be detectable by a non-expert court after scrutiny, then that is justification for overturning the decision, even though the court may lack more specific expertise. In my view, the standard in Orange Ltd. v. Director of Telecoms (No. 2) is the appropriate standard to apply here. As it happens, I do not believe this issue has much, if any, impact on the substance of the appellant's appeal, since the issue he raises is essentially an issue of law: it involves the application of a legal test to facts which are not significantly in dispute. However, since the matter is of general importance, I would hold that the Circuit Court is not required to allow a full appeal on the merits, or the narrower appeal permitted in Dunne v. Minister for Fisheries [1984] I.R. 230 . Instead, the court should apply the test in Orange Ltd. v. Director of Telecoms (No. 2) as outlined above. I would, however, emphasise that the argument here proceeded on the basis that the only options were the type of appeal in Dunne v. Minister for Fisheries, or the more limited form of review contemplated in Orange Ltd. v. Director of Telecoms (No. 2). No argument was addressed to the formulation of the test in Orange Ltd. v. Director of Telecoms (No. 2), which may yet arise in an appropriate case " (at [30]).

1. According to this passage, the appeal governed by the 1988 Act was neither a full appeal on the merits (a de novo appeal) nor the narrower appeal in Dunne but was " the more limited form of review contemplated in Orange ".  I note that O'Donnell J. does not refer to or adopt the categorisation advanced by Clarke J. in FitzGibbon.  On the contrary, O'Donnell J. compares the test in Orange with the question of the " type of appeal " in Dunne, signaling that the question of the applicable test or standard is itself a means of categorising the type of appeal provided for.  This is relevant to assessing the approach taken by the parties to the case before me, which similarly blends the typology of the appeal with the test to be applied.  As noted, this is the approach I have also taken in this judgment.  There are a number of additional points that may be noted about this important paragraph from the judgment of O'Donnell J.

2. First, as observed by O'Donnell J., the question of the applicable standard of review or form of appeal did not require determination in that case, the issue in the appeal being one of law.  This is not in dispute. While O'Donnell J. addressed the scope of the appeal under section 26, he did so because it was a matter of " general importance " not because it was necessary to decide the appeal.

3. Second, the consensus of the Supreme Court on this question is not clear.  Clarke J. echoed the concerns of O'Donnell J. regarding the lack of specificity in legislation providing for statutory appeals to the courts, while refraining from deciding the point in Nowak on the basis it was not necessary or appropriate to do so:

"[41].. Precisely where the test in Orange Ltd. v. Director of Telecoms (No. 2) [2000] 4 IR 159 , in whatever formulation, might fit into a more coherent structure (if at all) is a matter which may need to be considered in that context. However, it does not appear to me that this case is the appropriate vehicle to attempt so to do.

[42] For the reasons already noted it does not seem to me that it is, strictly speaking, necessary to decide such questions in order to resolve this case. I would prefer to leave those issues to a case in which same might prove decisive."

1. McKechnie, MacMenamin, Laffoy, Dunne and Charleton JJ. agreed with both O'Donnell J. and Clarke J., leaving it somewhat unclear as to whether they favoured the view that Orange was applicable, or the view that it was preferable not to decide that point.

2. Third, the parties in Nowak only presented a choice between Dunne or Orange.  There was no other position presented or apparent discussion about the formulation of those forms of statutory appeals, prompting O'Donnell J. to observe that the formulation of Orange " may yet arise in an appropriate case ".   The judgment may point towards a variation of the Orange test and its broadening beyond the strict version of that test advanced by the DPC here.

3. Fourth, the description of the test in Orange in the judgment of O'Donnell J. is interesting: " if an error is sufficiently clear and serious to be detectable by a non-expert court after scrutiny, then that is justification for overturning the decision..." This suggests that a decision can be overturned (even on a technical point) provided there is an error which is " detectable " by a non-expert court.  It also necessarily envisages the possibility of scrutiny of technical issues by the non-expert court.  Turning then to the position of the parties here: the DPC contends that the judgment of O'Donnell J. reflects, and has since been treated as, the decision of the Supreme Court on the scope of an appeal under section 26. To anchor this point, the DPC's written submissions rely on Peter Nowak v Data Protection Commissioner [2025] IECA 195 at [35]; Agnieszka Nowak v Data Protection Commissioner [2022] IECA 95 at [41] and Xerico at [48].

4. In the DPC's second supplemental submissions, it is further said that **

" It is difficult to gainsay that the emerging consensus in the case law and the analysis is that it is indeed O'Donnell J's judgment in Nowak *** (adopting Orange), rather than that of Clarke J., which has been adopted and endorsed as representing the law on the applicable standard for appeals under the Data Protection Act, 1988.  Contrary to LinkedIn's apparent position, all of that (in particular the Court of Appeal authority), should not be simply ignored by this Court* " (at [14]).

1. LinkedIn contends in its supplemental written submissions that Nowak **** was not an unambiguous approval of Orange **** by the Supreme Court and further points to textual differences between section 26 of the 1988 Act and sections 142 and 150 of the 2018 Act. The State in its written submissions contends that Nowak is of limited utility and cannot safely be regarded as an endorsement of Orange.

2. It certainly cannot be gainsaid that there are three Court of Appeal judgments in which the Supreme Court in Nowak has been referred to as endorsing Orange as the applicable standard in appeals under section 26 of the 1988 Act.  It does not however appear that the point was contested or argued in any of those three cases.  In Xerico, the interpretation of section 26 was not in issue (that case, as will be seen, concerned an appeal from the Residential Tenancies Board).  The interpretation of that section was noted, although not a point of appeal or disagreement, in an appeal under the 1988 Act in Peter Nowak v. DPC (at [35]). Similarly, in Agnieszka v. DPC, Faherty J. noted that the principles were agreed at the hearing to be those as set out in Orange and in Nowak.

3. As the applicable standard does not appear to have been argued or disputed in any of the cases cited by the DPC and neither the context nor the judgments in Nowak were considered, it is difficult to regard these Court of Appeal judgments themselves as decisive as to the ratio of the Supreme Court in Nowak. I agree with the supplemental submissions of LinkedIn in this respect.  However, I am not persuaded that it is necessary for me to decide this somewhat contentious point.

4. There can be no doubting the importance of the statements in the judgment of O'Donnell J. that a formulation of Orange did and should govern appeals under section 26 or the weight that should be attached to those statements on that specific point.  This does not however answer the more difficult question of the correct interpretation of section 142 and section 150 of the 2018 Act.  In weighing this question, the first point to note is of course the different statutory language introduced in the 2018 Act.  One particular feature of section 142 and section 150 that stands out when Nowak is considered is the power of the appellate court to substitute or replace the decision of the DPC with its own.

5. In Nowak, O'Donnell J. had referenced doubts about the efficiency or feasibility of an appeal which would require a court to obtain sufficient technical knowledge to be able to confidently substitute his decision on a technical issue for the decision originally made. This was part of the reasoning for preferring the Orange test rather than the approach in Dunne v. Minister for Fisheries. The judgment underlines the importance of the question of substitution.  Significantly, in Nowak itself, the court had no express power to substitute its decision for that of the decision-maker under the 1988 Act.

6. When the Oireachtas subsequently came to enact the 2018 Act, it elected by sections 142(3)(b) and 150(6)(b), to confer on the courts an express unqualified power to substitute or replace their decision for that of the DPC, without any reservation as to the technicality or otherwise of the decision or findings which may be so substituted. Section 142(3)(b) provides that the court may " replace the decision with such other decision as the court considers just and appropriate, including a decision to impose a different fine or no fine..." and section 150(6)(b) empowers the court to " substitute its own determination for the decision.."

7. The prudence of this approach and its divergence from other statutory appeal mechanisms may be questioned or criticised, but the task I face is one of statutory interpretation and that question of interpretation is materially different now to the question that was addressed in Nowak.

8. One of the judgments relied upon by O'Donnell J. - Dunne - was not specifically relied on by the parties before me, but I note for completeness that Costello J. there also emphasised the importance of the power of the court to substitute its opinion for that of the decision maker:

" When the Court is given by statute an appellate jurisdiction to confirm or annul a bye-law or other administrative act, it must construe the words used by the legislature to see whether the Court has power to substitute its own opinion for that of the administrative authority if, on the evidence before it, the Court considers that the impugned act was wrong on the merits and not merely wrong in law."

1. Costello J. proceeded to analyse the applicable legislation (primarily section 11 of the Fisheries (Consolidation) Act 1959), noting that the appeal was wider than an appeal on a point of law; the Court had the power to confirm or annul a bye-law; the appellant may not have had an opportunity to present a case to the Minister; and " to be effective " the court may have to allow and hear new evidence. Costello J. concluded that the court could reach a different view on the merits to that of the Minister, while noting that " the Court should be slow to substitute its own opinion for the opinion of the Minister in cases in which his department's experience and knowledge of the matter in issue would be an important element in reaching its formulation."  I would observe that there was no express power of substitution in section 11.

2. The second significant point of difference between Nowak and this case is the introduction in the 2018 Act of the power of the DPC to impose potentially sizable administrative fines.  In interpreting the scope of the appeal provided for in section 142, this new dimension to the legislation brings with it a series of important additional considerations and rights (as addressed further below).  These are firmly reflected in section 142 itself and are matters to which weight must be attached in the interpretation of that provision.   The legal landscape for assessing an appeal under the 1988 Act is simply different when compared to an appeal against the imposition of fines in the amount of €310 million by the DPC, as here.

3. Against this background, the final judgment which the parties consider to be of note is Xerico This judgment of the Court of Appeal addressed the scope of section 148AA of the Residential Tenancies Act 2004 (as inserted by section 28 of the Residential Tenancies (Amendment) Act 2019), which provides:

"(1) A landlord the subject of a decision under subsection (4)(a) or (5) of section 148X by the decision maker to impose a sanction on the landlord may, not later than 21 days from the giving by the Board to the landlord of the decision notice, appeal to the Circuit Court against the decision.

(2) The Circuit Court may, on the hearing of an appeal under subsection (1) by a landlord, consider any evidence adduced or argument made, whether or not adduced or made to an authorised officer or the decision maker.

(3) Subject to subsection (4), the Circuit Court may, on the hearing of an appeal under subsection (1) by a landlord—

(a)either—

(i) confirm the decision the subject of the appeal,

(ii)  set aside the decision the subject of the appeal, or

(iii) set aside that decision and replace it with such other decision as the Court considers appropriate, which may be a decision—

1. to impose a different sanction on the landlord, or

2. to impose no sanction on the landlord, and

(b) make such order as to costs as it thinks fit in respect of the appeal.

(4) The Circuit Court shall, for the purposes of subparagraph (i) or (iii)(I) of subsection (3)(a), take into consideration the matters referred to in section 148AD.

(5) An appeal under subsection (1) shall be brought before a judge of the Circuit Court for the time being assigned to the Circuit in which the appellant resides or carries on any profession, trade or business."

1. The Court of Appeal in Xerico addressed the following preliminary issues:

"(a) What is the proper scope of the appellate jurisdiction of the Circuit Court on a statutory appeal under section 148AA of the Residential Tenancies Act 2004?

(b) If the appeal is an appeal against error, in what circumstances may the Circuit Court consider evidence that was not before the Board?"

1. The appellant initially contended for a de novo appeal but ultimately pursued a "hybrid" appeal consisting of an appeal "on the record" with the possibility of additional evidence being adduced on appeal.  The appellant relied on Rye and contended that in enacting section 148AA(1) the Oireachtas envisaged something much more than an appeal against error particularly given that section 144AA(3) of the 2004 Act allows the Circuit Court to substitute its own view for that of the decision-maker.  Particular emphasis was placed by the appellant on the power of the Circuit Court to " set aside ... and replace " a decision in its entirety with a decision which the Circuit Court " considers appropriate " and on the power to call evidence on appeal.  Another factor emphasised by the appellant was the absence of a power of remittal, which was said to be " a fairly strong indicator " (albeit not conclusive) that an appeal other than an appeal against error was intended.

2. The RTB contended for an appeal against error.  The position of the RTB was that, because the legislation provided for the admission of new evidence, it was not consistent with an appeal on the record.  The RTB also contended that the facility to introduce new evidence was not determinative of the appeal not being an appeal against error (reliance was placed on M&J Gleeson, FitzGibbon, Rye Investments, Ulster Bank v. FSO and Carrickdale as relevant comparators).  The absence of a power of remittal was also said not to point against an appeal against error (relying on Carrickdale, Rye Investments and FitzGibbon). The same points are made by the DPC here.  The RTB relied heavily on its contention that the Act conferred extensive investigative and evidence-gathering powers on the investigator and conferred on parties the right to request an oral hearing.  The sharing of the draft investigation report with the landlord, the affording of an opportunity to make submissions on it and the receipt of such submissions, were also emphasised by the RTB.  As regards the relevance of participation in the process, heavy reliance was placed by RTB on the judgment of McKechnie J. in FitzGibbon.

3. Faherty J. (with whom the other members of the Court agreed), identified the distinction between the two forms of appeal as follows: "... in an appeal against error the task of the court is to look for error on the part of the decision maker, whereas in an appeal on the record, the task of the court is to make up its own mind 'as to the proper result of the issues before it'" (at [28]).

