Harmless Error

Relief for decision-makers?

This article is not written or assisted by AI.

Where a decision is challenged in Judicial Review proceedings, not every error on the part of a decision-maker will result in the quashing of that decision. At that news, the sense of relief in the hall of the conference of the Irish Planning Institute earlier this month was almost palpable.

Very briefly, a decision may not be quashed for error if the error is “harmless” and did not materially affect the decision.

Reference to a 'doctrine' of “harmless error” probably exaggerates the status of this approach but it is a helpful shorthand.

Paragraph references below (§) are to the judgment being discussed under each subheading unless otherwise stated.

Although the doctrine has only relatively recently been declared as a matter of common law in England and Wales, it has only become clearly visible in this jurisdiction even more recently. Perhaps we should not be surprised that this has coincided with concern that planning decisions, especially housing development decisions, are too vulnerable to technical legal objections.

Carrownagowan Concern Group v ABP (No 2) [2024] IEHC 300

This case offered the first clear sighting of the doctrine of harmless error in the Irish jurisdiction.

The primary question in Carrownagowan was whether a development consent should be quashed where there was uncontradicted expert evidence to the effect that an error in Appropriate Assessment (“AA”) made no difference to its conclusion (§1).

The case concerned the Board's consent (308799) for a windfarm approximately 7km northwest of Killaloe in Co Clare (§3/4).

The Applicant alleged inadequate screening for AA because the Board's Inspector had failed to identify a hydrological connection between the development site and the Lough Derg (Shannon) SPA (§170). The Court agreed that there had been an error (§177/9) but then asked whether the error warranted certiorari(quashing of the permission) (§180).

The Court held that this was a harmless error within the meaning of the CJEU case of Altrip C-72/12 (para 49) (§181) on the basis of uncontradicted expert evidence “to the effect that any error [was] harmless and that there [was] a near-zero prospect of any effect, let alone significant effect, on European sites.” (§182) In summary, “an element of the inspector's analysis was incorrect, but the evidence establishe[d] that this [was] immaterial.” (§191(vi))

Doyle v ABP (No 1) [2025] IEHC 158

In Doyle two Board permissions (one on appeal from a decision by Clare County Council) to grant permission for a data centre were challenged. The Board decisions concerned the proposed development of a data centre (314474, the appeal from a grant by the Planning Authority) and an associated 110kV electrical substation (313895, the subject of a separate SID application, §9) on a 60ha site approximately 4km northeast of Ennis, just east of J13 on the M18.

The Board agreed that both decisions should be quashed on the basis that it had failed to consider adequately the effects of the proposed development, specifically “the loss of a bat roost” (emphasis added) (§18 and §21 and see §41).

Despite the Board's concession, the Developer Notice Party sought to continue to defend the Board's decision. (Paid subscribers can read about this – still unusual – possibility in my post of 27 February 2024.) It did so expressly on the basis that the error relating to the bat roost in the Inspector's Report (which the Board adopted, including the error, see Shadowmill v ABP [2023] IEHC 157 §84) comprised “harmless error” (§42/3).

In upholding the Notice Party's submission that the error was harmless and immaterial to the Board's decision, Humphreys J helpfully summarised the law in relation to harmless error.

A decision should not be quashed for error “if the error was harmless and did not materially affect the result” (§45(i)). This is the basic statement of the doctrine and Humphreys J suggests that it is traceable to the Scottish decision of Walton v Scottish Ministers [2012] UKS 44. I consider this to be an uncomfortable stretch to a case very different in law and facts. Humphreys J cites subsequent English cases and then Sliabh Luachra Against Ballydesmond Windfarm Committee v ABP [2019] IEHC 888 but I consider these to be similar stretches.

The firmer foundation to which Humphreys J has repeatedly referred in respect of harmless error (perhaps first in Reid v ABP [2021] IEHC 362 §53) is the CJEU case of Gemeinde Altrip v Land Rheinland-Pfalz, C-72/12 which concerned the interpretation of the EIA Directive and the question of the effect of procedural error making no difference to final result (Altrip, §39).

In perhaps the crucial paragraph of its judgment, the CJEU held that;

“impairment of a right cannot be excluded unless...the court..is in a position to take the view...that the contested decision would not have been different without the procedural defect invoked...” (§53, emphasis added)

“The term “procedural defect” in Altrip has an autonomous EU meaning which..covers any defects committed in the impugned procedure...In any event there is no logical or rational basis to limit the Altripdoctrine to particular kinds of harmless error. Indeed generally speaking it would trivialise judicial review and be an exercise in self-indulgence to quash decisions because of things that make no difference, in the absence of a mandatory provision that compels such an approach.” (Doyle, §45(i))

The decision as to whether an error would have affected the result is for the Court on the facts of each case (§45(ii)).

An error will not be harmless if it gives rise to a “reasonable possibility that the error would have made a difference to the actual outcome.” (§45(iii)) Once an applicant has established the existence of error, it is for the party asserting harmlessness to prove same (§45(iv)). If the applicant establishes even doubt as to harmlessness, the error will not be harmless. These assessments will be made by the Court on the basis of all of the circumstances of the case (§45(v)).

On the facts in Doyle, Humphreys J held that “the conclusion that the inspector's report error was not material is irresistible”. Crucially, the possibility of there being some roosts was built into mitigation measures. The existence of a single roost at the time of the survey “was neither here nor there” (§85). “Potential roosts were accounted for in the surveys, the assessments and the conditions – whether they were actually occupied ant any given time doesn't make any difference...” (§92, emphasis in original)

It is worth noting that the overlooking of one roost was harmless not because it was only a single roost but because although one roost had been overlooked the possibility of roosts generally was built into mitigation measures.

Carrownagowan Concern Group v ABP [2025] IESCDET 9

Following the striking out of some of its Grounds of challenge (see [2023] IEHC 579 which decision was upheld in [2024] IECA 234), the Applicant in Carrownagowan did not succeed in the substantive hearing of what remained of their application for Judicial Review (above). It sought leave to appeal to the Supreme Court, in part against the decision of Humphreys J on the 'harmless error' point.

In considering the leave application, a Panel of the Supreme Court held;