4. The Court approached the matter on the basis that, as Costello J. noted in Dunne v Minister for Fisheries [1984] I.R. 230, " in every case the statute in question must be construed " (quoted by Faherty J. at [50]).  Having analysed the cases already considered in this judgment (among others) including Rye Investments, Barry v. Medical Council, Orange, M&J Gleeson, Carrickdale, Nowak, Ulster Bank v. FSO and FitzGibbon, the Court assessed the specific factors emphasised by the parties.  Faherty J. attached some weight to the fact that the power of the court under section 148AA(2) was a broad power to admit " any evidence " adduced or argument made whether or not before the decision-maker and contrasted cases such as Ulster Bank v. FSO.

5. When assessing the engagement during the investigation and decision-making process, Faherty J. noted that during the second, decision-making stage of the bifurcated process provided for in the 2004 Act, the landlord was not in fact given the opportunity to make submissions on the final investigation report or to attend an oral hearing.  Both were provided for as possibilities in section 148X(6), although the landlord emphasised that it was a matter at the election of the decision-maker under that provision.  The RTB in reply on this point contended that the legislation should be interpreted objectively and the nature and scope of the appeal should not be determined according to the procedures adopted in a particular case.

6. The Court reasoned that the scope of the appeal provided for in the Act must be determined by reference to all of the procedural routes for which it provides.   The approach taken by the decision-maker was entirely consistent with the Act but did not allow the landlord any opportunity to obtain the final investigation report or make submissions to the decision maker.  The Court found that, as the decision maker had a discretion as to whether to allow the opportunity to make submissions or conduct an oral hearing, s.148AA (1) " requires to be interpreted as empowering the Circuit Court on appeal to arrive at a different decision to that of the decision maker, and not just to review whether the decision is vitiated by a serious or significant error or a series of such errors "  (at [91]).  Allowing an appeal against error only would, in the view of Faherty J., " be extraordinarily restrictive."  The conclusion reached was as follows:

"... the manner in which s. 148X of the 2004 Act is framed, coupled with the nature of the remedies provided for in s. 148AA(3) (which includes provision for the Circuit Court to substitute its own adjudication for that of the decision maker something which Cooke J. in Rye Investments described as "of primary importance in determining the scope of a statutory appeal"), are, to my mind, compelling indicators that the appeal provided for in s. 148AA(1) falls to be construed as an appeal on the record which is supplemented by the express statutory provision in s. 148AA(2) for the adducing of additional evidence " (at [97]).

1. This finding is notable here, as it reflects a combination of an appeal on the record with the possible admission of new evidence.  Its emphasis on the power of the court to substitute its decision for that of the RTB is also significant.

2. What emerges from Xerico is the importance of assessing the particular legislative provisions at issue including the statutory decision-making regime that precedes the appeal.  In that case, there were various indicators that pointed towards an appeal on the record (particularly the absence of a power of remittal and the conferring on the court of a power to substitute the decision-maker's view with its own).  The Court also emphasised heavily the absence of any guarantee or right under the applicable legislation to make submissions, participate in an oral hearing, or have sight of the final decision-making report during the process overseen by the RTB.

(iv) Conclusions on the interpretation of the statutory language

1. Having considered the wording of section 142 in its immediate statutory context and in the light of the overall legal framework of statutory appeals in Ireland, I fail to see a basis in section 142 (or its surrounding provisions) to indicate that the legislature intended to adopt one of the distinct categories of appeals identified in the judgment of Clarke J. in FitzGibbon.  On the contrary, section 142 (read in its context) provides no more clarity than the various forms of statutory appeal provisions which preceded it and has necessitated precisely the type of preliminary scrutiny which the Supreme Court in Nowak and Kelly hoped could be avoided by a more coherent approach to the drafting of legislation governing statutory appeals.

2. What an analysis of the cases does reveal however are a serious of important propositions.  Several of these are highlighted in an important passage from the judgment of McKechnie J. in FitzGibbon on which the DPC placed particular emphasis:

"... it seems to me that in addition to the terminology used, one must also consider matters such as the nature of the body in question, the knowledge and expertise within it - by reference to its purpose - the subject matter of its remit, the type of decision(s) involved, and the impact and consequences thereof for the addressee(s); the proceedings and the type of process involved in, and giving rise to, the decision under appeal must also be considered. These are but some of the factors which will inform the provision's ultimate meaning " (at [75]).

1. First, the determination of the scope and standard of appeal is a matter of interpretation of the specific statutory regime being applied.

2. Second, several judgments recognise the difficulties that have been faced as a result of the introduction of divergent and sometimes sparse statutory appeal mechanisms.  Some judgments advocate for the introduction of consistent approaches, but there is also recognition that this is a matter of legislative policy and the primary question remains the interpretation of the language of the statute in its proper context.

3. Third, there is some debate as to the weight that falls to be attached to some specific judgments (most notably the judgment of O'Donnell J. (as he then was) in Nowak), but the unassailable fact is that there is no judgment authoritatively interpreting section 142, following argument and submission as to the scope or standard of review under those provisions.  There is a recent judgment of Lankford J. in O'Brien v. The Data Protection Commissioner (20 February 2026), in respect of which the parties furnished brief supplemental written submissions in March 2026.  In that judgment, the court approached an appeal under section 150(5) 2018 Act on the basis that it was an appeal against error with Orange as the applicable standard of review. Nowak was cited as authority for this proposition.  It is common case that there was no argument or dispute on this point in that case, there was no fine, and section 142 was not applicable.  The DPC contends that it is nonetheless relevant as an example of the application of Orange in an appeal against error under section 150(5) and assists its position here, whereas it is LinkedIn's submission that the approach taken in O'Brien is not relevant to the issues being decided in this judgment.  Several grounds are cited by LinkedIn to support this position, including that " a point not argued is a point not decided " (State (Quinn) v. Ryan [1965] IR 70 at 120); O'Brien was not concerned with section 142; no fine was imposed and the constitutional, ECHR and Charter arguments raised here were not engaged or argued (among other grounds).  The State broadly agreed with LinkedIn's analysis.

4. Having considered the judgment and the parties' written submissions in respect of it, I prefer the view of LinkedIn and the State.  The judgment proceeds on the basis of Nowak but without the benefit of the argument that was advanced before me or the authorities opened in the course of the hearing of the preliminary issues and - significantly - O'Brien did not concern the imposition of a fine so some of the considerations which are prayed in aid in the interpretation of section 142 here simply did not arise.

5. Ultimately, I agree with the observation made on behalf of the State in its supplemental submissions of 20 March 2026 that,

"The fact that the test in Orange may have routinely been applied to such appeals in that past (including in O'Brien) does not, however, assist in the exercise of statutory interpretation in which the Court is currently engaged. Necessarily that exercise is by its very nature an autonomous one that is wholly dependent upon the terms of the statute as distinct from the approach of any parties to such an appeal."

1. Fourth, whether or not there is a possibility of admitting new evidence in an appeal may point more towards an appeal de novo but there are also cases in which new evidence has been countenanced as a possibility in an appeal on the record or an appeal against error applying Orange.  I am satisfied that the question of whether new evidence can be admitted is - as I will explore further under the heading of Issue 4 - a matter within the discretion of the trial judge and does not of itself necessarily indicate or determine the nature or standard of review.

2. Fifth, the power to substitute or replace the decision of a statutory body on appeal is an important indicator of the nature of the appeal intended by the Oireachtas. **** The State relies heavily on this aspect of section 142 as pointing towards an appeal on the record.  The DPC disagrees and contends that this is about the remedy or the outcome, not about the form or standard of review or the depth of the scrutiny to be undertaken.  It is contended that the Oireachtas could have expressly provided for a full review on the merits, but did not do so and that the test under Orange does involve substantive scrutiny.  It points to Rye as a relevant authority and to the Central Bank Act 1942 and the FSPO as relevant examples of legislation in which wide discretion was conferred.

3. As has been seen from an assessment of the relevant judgments, the powers conferred on a court are an important indication of the breadth of the appeal.  The question of whether a court can substitute the decision of the statutory power is frequently cited as a relevant factor in discerning the nature of the appeal.  In Orange itself, for example, Keane C.J. determined that the appeal in that case " was not intended to take the form of a re-examination from the beginning of the merits of the decision appealed from culminating, it may be, in the substitution by the High Court of its adjudication for that of the first defendant."

4. In this case, the breadth of the court's powers under section 142, particularly the fact that the court can supplant not just the outcome but the decision of the DPC without limitation or reservation, does point towards a more fulsome appeal than an appeal against error on the basis of Orange. The formulation used - that the court may " replace the decision with such other decision as the court considers just and appropriate, including a decision to impose a different or no fine " - is in my opinion strongly indicative of a free-standing power of the court to make its own decision as to the just and appropriate outcome.  It is not a power to merely vary the fine.  Rather, the court is wholly untrammelled as to the correct decision (provided it is " just and appropriate ").

5. I agree with LinkedIn and the State that it is notable that there are few other legislative regimes in which a power to substitute or replace the underlying decision is conferred on the court.  The DPC relies on the breadth of the appellate power conferred by section 57CM of the Central Bank Act 1942 as applied in **** Ulster Bank v. FSO and the Copyright Act 1963 as applied in Carrickdale.

6. The DPC contends that the test applied in such cases was Orange, and that the use of the word "replace" is not different in any material way to the formulations in the legislation assessed in Orange, Ulster Bank v. FSO or Carrickdale, which gave broad powers to the Court.

7. This does not however answer the fact that those provisions contain no equivalent to section 142(3)(b) or section 150(6)(b).  There are broad appeal provisions to make such order as the court deems appropriate and specific powers to affirm, set aside or remit the underlying order (section 57CM) and even a power to vary the underlying order (section 40(1) of the Copyright Act 1963), but no power of substitution. I consider a power to vary an order to be somewhat of a different magnitude.  The specific importance of substitution is something that emerges from an analysis of the judgments on this topic, including the judgments of the CJEU, as will be seen.

8. It may be something the DPC regards as incongruous with the nature and extent of its specialised expertise, that a court would have such an extensive, unlimited power as to be able substitute the DPC's decision in its entirety with a new, different decision, but that legislative election was made by the Oireachtas.  It is striking that the only other legislative appeal regime cited by the parties in the hearing before me which confers a specific power to replace the decision of the statutory body with the decision of the court is section 148AA(3)(a)(iii) of the Residential Tenancies Act 2004 (as inserted by section 28 of the Residential Tenancies (Amendment) Act 2019).  This was the provision assessed in Xerico and which resulted in the appeal being treated as one on the record and not amenable to the application of the Orange test.

9. The sixth point that arose concerns the power of remittal (or absence thereof). The DPC contests that the absence of a power to remit has any relevance.  It is pointed that other regimes like in Carrickdale, Rye and FitzGibbon also had no power to remit, but were nonetheless appeals against error.  The State, by contrast, weighs this as a relevant factor in its interpretation that the form of appeal is an appeal on the record and not against error applying Orange.

10. I accept that the absence of a power to remit is not decisive as regards the nature of the appeal, but, coupled with the fact that the court has the power to replace the DPC's decision under section 142, it does point towards the court as the primary decision-maker and as having a full jurisdiction to make the decision regarding the imposition of fines, without being bound to weigh the decision of the DPC.

11. The seventh point that arises is the relevance of the nature of the statutory body and its relevant expertise and technical specialisation. **** This is undoubtedly a weighty factor in assessing the scope and nature of a statutory appeal (as emphasised by McKechnie J. in FitzGibbon at [75]).   The court hearing this appeal will have to carefully weigh whether and to what extent to defer to the undoubted and substantial expertise of the DPC on issues within the scope of that expertise.  However, when it comes to the imposition of fines, I agree with the State that courts are well-placed to undertake an analysis of several of the issues arising.  Further, the overarching task remains the interpretation of the language of the statute.  The Oireachtas was fully aware of the composition, qualifications and role of the DPC in enacting the 2018 Act and elected to confer on the courts the power to substitute or replace the decision of the DPC without reservation or limitation.  The question of curial deference will be addressed under a discrete heading below.

12. An eighth consideration to note is the importance of the impact of the impugned decision on the addressee.  There is no doubt that a decision of the DPC under Part 6 of the 2018 Act may have significant consequences for a processor or controller.  The scale of potential fines is high (depending on turnover), as illustrated by the LinkedIn Decision. However, this is precisely why the GDPR has the protections inbuilt throughout that measure, most notably in this context, Articles 83 and 78. It is imperative that sections 142 and 150 of the 2018 Act which give effect to the GDPR are interpreted so as to ensure the level of protection afforded meets the requirements of the GDPR and of the Charter.  As will be addressed later, I have concluded that the form and standard of appeal for which both sides advocate meets those requirements.

13. A further point that must be flagged is that, while it is undoubtedly the case that Orange is still part of the landscape of possible standards to apply in statutory appeals and has been the favoured approach in many cases, the question of the formulation of Orange has been raised (particularly in Nowak).  There is no universal consensus as to where exactly it does or should fit among the spectrum of possible appeal regimes and standards.  I note that the LRC Paper repeatedly raises the question, and seeks submissions as to whether Orange should be a guiding standard in statutory appeals.

14. Finally, I do regard the judgment in Xerico as supporting LinkedIn's case as to the applicable form and standard of appeal.  The DPC seeks to distinguish Xerico on the basis of the statutory procedure, contending that the procedure before the DPC is of a different nature with considerably more engagement and opportunity afforded to LinkedIn to participate.  LinkedIn's contentions that the process was truncated are disputed.  While I do not accept that the process engaged in by the DPC here can properly be characterised as truncated, this is not a valid basis to distinguish the findings in Xerico for three reasons:

• The first important point that must be noted is that the exercise of interpretation section 142 in its context must be fact neutral.  The facts of this case do not determine the correct interpretation of section 142.  As Xerico makes abundantly clear, the question is what the legislation provides for, not how it was operated in a given case.

• Second, whether the legislation provides for the conduct of an oral hearing is a relevant potential factor.  As already summarised, there is no guarantee of, or right to, an oral hearing in the 2018 Act.  In this case, while the process did take 6 years, there was no oral hearing.  This is not to minimise the extent or scale of the investigation and engagement that occurred in this case.  Indeed, by apparent contrast with Xerico, it is clear from the Decision (at [514a]) that LinkedIn was given a copy of the PDD; a copy of the CSA's comments on the Draft Decision; the changes the DPC proposed to make to the PDD; and an opportunity to make submissions in relation to the comments.  When interpreting the 2018 Act, however, it is the provisions of that Act and what it requires which fall for consideration.   The facts of this case or the formulation of the LinkedIn Decision cannot dictate the correct interpretation of the 2018 Act.  This is precisely what was found in Xerico and is a proposition I accept and apply here.

• Third, the existence of a right to make submissions or otherwise be heard and advance a case at a late, close to final, stage of the statutory process may be relevant in characterising the nature and scope of the appeal.  The fact that the landlord did not see the final report in Xerico was important to the decision that the appeal in that case must be an appeal on the record.  Here, the final opportunity which the 2018 Act confers on a data controller or processor to make submissions, is following receipt of the draft investigation report.  The 2018 Act does not confer a right to make submissions or be heard in relation to the draft decision of the Commission or the revised draft decision under Article 60.  This is the salient fact.  The fact that LinkedIn was given an opportunity to make submissions on the final draft decision is a fact of this case, but that does not convert it into a requirement or entitlement under the 2018 Act which can aid in its interpretation.

(f) GDPR, Charter and ECHR

1. The primary provisions of the GDPR invoked by LinkedIn are Article 78 and Article 83.  The latter has already been addressed. Article 78(1) is important.  It is headed " Right to an effective judicial remedy against a supervisory authority " and provides as follows:

"Without prejudice to any other administrative or non-judicial remedy, each natural or legal person shall have the right to an effective judicial remedy against a legally binding decision of a supervisory authority concerning them."

1. The parties also rely on recital 143:

"...each natural or legal person should have an effective judicial remedy before the competent national court against a decision of a supervisory authority which produces legal effects concerning that person. Such a decision concerns in particular the exercise of investigative, corrective and authorisation powers by the supervisory authority or the dismissal or rejection of complaints...

Proceedings against a supervisory authority should be brought before the courts of the Member State where the supervisory authority is established and should be conducted in accordance with that Member State's procedural law. Those courts should exercise full jurisdiction, which should include jurisdiction to examine all questions of fact and law relevant to the dispute before them."

1. Recital 148 then provides that

"In order to strengthen the enforcement of the rules of this Regulation, penalties including administrative fines should be imposed for any infringement of this Regulation, in addition to, or instead of appropriate measures imposed by the supervisory authority pursuant to this Regulation.... [t]he imposition of penalties including administrative fines should be subject to appropriate procedural safeguards in accordance with the general principles of Union law and the Charter, including effective judicial protection and due process".

1. Article 47 of the Charter provides in similar terms that,

"Everyone whose rights and freedoms guaranteed by the law of the Union are violated has the right to an effective remedy before a tribunal in compliance with the conditions laid down in this Article..."

1. Article 6(1) of the ECHR is in the following terms:

"In the determination of his civil rights and obligations or of any criminal charge against him, everyone is entitled to a fair and public hearing within a reasonable time by an independent and impartial tribunal established by law."

1. LinkedIn relies on each of these provisions.  It contends that Orange (combined with the application of deference) is incompatible with Article 6(1) ECHR and Article 47 of the Charter.  Reliance is placed on Case C 386-10 Chalkor AE Epexergasias Metallon v. European Commission ("Chalkor") and on a UK judgment, Competition and Markets Authority v. Flynn Pharma [2020] 4 All ER 934 (" Flynn "), among others.

2. The DPC in reply contends that there is no tension between Orange and the requirements of Article 47 of the Charter.  It is said that Chalkor and Flynn can be distinguished and that considerations akin to deference are mandated by the CJEU judgment in SCHUFA. It is said, relying on SCHUFA, that the concept of " full jurisdiction " is premised on there being a margin of discretion for the competent authority with which the reviewing court will not interfere.

3. The position of the State (albeit somewhat tentatively) is that there is no tension between the requirements of EU law and the form of appeal for which either side contends here (whether appeal on the record or an appeal against error applying Orang e).  According to the State, the provisions of the Charter and the ECHR do not therefore assist in the interpretation of sections 142 and 150.

4. Both LinkedIn and the State place particular reliance on Chalkor and SCHUFA. I will address those judgments in some detail.

Case C-386/10 P. Chalkor AE Epexergasias Metallon v European Commission.

1. Chalkor was a company incorporated under Greek law and listed on the Athens stock exchange and was engaged in the production of semi-finished copper products. Following an investigation, the Commission issued a decision determining tha t Chalkor had participated in agreements and practices with other market operators designed to agree prices, share markets and exchange confidential market information.   A total fine of €9.16 million was imposed on Chalkor. This was appealed to the General Court on several grounds and the fine was reduced.

2. On appeal, the General Court sought to verify whether Commission had exercised its discretion in accordance with the method set out in applicable guidelines (Guidelines on the method of setting fines imposed pursuant to Article 15 (2) of Regulation No 17 and Article 65(5) [ECSC]), and whether any departure from those guidelines was " justified and supported by sufficient legal reasoning " (General Court at [61]).

3. The approach of the General Court was that " in areas where the Commission has maintained a discretion... review of the legality of those assessments is limited to determining the absence of manifest error of assessment... " (at [63]).  It was noted that this discretion did not prejudge the exercise by the Court of its unlimited jurisdiction which empowered it to annul, increase or reduce the fine imposed by the Commission (General Court at [64]).

4. Chalkor then appealed to the CJEU on the ground (among others) that the General Court erred in the standard of review applied.  The complaint was that the General Court was wrong to limit its review to a mere review of legality.  It was further contended that Articles 47 and 49 of the Charter required a full review of Commission decisions as regards both matters of fact and law. The Commission in reply contended that the concept of ' full jurisdiction' for the purposes of the ECHR is not the same as that of the 'unlimited jurisdiction' under EU law and that the jurisdiction of the General Court to substitute its own assessment of the fine for that of the Commission goes beyond what is required by the ECHR.

5. The judgment of the CJEU separately addressed the question of the Court's jurisdiction to " review the legality " of an act as provided for in Article 263 TFEU (in respect of which the Commission has a margin of discretion) and the broader unlimited jurisdiction in respect of the amount of the fine, provided for under Article 31 of Regulation No 1/2003.  The latter provides "[t]he Court of Justice shall have unlimited jurisdiction to review decisions whereby the Commission has fixed a fine or periodic penalty payment. It may cancel, reduce or increase the fine or periodic penalty payment imposed."

6. The Court recorded that, "[i]n addition to the review of legality, now provided for under Article 263 TFEU, a review with unlimited jurisdiction was envisaged in regard to the penalties laid down by regulations " (at [53]).

7. The Court assessed the power to review the legality of Commission decisions as follows:

" As regards the review of legality, the Court of Justice has held that whilst, in areas giving rise to complex economic assessments, the Commission has a margin of discretion with regard to economic matters, that does not mean that the Courts of the European Union must refrain from reviewing the Commission's interpretation of information of an economic nature. Not only must those Courts establish, among other things, whether the evidence relied on is factually accurate, reliable and consistent but also whether that evidence contains all the information which must be taken into account in order to assess a complex situation and whether it is capable of substantiating the conclusions drawn from it (see Case C‑12/03 P Commission v Tetra Laval [2005] ECR I‑987, paragraph 39, and Case C‑525/04 P Spain v Lenzing [2007] ECR I‑9947, paragraphs 56 and 57 " (at [54]).

1. This is an important indication that, even when the legislation expressly limits the court's role to a review of legality, the court must still test the evidence relied upon, even on complex questions, and assess whether it suffices and substantiates the conclusions drawn.  The Court returned to the topic of the " review of legality " later in the judgment:

" In carrying out such a review, the Courts cannot use the Commission's margin of discretion - either as regards the choice of factors taken into account in the application of the criteria mentioned in the Guidelines or as regards the assessment of those factors - as a basis for dispensing with the conduct of an in-depth review of the law and of the facts " (at [62]).

1. This passage indicates that, even applying the margin of discretion to the Commission in a review of legality, the Court must still carry out an in-depth review of both the law and the facts.  The Court concluded that the review of legality under Article 263 TFEU, supplemented by the unlimited jurisdiction in respect of the amount of the fine, under Article 31 was " not contrary to the requirements of the principle of effective judicial protection in Article 47 of the Charter " (at [67]).

2. This decision was quoted at considerable length by both counsel for LinkedIn and counsel for the DPC.  LinkedIn relied on Chalkor as authority that the review of legality must be based on evidence put forward by the appellant; the CJEU cannot use the Commission's margin of discretion as a basis to dispense with an in-depth review of the law and the facts; and the Court may substitute its decisions for those of the Commission when carrying out a full review of the fine.

3. The points which the DPC seeks to extrapolate from Chalkor are that it supports the DPC position on the applicable standard of review; on the admissibility of new evidence on the fining decision alone; and on the possible bifurcation between review of infringement and fining decisions.  Emphasis is placed on the duty of the Court to conduct a review Commission's decisions for legality under Article 263, allowing a margin of discretion, and to then conduct a more extensive review of the fine, allowing for new evidence to be adduced on that issue.  This is said to be what the CJEU found to meet the requirements of Article 47 of the Charter.  It is said that this is very similar to the bifurcation between sections 142 and 150 of the 2018 Act, with the scope for a more extensive review with new evidence under section 142.

Assessment

1. Chalkor was decided in a very different factual context and on the basis of markedly different rules to those which apply in this case.  There was a mandated differentiation there between the " review for legality " of an act of the Commission under Article 263 TFEU and the expressly " unlimited jurisdiction " to review the fine fixed under Article 31 of Regulation 1/2003.  The Court's analysis of the nature and standard of review under each heading derives from those specific provisions, which have no direct analogue here.   Accordingly, while the DPC quite accurately points to a bifurcation as between the standards and rules applicable to the review against legality under Article 263 and the unlimited jurisdiction under Article 31 (Regulation 1/2003), that was a feature and consequence of those specific rules and I do not see that the approach of the Court or its analysis of those different provisions can be transferred to, or inform, this case in which the task is to interpret provisions of the 2018 Act.

2. This disconnect with Chalkor is reinforced by the fact that the parties here are agreed on one important point:  the same type of appeal and standard of review are applicable under both section 142 and section 150 of the 2018 Act.  The two different forms of review addressed in Chalkor do not reflect the legal regime under the 2018 Act as understood by the parties themselves here.

3. Another important comment about Chalkor is that it assesses the compliance with Article 47 of the Charter of the particular review provisions that were at issue in that case.  It does not purport to set out a guide as to the only form of specific provisions that can meet those criteria (provided the overarching requirements of effective judicial protection are met).

4. Bearing these points of distinction firmly in mind, there are nonetheless some useful points that may be derived from the judgment of the CJEU in Chalkor (several of which are repeated in later judgments also opened by the parties, including Case C-501/11P Schindler Holdings Limited v. Commission (" Schindler ") and Case C-603/13P Galp Energia Espana v Commission (" Galp ")).

5. First, the CJEU referred to the principle of effective judicial protection as " a general principle of European Union law to which expression is now given by Article 47 of the Charter " (at [52]), noting that Article 47 implements, or gives expression in EU law to, the protection afforded by Article 6(1).  The CJEU then proceeded on the basis that it was only necessary to consider Article 47.   Indeed, while Chalkor argued that competition proceedings are criminal proceedings within the meaning of the ECHR, it appeared to accept that was not relevant to the review by the Courts (as distinct from the review by the Commission) (CJEU at [40]).  This is an important point in my view and justifies placing emphasis on Article 47 when it is applicable rather than exploring the remit of Article 6(1).

6. Second, the judgment in Chalkor underlines that Article 47 requires review by a body with 'full jurisdiction' which means a body with the power to 'quash in all respects, on questions of fact and law, the challenged decision'.  A court must have jurisdiction to examine all questions of fact and law relevant to the dispute before it.  According to the Court, this " involves review by the Courts of the European Union of both the law and the facts, and means that they have the power to assess the evidence, to annul the contested decision and to alter the amount of a fine " (at [67]. It was the combination of the Court's role in reviewing legality under Article 263 TFEU and the unlimited jurisdiction under Article 31 of Regulation 1/2003 which was held to meet this requirement.

7. Third, it emerges from the CJEU judgment that, even when the Commission has a margin of discretion with regard to complex questions (such as economic matters), that does not displace the need for the court to review the Commission's interpretation of economic information when reviewing the legality of the decision.  This necessitates an in‑depth review of the relevant law and facts and the evidence adduced by the applicant to establish whether the evidence put forward is factually accurate, reliable and consistent; whether that evidence contains all the relevant and necessary data for appraising a complex situation; and whether it is capable of substantiating the conclusions drawn.  In short, even in a review for legality in which the Commission has a margin of discretion on complex questions, the court must undertake a full review of the law, facts and the evidence put forward (see [54] and [62]).

8. Fourth, the judgment does illustrate the potential overlap between the circumstances of the infringement and the factors that are relevant to the amount of the fine.  The judgment lists the factors which may be relevant to the determination of the amount of a fine as including the duration of the infringements, and all factors which may affect the assessment of gravity, such as the conduct of the undertakings, their role in the conduct in question, the value of the goods, the content and duration of the anti-competitive conduct; the number and intensity of incidents; the extent of the affected market; damage to the economic public order; and any repeated infringements (at [56]-[57]).  The CJEU commented that "[t] his large number of factors requires that the Commission carry out a thorough examination of the circumstances of the infringement " (at [58]).  That indicates that a review of the fining decision must necessarily also incorporate a review of the circumstances of infringement.

9. Fifth, the DPC submits that Chalkor supports its position that new evidence can only be admitted in challenging the imposition of a fine and not the underlying findings of infringement.  I note the following passage:

" ...in regard to infringements of the competition rules, it is for the Commission to prove the infringements found by it and to adduce evidence capable of demonstrating to the requisite legal standard the existence of the circumstances constituting an infringement. What the applicant is required to do in the context of a legal challenge is to identify the impugned elements of the contested decision, to formulate grounds of challenge in that regard and to adduce evidence - direct or circumstantial - to demonstrate that its objections are well founded." (at [65]).

1. The CJEU had previously commented, " the Courts must carry out the review of legality incumbent upon them on the basis of the evidence adduced by the applicant in support of the pleas in law put forward.. " (at [62], also emphasised in Schindler at [37]).

2. These paragraphs indicate that the appellant may adduce evidence to demonstrate that its objections to the findings of infringement are well-founded, even when the review is a review of legality in respect of the findings of infringement.

3. This interpretation is supported by the later judgment in Galp (on which the parties also rely) in which the CJEU found,

"...the scope of judicial review provided for in Article 263 TFEU extends to all the elements of Commission decisions relating to proceedings applying Articles 101 TFEU and 102 TFEU which are subject to in depth review by the General Court, in law and in fact, in the light of the pleas raised by the appellants (see, to that effect, judgments in KME Germany and Others v Commission, C-272/09 P, EU:C:2011:810, paragraphs 102 and 109; Chalkor v Commission, C-386/10 P, EU:C:2011:815, paragraphs 62 and 82; and Telefónica and Telefónica de España v Commission, C-295/12 P, EU:C:2014:2062, paragraphs 56 and 59) and taking into account all the elements submitted by the latter, whether those elements pre-date or post-date the contested decision, whether they were submitted previously in the context of the administrative procedure or, for the first time, in the context of the proceedings before the General Court, in so far as those elements are relevant to the review of the legality of the Commission decision (see, to that effect, judgment in Knauf Gips v Commission, C-407/08 P, EU:C:2010:389, paragraphs 87 to 92)" (at [72]).

1. Galp is an interesting illustration of this point in operation: there, the General Court had refused to consider new evidence in the context of its Article 263 review of legality jurisdiction and found that participation by the appellant in the compensation and control aspects of a cartel related to the marketing of penetration bitumen on the entire Spanish territory (excluding the Canary Islands) was not proved.  However, the General Court then received that evidence in the exercise of its unlimited jurisdiction under Article 31, Regulation 1/2003 and concluded that it did demonstrate the appellant's awareness of the participation of other members of the cartel in those compensation mechanisms and that they could have foreseen such participation in the monitoring system.  The General Court therefore concluded the appellants could be held liable for those aspects of the infringement and that there was no need to vary the starting point of the fine on that aspect of the infringement.  The CJEU found this aspect of the decision to be an error of law, as the General Court in the exercise of its unlimited jurisdiction should assess the fine and not make findings as to liability.  The other aspects of the General Court's judgment were upheld.

2. This shows the distinction between the role of an appellate court in reviewing fines and its role in assessing liability (which is relevant to Issue 1 here).

3. While the facts and law of Chalkor and Galp are significantly different, the DPC does rely on Chalkor for the contention that the admission of new evidence is limited to the fining element of the decision.  I do not see that this contention is well made.  The judgments in Chalkor and Galp rather support the admission of evidence that is relevant to the review of legality under Article 263 TFEU.

4. A related point made by the CJEU is that, even when a court is exercising its " unlimited jurisdiction ", the proceedings are inter partes and " it is for the applicant to raise pleas in law against that decision and to adduce evidence in support of those pleas " (at [64]).  This has some resonance in the context of the preliminary issues I am deciding: whatever standard of review is applied, it is an appeal inter partes against a statutory decision and the appellant will bear the burden of making out specific grounds of appeal (it being accepted that this is not an appeal de novo).

5. Finally, the CJEU places some weight on the power of the Court to substitute its appraisal for that of the Commission when assessing whether the criteria of Article 47 were satisfied.  The Court addressed the unlimited jurisdiction to review the fine fixed as follows:

" That jurisdiction empowers the Courts, in addition to carrying out a mere review of the lawfulness of the penalty, to substitute their own appraisal for the Commission's and, consequently, to cancel, reduce or increase the fine or penalty payment imposed (see, to that effect, Joined Cases C‑238/99 P, C‑244/99 P, C‑245/99 P, C‑247/99 P, C‑250/99 P to C‑252/99 P and C‑254/99 P Limburgse Vinyl Maatschappij and Others v Commission [2002] ECR I‑8375, paragraph 692)" (at [63]).

1. This indicates the importance of the court being able to substitute the Commission's appraisal with its own: there is a correlation drawn between the power to alter the order made and this power to substitute the underlying reasoning or " appraisal ". It highlights the importance of a court being conferred with the power to substitute the decision of the statutory body (a point which has already been seen in Irish cases interpreting statutory appeal mechanisms).

2. I pause for completeness to observe that, in its oral submissions, the DPC referenced a statement that the possibility of substitution in Chalkor went beyond what was required by Article 47.  However, as noted in LinkedIn's replying submissions, that was a submission by the Commission in Chalkor and not the view of the Court.

3. It is however important to note that, in Galp, the CJEU confirmed that the General Court cannot substitute its reasoning for that of the Commission in a review of legality under Article 263 TFEU, that only being possible in the exercise of the unlimited jurisdiction to review the amount of the fine under Article 261 TFEU and Article 31 Regulation 1/2003 (Galp, at [73]-[76]).  Of course, the Oireachtas here has made the significant decision that the court can substitute its decision for that of the DPC in respect of both decisions on infringement and on fines.

Case C-26/22 and C-64/22 UF, AB v. Land Hessen and SCHUFA Holding AG

1. Case C-26/22 and C-64/22 SCHUFA Holding is a judgment of the CJEU concerning the interpretation of provisions of the GDPR on which both LinkedIn and the DPC rely. The applicants were granted early discharge from their debts in insolvency proceedings and they applied to SCHUFA (a private credit information agency) to have the entries related to them deleted. The SA in Hamburg upheld the legality of the data processing by SCHUFA.  The applicants challenged this decision before the German courts, arguing that the SA was obliged to take measures to ensure the entries were deleted. The SA argued that the court review would be confined to examining whether it handled the complaint and informed the complainant of the progress and outcome of that complaint and that it was not the responsibility of the court to review the substantive correctness of the decision.

2. The German court referred a question to the following effect (as paraphrased by the CJEU at [47]):

" whether Article 78(1) of the GDPR must be interpreted as meaning that judicial review of a decision on a complaint taken by a supervisory authority is limited to the question whether that authority has handled the complaint, investigated the subject matter of the complaint to the extent appropriate and informed the complainant of the outcome of the investigation, or whether that decision is subject to a full judicial review, including the power of the court seised to require the supervisory authority to take a specific measure."

1. Rejecting the position advanced by the German SA, the CJEU confirmed that " full judicial review" was necessary and the court undertaking that review must be able to exercise " full jurisdiction, which should include jurisdiction to examine all questions of fact and law relevant to the dispute before them " (at [52]).  According to the Court "limited judicial review" would not meet the requirements for effective judicial protection under the GDPR:

"...in view of the wide-ranging powers vested in the supervisory authority under the GDPR, the requirement for effective judicial protection would not be met if decisions concerning the exercise by such a supervisory authority of powers of investigation or the adoption of corrective measures were subject only to limited judicial review " (at [59]).

1. The points made by the DPC in reliance on SCHUFA are that, while the CJEU found that " full judicial review " is required, this was not defined and was selected over an alternative, very limited form of review.  It was also contended that " full jurisdiction " is not equivalent to the appeal proposed by Linkedln, but rather better accords with the Orange test.  For this proposition, the DPC relies heavily on paragraph 69 of SCHUFA.  In its written and oral submissions, the DPC interprets paragraph 69 as a determination that this court on appeal is not entitled to and should not substitute its own assessment for that of the DPC, but is rather limited to examining whether the DPC " complied with the limits of its discretion even in the context of the fines ".

2. The position of LinkedIn, by contrast, is that the judgment in SCHUFA confirms the need for " full " judicial review and that the guarantee of independence is not impaired by same.  With regard to paragraph 69, LinkedIn's position is that this paragraph addresses the discretion of the SA as to the choice of corrective powers provided for in Article 58(2) GDPR and must be read in the context of the specific corrective orders sought to be compelled in that case (written submissions at [62]).   I pause here to consider these contrasting interpretations.

3. First, there is no ambiguity in the judgment as regards the scope of the appeal required by Article 78: an appeal court must have full jurisdiction to review all relevant questions of fact and law.  The answer furnished by the CJEU to the question referred is an unqualified one:

" In the light of all the foregoing considerations, the answer to the first question is that Article 78(1) of the GDPR must be interpreted as meaning that a decision on a complaint adopted by a supervisory authority is subject to full judicial review " (at [70]).

1. The Court makes clear that limited review would not be sufficient for either " decisions concerning the exercise.. of powers of investigation" or "the adoption of corrective measures " (at [59]).

2. Second, the only " discretion " cited by the CJEU is a specific discretion as regards the choice of corrective measures under Article 58(2).   The Court noted that " while... the supervisory authority must deal with a complaint with all due diligence, that authority has, as regards the remedies listed in Article 58(2) of the GDPR, a margin of discretion as to the choice of appropriate and necessary means..." (at [68], emphasis added).  The only discretion cited in the judgment - a discretion as to the appropriate and necessary corrective measures to impose - is not one of the issues I have to decide here.

3. Third, the DPC seeks to interpret paragraph 69 of the judgment as supporting the Orange standard in an appeal against error as opposed to an appeal on the record.  The relevant wording in paragraph 69 is as follows:

" While the national court hearing an action under Article 78(1) of the GDPR must, as noted in paragraph 52 of this judgment, have full jurisdiction to examine all questions of fact and law relating to the dispute concerned, the guarantee of effective judicial protection does not imply that it is entitled to substitute its assessment of the choice of appropriate and necessary remedies for that of that authority, but requires that court to examine whether the supervisory authority has complied with the limits of its discretion."

1. The DPC submits that the effect of this paragraph is that this court on appeal is limited to examining whether the DPC " complied with the limits of its discretion even in the context of the fines " (day 3, page 41).  The DPC also contends that SCHUFA precludes the substitution of the SA's assessment.  I do not believe the judgment in SCHUFA supports the breadth of that interpretation.   As already noted, the fundamental finding in SCHUFA is that there must be an avenue for full review of all relevant questions of law and fact.  The only discretion addressed concerns the selection of corrective measures.  The facts and findings in SCHUFA are not addressed to the question of interpretation with which I am concerned here.

2. Fourth, the judgment references the procedural autonomy conferred on Member States.  This deference to Member States must be borne in mind when the language of the judgment - particularly paragraph 69 - is being appraised.  I consider that, if the CJEU intended to impose - or interpret Article 78 as imposing - a limit on the powers a Member State may confer on a court, this would need to be clearly and specifically stated and there is no such statement in the judgment.

3. Fifth, I do not believe it is necessary to parse the words used in paragraph 69, given the findings I have already made as regards the context and import of the judgment and its findings.  For completeness, however, if these words are scrutinised in light of the judgment as a whole, they may be interpreted more narrowly than the DPC contends.  In particular, the CJEU in paragraph 69 states that,

"... the guarantee of effective judicial protection does not imply that it is entitled to substitute its assessment of the choice of appropriate and necessary remedies for that of that authority, but requires that court to examine whether the supervisory authority has complied with the limits of its discretion. "

1. This could be read as a statement that the guarantee of effective judicial protection (or the right to " effective judicial remedy " as it is framed in Article 78(1)) does not of itself imply or necessitate that the court must be entitled to substitute the SA's view of the appropriate remedies with its own, provided the court can confirm whether the limits of that discretion were complied with.  It is a matter for each Member State, as already noted, to frame its own appeal regimes.

2. To deviate briefly from SCHUFA, the DPC in its written and oral submissions relies on the similar statement of the law by the European Court of Human Rights (" the ECtHR ") in Ramos Nunes de Carvalho e Sá v Portugal (Application nos. 55391/13, 57728/13 and 74041/13) (a case I will revisit under the heading of Issue 4):  " It can be derived from the relevant case-law that it is not the role of Article 6, in principle, to guarantee access to a court which can substitute its own assessment or opinion for that of the administrative authorities " (at [178]). This is said to support Orange as the applicable standard of review.  I read this statement - as I do paragraph 69 of the judgment of the CJEU in SCHUFA - as indicating a limit as to what is imposed by Article 78 of the GDPR or Article 6(1) ECHR.  I do not understand the CJEU or the ECtHR to be dictating a limit on what Member States can do.

3. In this jurisdiction, by the enactment of section 150(6), the Oireachtas has chosen to give a reviewing court the power to substitute its decision - without limitation - for that of the DPC including as regards the exercise of a corrective power.  I would find it surprising if the GDPR, the Charter or the ECHR had the effect of reducing the scope of that power of review in a manner not easily reconcilable with apparent from the statutory language itself.  I also see nothing in those measures that imposes or even favours that outcome.

4. In short, I am not persuaded that SCHUFA can be read as favouring the more limited form of appeal for which the DPC contends or as precluding the form of appeal for which LinkedIn advocates. The overriding import of SCHUFA is that courts hearing appeals against decisions of SAs must have full jurisdiction to review all relevant questions of fact and law and cannot be confined to a form of limited review (such as was suggested by the SA in SCHUFA). There was no question put to the CJEU as to the forms of appeal contended for by the parties to these proceedings.  If the CJEU intended to make a more general pronouncement about the scope of an appeal for which Member States are permitted to legislate, it may be expected that this would have been said with specificity and clarity.  I do not see any such statements in SCHUFA or indeed in the GDPR itself.

Other Judgments

1. Another judgment on which LinkedIn places reliance is Case C-132/21 BE v Nemzeti Adatvedelmi es Informachiszabadsag Hatosag (" BE ").  There, a specific issue arose about the interaction between the review by domestic courts of decisions of SAs under Articles 77, 78 and 79 GDPR.   The context, broadly, was that a dispute arose between an individual "BE" and a public limited company of which he was a shareholder regarding access to recordings of a general meeting of the company.  BE unsuccessfully challenged the refusal to grant access before the Hungarian SA and then appealed the decision of the SA to the referring court under Article 78.  While that appeal was pending, another Hungarian court determined a parallel challenge which BE brought against the company directly under Article 79.  The referring court sought answers as to the priority as between the two cases.  The CJEU formulated the questions referred as follows:

" whether Article 77(1), Article 78(1) and Article 79(1) of Regulation 2016/679, read in the light of Article 47 of the Charter of Fundamental Rights ('the Charter'), are to be interpreted as meaning that the remedies provided for in Article 77(1) and Article 78(1) of that regulation, on the one hand, and Article 79(1) thereof, on the other, are capable of being exercised concurrently with and independently of each other, or whether one of them has priority over the other."

1. Paragraph 41 is the most important general statement in that judgment: "... it follows from Article 78(1) of Regulation 2016/679, read in the light of recital 143 of that regulation, that courts seised of an action against a decision of a supervisory authority should exercise full jurisdiction, which should include jurisdiction to examine all questions of fact and law relevant to the dispute before them."

2. Another important finding is in paragraph 45, which points firmly to the importance of a Member State's procedural autonomy:

"In the absence of EU rules governing the matter, it is for each Member State, in accordance with the principle of the procedural autonomy of the Member States, to lay down the detailed rules of administrative and judicial procedures intended to ensure a high level of protection of rights which individuals derive from EU law."

1. The Court reiterated that it is for the domestic court to decide, on the basis of national law, how the remedies under the GDPR are to be implemented, provided the rules for the implementation of the remedies ensures effectiveness and effective protection of the rights guaranteed.  The following paragraphs address the interaction between domestic law and rules and the requirements of the GDPR:

"[48] Those detailed rules must not be less favourable than those governing similar domestic actions (principle of equivalence); nor must they render practically impossible or excessively difficult the exercise of rights conferred by EU law (principle of effectiveness) (see, to that effect, judgment of 14 July 2022, EPIC Financial Consulting, C-274/21 and C-275/21, EU:C:2022:565, paragraph 73 and the case-law cited).

[49] Under the principle of sincere cooperation laid down in Article 4(3) TEU it is for the courts of the Member States to ensure judicial protection of a person's rights under EU law. In addition, Article 19(1) TEU requires Member States to provide remedies sufficient to ensure effective judicial protection in the fields covered by EU law (judgment of 27 September 2017, Puškár, C-73/16, EU:C:2017:725, paragraph 57).

[50] In particular, when the Member States set out detailed procedural rules for legal actions intended to ensure the protection of rights conferred by Regulation 2016/679, they must ensure compliance with the right to an effective remedy and to a fair trial, enshrined in Article 47 of the Charter, which constitutes a reaffirmation of the principle of effective judicial protection (see, by analogy, judgment of 27 September 2017, Puškár, C-73/16, EU:C:2017:725, paragraph 59)."

1. The Court regarded the risk of conflicting decisions between the courts of a Member State under Article 78 and Article 79, as a situation which could endanger the consistent and homogenous application of the provisions of the GDPR, create uncertainty and weaken protection under that Regulation.  The answer to the question referred was that it was for Member States in the exercise of their procedural autonomy to lay down detailed rules as regards remedies to ensure the effective protection of rights under the GDPR and the rights referred to in Article 47.

2. This illustrates the extent of the procedural autonomy of Member States in the implementation of the guarantees and protections enshrined in the GDPR.  The rules governing remedies and effective protection of the rights under the Regulation are a matter for Member States, provided the requirements of equivalence and effectiveness are met and there is an effective remedy before a court in accordance with Article 47 of the Charter.

3. It is important to remain mindful of this procedural autonomy of Member States under the GDPR when there is a dispute - as there is between LinkedIn and the DPC - regarding the scope of the appeal provided for in domestic legislation.

4. For completeness, I note another judgment cited by LinkedIn is the judgment of the UK Court of Appeal in Flynn.  This is not directly relevant here, concerning as it does the interpretation of English Competition Act 1998.   However, insofar as LinkedIn relies on it as authority that a full merits review is mandated by Article 6(1) and is (apparently) incompatible with any margin of discretion (citing [136] in its written submissions at [58]),  I do not see Flynn as necessarily supporting an argument against the application of Orange or against the affording of a margin of deference.  The following paragraph from the judgment of the UK Court of Appeal may be noted,

"[143] In T-Mobile (UK) Ltd v Office of Communications [2008] EWCA Civ 1373, [2009] 1 WLR 1565 it was observed that the task of the Tribunal was not to serve as a 'fully equipped duplicate regulatory body waiting in the wings just for appeals'. It is to 'look into whether the regulator [has] got something materially wrong'. The reference to materiality is important. The Tribunal should interfere only if it concludes that the decision is wrong in a material respect. "

1. These findings in Flynn point towards an appeal which requires material error by the decision-maker and do not necessarily disfavour the application of the Orange test or of a margin of appreciation.

Conclusions

1. The following points may be derived from the CJEU judgments just discussed:

• The principle of effective judicial protection is " a general principle of European Union law to which expression is now given by Article 47 of the Charter " (Chalkor at [52]);

• As Article 47 implements the protection afforded by Article 6(1), it is only necessary to consider Article 47 and Article 6(1) does not need to be assessed (Chalkor at [51]);

• Article 47 requires review by a body with 'full jurisdiction' which means a body with the power to ' quash in all respects, on questions of fact and law, the challenged decision '.

• A reviewing court must have jurisdiction to examine all questions of fact and law relevant to the dispute before it (BE at [41]);

• Even in a review for legality in which the Commission has a margin of discretion on complex technical questions, the court must undertake a full review of the law, facts and the evidence put forward (Chalkor at [54]);

• It is a matter within the procedural autonomy of Member States to decide the procedural rules for legal actions, provided the requirements of effectiveness, equivalence and the obligation to protect rights under the GDPR and Article 47 are satisfied.

1. Looking at these points, I fail to see how it is asserted that the application of the test in Orange in an appeal under section 142 is deficient or would fail to properly respect the requirements of the GDPR, the Charter or the ECHR.   Sections 142 and 150 confer a full right of appeal to court, with no limitation on the issues which may be appealed, and with the court having the power to substitute the decisions of the DPC as to both infringement and fines.   As will be seen, the position of the DPC is that there should only be deference on certain issues, to be determined by the judge hearing the appeal.  There is no argument being advanced that deference is owed to the DPC on all aspects of its decision as a matter of principle (regardless of whether Orange applies).

2. The DPC disputes the requirement to undertake a review of all facts and law relevant to the appeal or issue would be displaced if Orange was the applicable standard.  The DPC points to the finding in Ulster Bank v FSPO [2024] IECA 231 (Meenan J., Allen and O'Moore JJ.in agreement) that, when the court is reviewing a decision of the FSPO on the construction of a contract, it should first reach its own conclusion as to what that proper construction is before considering whether the test in Orange is met, with deference being " limited to facts of a specialist or technical nature " (at [45]).   This indicates the independent need for a court to undertake a review of the facts and law relevant to an appeal, before testing that against the applicable standard of review or margin of deference owed.   The judgment of the CJEU in Chalkor would point towards a similar interpretation of the scope of appeal as a matter of EU law.

3. If Orange was found to be the applicable test (and that is not my view of section 142 or section 150), I do not see that there would be a valid basis in the GDPR, Article 47 of the Charter or Article 6 ECHR to displace that interpretation.  On the contrary, from the assessment of the judgments of the CJEU on which the parties place most emphasis, I do not see cognisable support for either form of appeal advocated by the parties here.  The judgments instead emphasise the common threshold that must be met (primarily the requirement that the court must have jurisdiction to examine all questions of fact and law relevant to the dispute) and the importance of Member States' procedural autonomy.  It has not been shown to me how an appeal against error with Orange as the standard does not meet that threshold.

4. Given these findings, the approach I take is that favoured by O'Donnell J. in Nowak, namely that it is sufficient to conclude by the application of " traditional techniques of interpretation " what an appeal under those provisions entails, given that any of the options being argued for are at a minimum, not inconsistent with the requirements of European Union law (see Nowak at [25]). The very fact that this finding was made in a case to which Article 47 applied, and in which O'Donnell J. favoured the application of Orange as the standard of review, is one to which heavy weight must be attached here.  As the DPC quite accurately points out, this is a firm indication that an appeal in which Orange is applied as the standard of review does afford an effective judicial remedy for the purposes of Article 47.

5. In light of these conclusions, and given that Article 47 transplants the protections under Article 6(1) into EU law, I do not believe it is necessary or appropriate to analyse the question of whether the process or administrative fines are criminal in nature for the purpose of deciding whether Article 6(1) applies.     As extensive submissions were made on these points, it may assist if I further explain the reasons for my decision not to examine whether the criminal scope of Article 6(1) is engaged.

6. First, LinkedIn confirmed repeatedly during its submissions that it does not seek a ruling on the question of whether the process or the fines are criminal in nature or whether the DPC process amounts to the administration of justice and not the ex ercise of limited functions and powers of a judicial nature, within the meaning of Article 37 of the Constitution.

7. Second, the criminal characterisation of the process and the administrative fines are substantive issues in these proceedings and in the judicial review proceedings.  The grounds pleaded in these proceedings (ground 8) and in the judicial review proceedings (core grounds 1, 2, 3) raise the question of whether the fines imposed by the DPC amount to the imposition   of criminal sanctions.  It is not appropriate to veer into determining such questions that are part of those proceedings and which have not been identified as, or agreed to be, suitable for preliminary determination.

8. Third, this judgment addresses four specific preliminary issues which the parties selected and agreed to put forward on the basis they meet the strict criteria for be tried as such.  It is a departure from the default unitary trial and I am acutely conscious of the need not to expand my determination beyond the four corners of those issues.  I am also conscious that, apart from these proceedings, there are judicial review proceedings pending between LinkedIn and the DPC, and several other cases pending between the DPC and other parties in which some of the important questions ventilated in the hearing before me may also arise for consideration.

9. In light of the determinations made (among other factors), the application of Article 6(1) does not need to be determined.   I therefore avoid making determinations in respect of, or even comments on, weighty questions of law which go beyond, and have transpired not to be necessary for the determination of, the four preliminary issues.  Any such questions can be properly ventilated and determined at the appropriate time and in the appropriate forum.

10. Fourth, the question of whether the GDPR provides for criminal processes and sanctions is an issue of real complexity and importance and one that has not previously been decided by the CJEU.  I note the opinion of Advocate General Pikamae in Case C-768/21 TR v. Land Hessen and the findings in Case C-667/21 ZQ at [85] relied upon by LinkedIn (the CJEU in ZQ refers to Article 83 having a " punitive purpose ") but the fact remains that the designation of administrative fines under the GDPR as criminal has not been the subject of considered final determination.  I consider it inappropriate to determine such matters in a preliminary hearing absent full argument by both sides and by the State on this specific topic.  That will appropriately occur during the substantive proceedings.

11. Fifth, given the findings already made - including that Orange is not the applicable standard of review in an appeal under the 2018 Act; that Article 47 adopts Article 6(1) into EU law; and that an appeal against error applying Orange would not in any event fail to meet the criteria of Article 47 –  there is no need to assess whether Article 6(1) applies or whether its requirements are met.   Questions about Article 6(1) would not add to the analysis or assist in the provision of answers to the preliminary issues presented.

12. Sixth, there is a related constitutional element to these questions, given the argument that the exercise of the enforcement powers of the DPC constitutes the administration of justice and is a criminal matter and accordingly not saved by Article 37(1).  This is part of the pleaded case. Insofar as LinkedIn seek to argue the criminal characterisation of the process in the context of Article 37(1) of the Constitution, that is no part of the preliminary issues before me and does not need to be decided in order to interpret sections 142 and 150 (it being confirmed by the State that double construction will not be argued in the substantive proceedings).  It is not appropriate to engage in such constitutional questions unless necessary to do so and I am satisfied this is not necessary for the issues I have to decide here.

13. Seventh, I can see a danger in proceeding to interpret the 2018 Act on the premise or assumption that it involves a criminal process and sanction, without that point having been decided in properly contested proceedings.   I do not therefore decide this issue or proceed on any assumptions as to its outcome in this judgment.

14. Eighth, the different treatment of the ECHR under section 2 of the European Convention on Human Rights Act 2003, and the more direct impact of the Charter and the GDPR, further support this approach. I agree with the State and the DPC in this respect.

15. There was an additional reason raised by the DPC as to why the question of the administrative fines being criminal in nature should not be determined here.  It was said that there was insufficient notice of the intention to include this point among the preliminary issues and that, if DPC had received fair notice, it may have sought to lead evidence such as comparative evidence regarding other statutory bodies.

16. I do not believe a valid complaint of procedural prejudice necessarily arises, given the question of whether the process or fines are criminal in nature was flagged by LinkedIn as part of its case on these topics since the delivery of written submissions.  Nonetheless, for the various reasons already given, I am satisfied that it is neither necessary nor appropriate for this important question to be determined here.

17. There are two further related topics which I do not consider it necessary or appropriate to decide and which I do not therefore decide in this judgment.  Several of the reasons for not deciding the Article 6 point are equally applicable here.

18. The first topics concerns the asserted lack of independence of the DPC as a structural matter.  These arguments are made by LinkedIn by reference to Article 6(1) ECHR and rely heavily on the judgment of the ECtHR in Grande Stevens v. Italy (Application no. 18640/10) (at [132] to [139]). However, as the DPC points out, the requirement of institutional independence is also contained in the GDPR (including Article 52), and the more pressing issues here are the need to have an appeal mechanism that complies with the GDPR and whether and how the 2018 Act can be interpreted in line with that requirement.  I do not see that an assessment of the structure of the DPC assists in that task of statutory interpretation and I leave any controversy about the characterisation of the DPC (as well as the characterisation of the fines it imposes) over for determination in the proper forum, namely the substantive proceedings.  In short, I consider it unnecessary and therefore inappropriate for me to determine any question about what LinkedIn terms the "structural" lack of independence or impartiality of the DPC. This on the understanding that there is no dispute but that the requirements of the Charter and the GDPR are applicable to, and must guide and inform the interpretation of, sections 142 and 150.

19. The second topic addressed in submissions but not decided here is the relevance of Article 37(1) of the Constitution as interpreted in the judgment of the Supreme Court in Zalewski v. Adjudication Officer [2022] 1 IR 421.  This seminal judgment is undoubtedly part of the landscape in which these proceedings and the judicial review proceedings will fall to be determined.  The judicial review proceedings clearly engage these questions (amended statement of grounds at [34]).   It is not however properly part of the preliminary issues in these proceedings.  The position advanced by LinkedIn in oral submissions is that I do not have to decide whether the process before the DPC constitutes the administration of justice within the meaning of Article 34 of the Constitution or whether it falls within Article 37(1) as " limited functions and powers of a judicial nature ".  Rather it is said that I should have regard to Zalewski as part of the legal context within which to interpret sections 142 and 150.  The argument that appears to be advanced is that an appeal against error applying Orange as the standard of review is not sufficient to satisfy Zalewski and that this should be factored into the interpretation of sections 142 and 150.  This is a significant question that could have ramifications beyond this case.  I am also of the view that it is not appropriate or necessary to seek to predict or assume the outcome of this aspect of the substantive proceedings.  Indeed, the LRC confirms the uncertainty as to impact of Zalewski:

" ... the implications of Zalewski are somewhat uncertain. The Supreme Court relied on a range of contextual factors, both in terms of setting the ambit of the "administration of justice" under Article 34 and defining the scope of permissible 'limited' functions for the purposes of Article 37. Contextual analysis has many advantages, but one disadvantage in this context is that it can be difficult to say with certainty whether a body is engaged in the "administration of justice" or exercising "limited" functions " (LRC Paper at [1.40].

1. I do not propose to engage in this interpretative task here in the absence of argument by the parties, and in light of the fact that it is squarely for determination in the pending statutory and judicial review proceedings.   In short, it does not appear to be either helpful or necessary to have regard to the Supreme Court judgment in Zalewski or the constitutional questions addressed there in the context of deciding the preliminary issues before me.  In any event - and without any comment on the merits of the arguments made in reliance on Zalewski -  my decision is that the correct interpretation of section 142 is not based on the application of Orange, and that appears to remove the only anchor for LinkedIn's reliance on Zalewski.

2. This is not a question which I need to determine, given my decision that an appeal against error applying Orange is not the appeal provided for in the 2018 Act.   The question of whether or how Zalewski would affect that decision is one which I do not need to - and therefore should not - answer or even venture a view on in this judgment.

(c) Curial Deference

1. The topic of curial deference recurs throughout the submissions of the parties.  It is important to identify with clarity what they say it means and how they contend it intersects with the different forms of appeal being discussed.

(i)  Parties' positions

1. LinkedIn's primary case is that there is no scope for curial deference in an appeal against the DPC's decision.  It is said that the nature of the appeal for which LinkedIn contends is incompatible with such deference.  In LinkedIn's written submissions the position presented is that the court is " not obliged to defer to the decision of the DPC " and it is asserted that a court must not be constrained on grounds of deference or margin of discretion from reviewing findings by the DPC.  This is said to be necessary to satisfy Article 6(1) of the ECHR and/or Article 47 of the Charter.  It was however conceded in oral submissions that, if the primary case was rejected and there was to be scope for deference, it should only play a limited, minor role in areas of specific expertise (as countenanced in Xerico) and without altering the appropriate standard of review.   As part of this alternative argument, it was submitted that affording any such deference would be a matter for the judge at trial to decide on the basis of the specific points being argued and that no deference is appropriate on points of law or of fact, both of which are for the trial judge to determine.

2. The DPC's position in written submissions is that curial deference is " part of the Orange test " (at [84]), but then it also contends between written and oral submission that deference has its place; that it will not apply to every appeal point; and that whether it arises will depend on the particular errors being asserted and the nature of the DPC determination being assessed.  It is contended that Orange " affords some level of deference " to the DPC on appropriate matters (written submissions at [3(b)]), but then it is also said that an appeal on the record does not necessarily exclude deference (citing Xerico) and that the Orange test can apply even if there is no room for deference.  So, while deference is part of the Orange test, Orange can apply independently of deference and vice versa.  This points towards a distinction between the question of deference and the applicable standard of review.  The DPC relies heavily on the context of the GDPR, the Charter and the TFEU, contending that they heighten the need for deference, and that the adoption of LinkedIn's position would risk denuding or undermining Article 8(3) of the Charter and Article 16 TFEU and might call into question the rationale for having SAs.  The DPC also contends that curial deference does not clash with Article 78(1) GDPR and that excluding such deference would in fact clash with the judgment in SCHUFA (this has already been considered).

3. The State's stance is that deference is conceptually distinct from the standard of review and that it can have a free-standing role.  It is said that there are areas in which a court is well-equipped to deal with the issues before the DPC such as severity of sanction, but that the need for deference can only be approached on an issue by issue basis rather than as a matter of doctrine.  It cites an increased caution since Orange about paying excessive deference to regulatory bodies.  According to the State, the role of deference will vary depending on the facts and issues at stake and should most appropriately be addressed by the judge presiding over the substantive hearing of the proceedings.

(ii) Interaction between standard of review and curial deference

1. I note the positions of the DPC and the State that the nature or standard of review do not necessarily decide the question of curial deference (or vice versa).   I note too the finding by the Court of Appeal in Xerico that " whatever curial deference might be afforded... that factor cannot be determinative of the nature of the appeal provided for in s. 148AA " (at [107]).

2. It is clear that the question of deference is not necessarily synonymous with the nature or standard of review and that it must be assessed on the basis of specific facts.  The expert nature of the statutory body cannot decide the nature of the appeal.  It is for that reason that I assess the topic of curial deference discretely here.  However, I do have a slight misgiving about the DPC's position that Orange is the universally applicable test on this appeal - including on points of law - while the role of deference is limited to those issues in respect of which it is specifically warranted.  I note too that, in Stanberry Investments Limited v. Commissioner of Valuation [2020] IECA 33 ** (" Stanberry "), Murray J. traced the connection between deference and the development of the Orange test via Canada (Director of Investigation and Research) v. Southam Inc. (1997) 1 S.C.R. 748 and M&J Gleeson (at [47]).  There, Murray J. described the judgment of Kearns J. in M&J Gleeson as positing " what was, essentially, a sliding scale of review by reference to the degree of specialisation of the decision making body. "  In this case, I have already found that sections 142 and 150 do not enshrine a form of appeal against error based on Orange.  If there is a tension between Orange as a universal standard of review and the more nuanced approach to deference heralded in Stanberry, this may support these findings.  However, this is not a case in which there is no conceivable role for curial deference, I have already found that Orange is not the applicable standard of review here and, moreover, this point was not argued before me in those terms, so I do not consider it further here.

(iii) Relevant authorities

1. In Henry Denny & Sons (Ireland) Limited v. Minister for Social Welfare [1998] 1 IR 34 (" Henry Denny "), Hamilton C.J. expressed the following views (in a case in which the appeal was limited to an appeal on a point of law):

"The courts should be slow to interfere with the decisions of expert administrative tribunals. Where conclusions are based upon an identifiable error of law or an unsustainable finding of fact by a tribunal such conclusions must be corrected. Otherwise it should be recognised that tribunals which have been given statutory tasks to perform and exercise their functions, as is now usually the case, with a high degree of expertise and provide coherent and balanced judgments on the evidence and arguments heard by them it should not be necessary for the courts to review their decisions by way of appeal or judicial review."

1. I pause here to observe that Hamilton C.J. carved out of the reach of deference both errors of law and unsustainable findings of fact.  His comments on deference are premised on decision makers having given " coherent and balanced judgments on the evidence and arguments heard by them ".   Murray J. in Stanberry Investments Limited v. Commissioner of Valuation [2020] IECA 33 (" Stanberry") ** also emphasises this formulation: "' curial deference' is thus properly understood as depending on the Tribunal having provided a properly reasoned decision, not as affording a mechanism for compensating where the decision is not so reasoned " (at [52]).

2. In Carrickdale, Laffoy J. considered Henry Denny, among other judgments, noting " a perceptible change in the attitude of the courts to the review of decisions of expert administrative tribunals" with a heightened deference towards the decisions of such bodies.

3. The DPC cites Millar v Financial Services Ombudsman [2015] IECA 127 which - like Ulster Bank Investment Funds Limited v. Financial Services Ombudsman - concerns sections 57CL and 57CM of the Central Bank Act 1942 (as amended).  The appeal in Millar concerned the rejection by the Ombudsman of complaints about increases in variable interest rates being in breach of loan agreements.  The High Court allowed the appeal, a decision which was overturned by the Court of Appeal.  In the course of her judgment Finlay Geoghegan J. noted that no deference should be paid to a decision of the Ombudsman on a pure question of contract law.  As Finlay Geoghegan J. observed, "[w] ith respect to the Ombudsman he does not have expertise or specialised knowledge, certainly relative to the High Court, in deciding questions of law. "  This was said to follow from the judgments in Orange and Ulster Bank v. FSO.

4. Another important judgment is that of the Court of Appeal (Murray J., with Haughton and Power JJ. in agreement) in Stanberry Investments Limited v. Commissioner of Valuation [2020] IECA 33 ** (" Stanberry "). The appeal provisions at issue were sections 39(1), (2) and (5) of the Valuation Act 2001, which provided for a case to be stated to the High Court on a point of law.  The form of appeal and the context of the case (the valuation of car parks) have no real resonance here.  One question presented was whether the Tribunal had made an error of fact that was vitiated by significant errors of fact amounting to an error of law.  The following observations are important:  First, as found in Millar, there is no scope for deference on pure questions of law. Second, " errors of fact simpliciter do not present any issue of curial deference either. "  Third, in judicial review proceedings and appeals on a point of law, the deference to the expert body is built into the procedure and the test for review and " [i]t follows that in both judicial review proceedings, and appeals on a point of law, the scope for 'deference' is limited " (at [49]). Fourth, the Oireachtas in prescribing an appeal on a point of law, must be assumed to do so on the basis that issues arising before the valuation tribunal " are peculiarly suited to the expert determination of the specialist body " (at [51]).  If the appeal on a point of law presents issues of fact or inference which are within the proper " zones of expertise " of that body (such as methods of valuation or the application of particular valuation concepts), " very significant weight " should be afforded to the decision of the expert body (at [51]).  Finally, there is no " supercharged presumption of validity ".  On the contrary,

"Deference means that in those areas touching on the Tribunal's expertise, the Court should be slow to interfere with the Tribunal's reasoning. It does not mean that where the Tribunal's reasoning is unclear so that there are differing possible interpretations of its decision the Court must simply assume that it was correct in the conclusion it reached" (at [52]).

1. In Xerico, the Court of Appeal noted that deference is not incompatible with an appeal on the record and that it would be a matter for the trial judge to determine whether deference is appropriate and warranted on a particular issue: " I do not propose in this Case Stated to opine on the degree of deference that might be afforded by the Circuit Court to a decision under appeal pursuant to s. 148AA, it being a matter for the Circuit Court (in the first instance) to adjudicate on the level of expertise of either the AOs or the decision maker " (at [105]).  This was caveated by the observation that, in an appeal on the record, there should be " limited deference " as the court is reaching its own conclusions on the issues.

2. In Central Bank of Ireland v. CD, Barniville P. noted the " statutory requirement of deference " in section 45(6) of the Central Bank Reform Act 2010, which references " expertise and specialist knowledge " possessed by the Central Bank and which a court must take into account when deciding whether to confirm a prohibition notice issued by the Bank.  Addressing this provision, Barniville P. noted " the statutory deference is directed to those areas in which the Central Bank and its officials have expertise and specialist knowledge, such as in relation to the financial system in the State and financial services " and no deference was warranted as to whether there was compliance with the requirements of fair procedures (at [316]).  Accordingly even when curial deference is enshrined in legislation, it is limited to the sphere of specific expertise of the statutory body and has its place (as the DPC itself accepts).

(iv) Article 60

1. Before concluding the assessment of the question of deference, there is a specific feature of the GDPR regulatory regime to which the DPC attaches weight.  The point that is made is that the cooperation, " one stop shop " mechanism reflected in Article 60 GDPR heightens the role for curial deference and that an appeal on the record by a generalist court could undermine the cross-border cooperative mechanism required by the GDPR.   Reliance is also placed on the special position of SAs under Article 16(2) TFEU and Article 8(3) of the Charter. The DPC points to the decision of the CJEU in Case C-645/19 Facebook Ireland Ltd v. Gegevensbeschermingsautoriteit, in which the CJEU describes the " one stop shop " cooperation mechanism between SAs under the GDPR, the objective of which is to ensure consistency and consensus as between SAs.  It is contended a LSA such as the DPC is not " the sole enforcer of the GDPR in cross-border situations " (relying on the opinion of Advocate General Bobek at [111]).  The DPC contends that the form of appeal for which LinkedIn contends is at odds with this heightened deference under the collective decision-making procedure and that it favours the Orange test.

2. LinkedIn disagree that Article 60, the TFEU or the Charter lean against the appeal being an appeal on the record.  It is said that there is no authority for such a proposition and no judgment of the CJEU supporting it; that it cannot be correct; and that the decision of the DPC is a decision of the DPC alone, which is subject to appeal in accordance with domestic law and procedures regardless of the cooperation regime or the status of the DPC under the GDPR or the Charter.  Reliance is placed on the judgment of Mulcahy J. in TikTok Technology Limited v. Data Protection Commissioner [2025] IEHC 619 (" TikTok v. DPC ") (which I address below).   LinkedIn contends that the cooperation regime cannot undermine the right to a remedy or mandate a different standard of review than that which would otherwise be applicable as a matter of domestic law.   LinkedIn also notes the rejection of an argument based on the independence of the SA in SCHUFA.  There, the CJEU found that " guarantees of independence are in no way compromised by the fact that the legally binding decisions of a supervisory authority are subject to full judicial review " (at [65]).

3. The State's position is that the Article 60 process does not make the DPC's decision sacrosanct but may be relevant to the degree of deference to be paid to the DPC's decision in particular circumstances.

Assessment

1. An important judgment on this question, on which LinkedIn places some reliance, is the judgment of Mulcahy J. in TikTok v. DPC.  There, Mulcahy J. considered a contested application for a stay on certain corrective measures that had been ordered by the DPC pending the determination of an appeal under sections 142 and 150.  The primary issue addressed in the judgment was the applicable test. The DPC relied upon the GDPR context and the system of cooperation between SAs to justify a heightened test applied by the CJEU in applications for interim measures.  It was argued that, because of the Article 60 mechanism, the decision challenged was effectively a joint decision of all SAs and, the procedural autonomy of Member States notwithstanding, the same standards that govern interim measures at EU level must apply. The test for which the DPC contended was that established by the CJEU in Cases C-143/88 and C-92/89, Zuckerfabrik Suderdithmarschen v Hauptzollamt Itzehoe and Zuckerfabrik Soest GmbH v Hauptzollamt Paderborn (" Zuckerfabrik "). TikTok by contrast contended for the well-established test applicable to national law measures generally (referred to as "the Okunade test" and derived from Okunade v Minister for Justice [2012] 3 IR 152).

2. Mulcahy J. noted that the decision of an SA is only binding on other SAs (at [148]) and that a challenge to the decision of an SA must be brought in the Member State in which that authority is established (at [153]). ** He also placed considerable emphasis on the procedural autonomy conferred on Member States under the GDPR, observing that, "[i]t is true that the GDPR seeks to ensure consistency in decision making by supervisory authorities, by requiring that they co-operate with each other. By contrast, however, it confers jurisdiction on national courts to provide effective remedies in relation to such decisions, with no reference to co-operation or harmonisation" (at [155]). Mulcahy J noted the important point (also emphasised by LinkedIn here) that the decision of the lead SA remains a decision of that Authority alone.  The conclusion in TikTok v. DPC was that an application for a stay on a decision of the DPC was governed by domestic procedural law and Okunade rather than Zuckerfabrik was the governing test.

3. While the issue in TikTok v. DPC was different to that under consideration here, it is an important confirmation in this jurisdiction that the expertise and independence of SAs and their interaction through the consensus-building cooperation regime are undoubtedly important features of the GDPR system.  It is also an important recognition of the role of Member States' procedural autonomy, the cooperation regime notwithstanding.

4. I take a similar view here. It is a necessary corollary of the " one-stop shop " regulatory system that the lead SA will be subject to effective judicial oversight in its own Member State.  This is underlined by Articles 78, 83(8) and 83(9) GDPR.  When that jurisdiction is invoked, it must be done on the basis of the rules and procedures governing litigation in that home Member State.  That is the procedural autonomy recognised and respected by the GDPR (subject of course to compliance with the requirements of effective judicial protection, equivalence and effectiveness).  I fail to see how Article 60 can alter the rules which a Member State applies to its legal proceedings and no basis in the GDPR (or the judgments of the CJEU opened to me) to form the view that it does.

5. Contrary to the arguments of the DPC, I do not see that the cooperation mechanism of Article 60 has any general, overarching relevance in determining the scope or extent of judicial oversight in a Member State.  In particular, I do not see it as a valid argument against the appeal on the record for which LinkedIn contends.

6. That is not to say that a high degree of deference may not be warranted when another SA provides comments through the Article 60 process on specific technical points within their expertise - there may clearly be a good argument for such deference - but this is something to be assessed on the basis of specific facts and not as a matter of doctrine or general principle.

---

Conclusions

1. In light of the judgments and submissions relied upon, the following propositions may be stated with regard to the question of deference:

• First, " very significant weight " should be afforded to findings of fact or inferences which are within the proper " zones of expertise " of a statutory body (Stanberry at [51]).

• Second, there is no scope for deference as a matter of Irish law in respect of questions of law or errors of fact (Henry Denny, Millar, Stanberry).

• Third, courts are well equipped to assess the reasoning of specialised statutory bodies (per O'Donnell J. in Nowak).

• Fourth, there is no presumption of correctness in a statutory appeal (Stanberry at [52]).

• Fifth, it is a matter for the judge hearing a statutory appeal to assess whether a finding of fact or an inference made by a specialised statutory body, which is properly reasoned and not based on questions of law or errors of fact, but rather on the application of specialised technical expertise, warrants deference on that basis (Xerico at [105]).

• Sixth, the fact that trial judge may determine that deference is owed to the expert body on a specific technical finding or findings does not mean the appeal is not an appeal on the record or even an appeal de novo (Xerico).

• Finally, in the specific context of sections 142 and 150, if and to the extent comments of technical nature are received from CSAs and reflected in the DPC's decision pursuant to Article 60, that may well be a factor that a trial judge may weigh in the balance when determining whether and to what extent to afford deference to the findings made.

1. I am satisfied that these are the correct principles and approaches to apply in this appeal and, most particularly, that it is a matter for the judge at trial to assess whether and to what extent the threshold for affording deference to the DPC on an issue or issues within the specific sphere of expertise of the DPC has been met.

2. In summary, I agree with the position being advanced by the DPC here that deference has its place and needs to be assessed on an issue-by-issue basis.   The State shares this view and LinkedIn falls back on it as an alternative proposition.  These positions evidencing a more qualified approach to curial deference are entirely appropriate, in my view, and are consistent with the evolution in the treatment of curial deference in the judgments that have been summarised here.

ISSUE 4: ADMISSIBILITY OF NEW EVIDENCE AND ARGUMENT

1. Issue 4 concerns the admissibility of new evidence or argument in an appeal under section 150.  It arises in these proceedings as a result of objections which the DPC has raised to the attempts by LinkedIn to rely on expert evidence of Prof. Stephens and technical evidence of Mr Abishek Shrivastava (vice president of product management at LinkedIn).  LinkedIn asserts that this evidence goes to the issue of infringement as well as the imposition of fines.   Neither the report of Prof Stephens nor the affidavit of Mr Shrivastava were opened to me and I make no comment on them.  Certain objections have also been raised by the DPC to what it regards as new arguments.

2. The position of LinkedIn is that, while this is an appeal on the record, there must be a facility to introduce new evidence and argument, whether the appeal is governed by section 142 or section 150.  During oral submissions, it was acknowledged that the threshold for admitting new evidence may be lower under section 142, given the express terms of section 142(2), but that this does not detract from the discretion of a court to admit new evidence in an appeal under section 150.

3. LinkedIn points to the absence of any prohibition in section 150  on the admission of new evidence or argument and relies on what is referred to as "the trend" towards admitting such material on appeal, when in the interests of justice to do so.  The judgment of Finnegan P. in Ulster Bank v. FSO is cited as evidence of this trend.  Reliance is also placed on two judgments of the ECtHR, which are addressed at the end of this section of the judgment.  It is also said that it would be invidious to admit new evidence in respect of the fining element of a decision and not the infringement findings (although I note that, conversely, LinkedIn relies on that precise scenario to justify its position on Issue 1).

4. The DPC by contrast contends that the absence of a provision akin to section 142(2) in section 150 indicates a legislative intention to exclude such evidence or argument in appeals under that provision, citing the maxim expressio unius est exclusio alterius.  The DPC also relies on " the general thread " in FSO/FSPO cases since Ulster Bank v. FSO not to admit new evidence or arguments.  Decisions of McMenamin J. in Hayes v. FSO (unreported,3 November 2008) and Ryan v. FSO (unreported, 23 September 2011), among other judgments, are said to be evidence of this thread.   The DPC also refers to Rotunda Hospital v. Information Commissioner [2013] 1 IR 1 and Minister for Communications, Energy and Natural Resources v. Information Commissioner [2022] 1 IR 1 as supporting its position.

5. The State agrees with LinkedIn that section 150 does not exclude new evidence or argument, but it does contend that the omission of an equivalent to section 142(2), coupled with the proposition advanced that the appeal is an appeal on the record, do lean against the admission of fresh evidence.  This is the position pleaded by the State in the judicial review proceedings: that there is no rule against the admission of new evidence and it will be a matter for the judge presiding over proceedings to make the determination as to admissibility.

6. The State makes an interesting point about the difference between an appeal against fines and appeals against infringements justifying the different treatment of additional evidence: it is said that a decision on infringement relates to past events and there is no reason for a party not to have brought forward all relevant evidence and arguments on infringement before the DPC.  A fining decision, by contrast, may engage proportionality and other considerations of a more forward-looking nature which may justify permitting fresh arguments and material to be adduced.

7. In oral submissions, the State points out the narrow body of European caselaw relied upon by LinkedIn and observes that the absence of any determination of the CJEU on this topic makes it difficult to discern a principle of EU law as to the admissibility of evidence in statutory appeals.

Assessment

1. Section 142(2) provides that,

" The court, on hearing an appeal under subsection (1), may consider any evidence adduced or argument made by the controller or processor concerned, whether or not already adduced or made to an authorised officer or the Commission."

1. As a matter of construction of the language used in the 2018 Act, it is plain that the Oireachtas specifically contemplates that new evidence and argument may be admitted in an appeal under section 142, but makes no such express provision for appeals under section 150.  However, I do not accept the DPC's assertion that  the discretion of a court to admit new evidence is excluded in section 150,  whether pursuant to the maxim expressio unius est exclusio alterius or otherwise .  It would be far-reaching for the Oireachtas to legislate to remove any discretion on the part of the court to admit new evidence or argument in an appeal and questions of compatibility with the ECHR may well arise if there was such a blanket prohibition (as will be seen below).  I certainly do not believe such an intention can be gleaned from the omission in section 150 of a provision similar to section 142(2).

2. Further, the authorities opened by the parties support the existence of such a discretion in statutory appeals.   In M&J Gleeson, Kearns J. allowed each party to call one expert witness, a decision explained as follows: " I do not think that this Court, which is not provided with any equivalent assistance or with the benefit or guide of precedent, should deny itself where necessary the assistance of expert economists."  He further noted, " it could conceivably impair, hamper or render extremely difficult the efficacy of this hearing if the court were to decide in advance that oral evidence from experts is in all cases to be ruled out " (at page 412).

3. I would note here that section 153 of the 2018 Act specifically provides for the admission of expert evidence in proceedings under the Act " as regards any matter calling for expertise or special knowledge that is relevant to the proceedings " (section 53(1)).  While this was not the subject of submissions before me, it does seem to be consistent with the decision in M&J Gleeson on this point.

4. Kearns J. addressed the question of factual evidence somewhat differently (at page 412):

" I do not propose to allow witnesses be called from the applicants, or the notice parties themselves because it seems to me that such evidence as could be given by any such witness could only be by way of revisitation of the primary facts or by way of new evidence. There seems to me to be no good reason why any evidence from the principals themselves could not have been submitted to the Competition Authority prior to the making of its decision. Of course, there is one qualification to what I have just said, namely that I will allow new evidence from such witnesses on the same conditions and those conditions only that the Supreme Court will allow such new or additional evidence to be given in an appeal before it."

1. This points towards a discretion on the part of a court to decide whether to admit new evidence, although not an unlimited one.

2. Another judgment that casts some light on this question is the judgment of McKechnie J. in FitzGibbon, in which the following important - and unassailable - statements were made:

"[88] Even if the Rules of Court did not so provide, I would be perfectly satisfied that the High Court has full jurisdiction to regulate the manner in which issues before it are dealt with: this must follow from the mandatory obligation on every court to ensure that constitutional justice and fair procedures are applied to any justiciable controversy determined by it. This duty takes effect once the court has seisin of the issue and continues until that court becomes functus officio (see the judgment of Walsh J. in The People v. Lynch [1982] I.R. 64, p. 84 and the judgment of the Court of Criminal Appeal in Director of Public Prosecutions v. M.J. [2014] IECCA 21, (Unreported, Court of Criminal Appeal, 3rd June, 2014) at paras. 24 to 26, pp. 10 to 11). This means that in any given case, the court can and will respond to what is necessary to ensure the integrity of a person's rights. It would however entirely defeat the purpose of a limited appeal if the exercise before the High Court routinely became a re-run of what transpired before the Committee. That will not be required in a great number of cases which will be dealt with in the manner provided by the Rules. On occasion however, the giving of oral evidence or the adoption of some other step may be inescapable, if a material issue cannot be otherwise resolved in a satisfactory and just way. Such a decision will of course be one for the appeal court."

1. I would also note the treatment of this question of new evidence in the judgment of Barron J. in Orange.  While rejecting a complaint made about the refusal to admit evidence going to reasonableness in the High Court action, Barron J. did address the benefit of admitting new expert evidence:

" The reason for the admission of expert evidence is to put a court which has to decide whether certain conduct on the part of experts is appropriate or not, into a position where it can understand the parameters of such conduct. If that is in issue, then a court could have no way of determining such an issue in the absence of relevant evidence " (at page 239).

1. Finnegan P. in Ulster Bank v. FSO considered the materials that should be  before a court on a statutory appeal:

" Having regard to my decision on the standard of review it is appropriate that the appeal should proceed on the basis of the materials which were before the Financial Services Ombudsman only. The Court however has a discretion on application to permit further evidence to be introduced where it is satisfied that this is necessary or appropriate in the interest of justice."

1. I regard these judgments are persuasive indications that the admissibility of new evidence and argument in a statutory appeal must be a matter within the discretion of the judge presiding over the proceedings.  This is so whether the appeal is brought under section 142 or section 150(5), although, as both LinkedIn and the State acknowledge, an application to admit new evidence or argument may be easier in a section 142 appeal given the terms of section 142(2).  I can also see merit to the State's view that the possibly easier admission of new evidence in an appeal under section 142 may be justified by the nature of the issues likely to arise.  Questions of proportionality of the fines may require an assessment of current information, for example.  The cases cited above also demonstrate a sound reason why expert evidence may well be admissible in a statutory appeal under either section 142 or section 150, as indeed appears to be contemplated by section 153 of the 2018 Act.

2. Insofar as the DPC cites a " the general thread " in FSO/FSPO cases since Ulster Bank v. FSO not to admit new evidence or argument, I would note that in the first judgment cited in support of this argument, Hayes v. FSO McMenamin J. did express misgivings " as a matter of logic " about new issues being raised on appeal in respect of which the decision maker was not on inquiry.  He noted, " a court should be slow to have regard to a point which might (but was not) have been made at first instance " (at [43]).  However, this was concerned with issues and not evidence and there is no prohibition expressed, more a reservation. On the question of new evidence, he recorded that " no application has been made by the applicant to submit any new evidence in this appeal. Therefore, no such evidence has been adduced " (at [41]).  I do not see it as supporting a move away from the findings in Ulster Bank v. FSO as to the discretion to admit evidence, as the DPC contends.

3. Two of the cases relied on by the DPC involve appeals on a point of law and are distinguishable on that basis (Rotunda Hospital v. Information Commissioner [2013] 1 IR 1 and Minister for Communications, Energy and Natural Resources v. Information Commissioner [2022] 1 IR 1).  The proposition that an appeal on a point of law should not concern points of law which were not before the decision-maker is somewhat different to the treatment of new evidence or arguments in a broader appeal (although the proposition stated in Rotunda Hospital is broader and references " any appeal process ").   It is notable that the points made in those cases concern the raising of arguments and not new evidence.   The focus of the submissions on Issue 4 was more on the admission of new evidence than argument, given that it originated with objections to the admissibility of the reports and evidence of Prof Stephens and Mr. Shrivastava.

4. Another argument made by the DPC is that Chalkor is authority that the admission of new evidence is limited to the fining element of the decision.  I have found that contention not to be well made.  As already noted, I am rather of the view that the judgments in Chalkor and Galp support the admission of evidence that is relevant to the review of legality under Article 263 TFEU.

5. In light of this interpretation of section 150, I do not believe it is strictly necessary to address the ECHR principles on which LinkedIn relies but will for completeness assess the two cases cited.

6. T he first is Produkcija Plus storitveno podjetje d.o.o. v. Slovenia ((Application no. 47072/15, 23 January 2019) ("Produkcija plus ").  This was a decision under Article 6(1) regarding a competition law investigation and decision in which a relevant conflict of fact arose.  The material conclusion by the ECtHR was that:

" the Supreme Court made no reference to any other evidence than the impugned decision itself. It did not hear the evidence requested by the applicant company aimed at proving the [material fact]. In fact, in connection with the submitted evidence, the Supreme Court only referred to the rule that no new evidence and facts were allowed at that stage of the proceedings... Furthermore, despite the applicant company expressly requesting that a hearing be held, the Supreme Court neither acknowledged the request nor gave any reasons for not granting it (see paragraph 13 above, Mirovni Inštitut v. Slovenia no. 32303/13, § 44, 13 March 2018)" (at [58]).

1. The ECtHR concluded that the " the applicant company was deprived of a right to have the factual aspects of the administrative decision issued against it reviewed by the tribunal with full jurisdiction " and that there was a violation of Article 6(1) (at [59] - [60]).

2. A similar approach can be seen in the decision of the ECtHR in Ramos Nunes de Carvalho e Sá v Portugal ( Applications nos. 55391/13, 57728/13 and 74041/13, ** 6 November 2018) on which both parties rely.  That concerned a disciplinary process involving a judge which was found by the ECtHR to be a civil process for the purposes of Article 6(1).   There was a factual conflict regarding a particular telephone conversation, which involved questions of credibility.  There was no hearing before the disciplinary body or the court hearing the appeal and the court did not admit evidence regarding that conversation.  The court instead held that it lacked jurisdiction to gather evidence or examine the disputed facts (even on the basis of the file submitted to it) and was limited to a review of lawfulness of the decision applying a standard of manifest error.

3. The ECtHR found that it was not a criminal process and noted that no violation of the ECHR can be found " if any structural or procedural shortcomings identified in the proceedings before an administrative authority are remedied in the course of the subsequent control by a judicial body that has full jurisdiction " (at [132]). In that case, the ECtHR weighted various factors, including the seriousness of the penalty, the absence of a hearing, the fact that credibility was a decisive element in the case, and evidence was not admitted on appeal, before concluding that the case was not heard in accordance with the civil requirements of Article 6(1) (at [214]).

4. The DPC points out - quite accurately - that these cases are distinguishable, given that they are based on direct factual conflicts, whereas there is no such conflicts identified in these proceedings. I am nonetheless of the view that these ECtHR judgments are an important indication that, if there is a relevant factual conflict in respect of which evidence is sought to be adduced in an appeal from a decision of a statutory body, but a blanket rule of domestic law or procedure prevents that evidence being admitted, there may be a violation of Article 6(1).

5. In this case, I am satisfied that there is no such exclusionary rule in an appeal under sections 142 or 150.  These decisions of the ECtHR are of limited direct relevance, but they do highlight the importance of judicial oversight in ensuring effective protection of rights under the ECHR and - of particular relevance here - they point to the importance of judicial discretion in the conduct of an appeal from an administrative decision-maker and point firmly away from any form of absolute rule prohibiting the admission of new evidence.

6. One discrete issue that has been concerns the impact of the findings regarding admissibility on the other preliminary issues raised.  In this regard, the DPC makes the case that Orange can be applicable even when new evidence is admissible, and LinkedIn similarly argues that an appeal on the record can be supplemented by new evidence without altering the character of the appeal or translating it into an appeal de novo.   I have already referenced under Issue 2 the principle - which I accept - that the admission of new evidence or argument does not dictate the standard of review, although it may well indicate that the appeal is a " hybrid " rather than falling neatly into any one of the FitzGibbon categories.

7. A final comment on Issue 4 arises from the supplemental submissions by the DPC, which refer to misgivings raised in the LRC Paper about the admission of new evidence in a statutory appeal presenting the risk of elongating and delaying the appeal process and placing an additional burden on the court (LRC Paper at [7.176]).  The admission of new evidence in an appeal undoubtedly does present a risk of appeals being extended and expanded, but the Oireachtas has expressly legislated for the possibility of such new material in section 142(2).  Moreover, the risk of delay and elongation must be weighed against the need to ensure a fair process and effective judicial remedy.  I do not believe that a blanket rule in appeals under section 150 against the introduction of new evidence is consistent with those requirements.  This does not mean a carte blanche to make new cases and introduce new material: quite to the contrary.  A reasoned application would need to be made to so expand the record for the purposes of the appeal, whether under the (possibly easier) rubric of section 142(2) or in an appeal under section 150.

Conclusion on Issue 4

1. There can be no presumption or blanket view taken in the abstract about the admission of new evidence or arguments in an appeal under section 150(5) (or indeed section 142).   It will be a matter within the discretion of the judge presiding over the proceedings to assess any application for the admission of new evidence or argument.   The burden may however be a lighter one when such an application is made in an appeal under section 142, given the specific statutory provision governing this question in section 142(2).

CONCLUSIONS

1. For the reasons explained in this judgment, I would answer the preliminary issues presented as follows:

Issue 1: Is the Appellant entitled to appeal the decision of the Respondent dated 22 October 2024 (the "Decision") under section 142 of the Data Protection Act 2018 or are there aspects of the within appeal which can only proceed as an appeal under section 150 of the 2018 Act?

Section 142 of the 2018 Act provides solely for an appeal from a decision to impose a fine.  A decision that there was an infringement or infringements of the GDPR or a decision as to the exercise of other corrective powers under the 2018 Act cannot be appealed under section 142.

Issue 2: What type of appeal does section 142 of the Data Protection Act 2018 provide for and what is the standard of review/assessment to be applied by the Court in an appeal under section 142?

The form of appeal provided for in section 142 of the 2018 Act is an appeal on the record, with the possibility of new evidence or argument being admitted.  Deference may be appropriate in respect of issues which are within the sphere of the DPC's technical expertise.

Issue 3:  If there are aspects of the Appellant's appeal which can only proceed as an appeal under section 150 of the Data Protection Act 2018, is the standard of review/assessment to be applied by the Court the same under sections 142 and 150 of the 2018 Act or does a different standard of review/assessment apply in an appeal under section 150 of the 2018 Act?

The answer to Issue 2 applies equally to an appeal under section 150(5).

Issue 4: Having regard to the answers to the foregoing questions, is the Appellant entitled to rely on evidence adduced and/or arguments made in the Appeal which were not already adduced or made to the Respondent during the inquiry process?

It is matter within the discretion of the court to determine whether to admit new evidence or argument in an appeal under section 150(5), with the burden being possibly lighter when such an application is made in an appeal under section 142.

1. This matter will be listed before me at 10.30 am on 28 April 2026 at which stage the parties can indicate whether there is a need for written submissions on the question of costs and, if so, propose an appropriate timetable for that purpose.

---

BAILII: Copyright Policy | Disclaimers | Privacy Policy | Feedback | Donate to BAILII
URL: https://www.bailii.org/ie/cases/IEHC/2026/2026IEHC235.html