In recent decisions of the Federal Court (Wigney J) and the NSW Court of Appeal (Bathurst CJ), unreasonableness jurisprudence has been relied on to reject the argument that the “illogicality” ground of judicial review is solely concerned with the end result, as opposed to findings or reasoning “on the way”.


Traditionally, for fear of “merits review”, administrative fact-finding was reviewable only where there was literally “no evidence” to support a particular finding of fact. At the forefront of the attack upon this orthodoxy has been the so-called “illogicality” ground of judicial review.

Minister for Immigration and Citizenship v SZMDS (2010) 240 CLR 611 concerned an attack upon the reasoning process that caused the Minister for Immigration to decide that he was not “satisfied” as to the criteria for the exercise of his power to grant a visa. Crennan and Bell JJ observed that:

…the contemporary invocation of “illogicality” or “irrationality” as a basis for judicial review may well have first emerged in Australia, as intimated by Gleeson CJ in Applicant S20, as a reaction to the ouster of the review ground of “Wednesbury unreasonableness” in immigration law. [citation omitted]

This constraint had long since vanished, but the doctrinal split had persisted. Indeed, it gained new significance because Crennan and Bell JJ went on to find that:

…“illogicality” or “irrationality” sufficient to give rise to jurisdictional error must mean the decision to which the Tribunal came, in relation to the state of satisfaction required under s 65, is one at which no rational or logical decision maker could arrive on the same evidence.

It has often been argued, in reliance on these remarks, that illogicality only goes to jurisdiction where the end result, at the “point of satisfaction”, is one at which no logical decision-maker could have arrived on the evidence. This argument – which goes a long way towards folding “illogicality” back into “no evidence” – was apparently accepted by the Full Court of the Federal Court in SZOOR v Minister for Immigration and Citizenship (2012) 202 FCR 1.

By contrast, the High Court in Minister for Immigration and Citizenship v Li (2013) 249 CLR 332 made it clear that where the exercise of a statutory discretion (as opposed to the attainment of a state of satisfaction as a statutory precondition to the exercise of a power) is concerned, a decision will be legally unreasonable not only where the end result is one at which no reasonable decision-maker could have arrived, but also where there is an identifiable “error in reasoning”, including where the decision-maker has “reasoned illogically or irrationally”.

Recent decisions

Some recent decisions have cast doubt not only upon SZOOR, but also the view that Li unreasonableness is limited to the review of the exercise of statutory discretions.

In SZWCO v Minister for Immigration and Border Protection [2016] FCA 51, Wigney J took the view that Crennan and Bell JJ in SZMDS plainly did not limit “illogicality” to the “end result”. Rather, findings or reasoning “on the way” to the ultimate decision were reviewable. Whether any illogicality or irrationality goes to jurisdiction then depends on its “materiality” to the conclusion reached.

Wigney J returned to this topic, and made similar remarks, in Minister for Immigration and Border Protection v SZUXN [2016] FCA 516. But this time his Honour was sitting on appeal from a decision in which the trial judge had characterised certain fact-finding as “objectively unreasonable” and concluded that the entire decision was therefore “legally unreasonable”. This squarely raised a fundamental question as to whether Li unreasonableness has anything to say about fact-finding towards a “state of satisfaction” and, indeed, the same state of satisfaction that was in issue in SZMDS. Yet his Honour did not tackle this question. To the contrary, his Honour began with a discussion of Li unreasonableness and moved seamlessly to a discussion of SZMDS and the “illogicality” ground. Similarly, his Honour concluded that the Tribunal’s decision was not irrational or illogical but that, in any event, any irrationality or illogicality would have been immaterial, such that “it could not be concluded that the Tribunal’s decision was legally unreasonable”.

In Duncan v Independent Commission Against Corruption [2016] NSWCA 143, Bathurst CJ considered the scope of the illogicality ground in the context of findings of corrupt conduct that depended “on the factual finding that the conduct could adversely affect the exercise of official functions by a public official”. In this different statutory context, his Honour specifically considered the same question as Wigney J and others before him. His Honour noted SZMDS, but decided the question wholly by reference to Li. His Honour concluded that:

…the decision in Li, particularly with its emphasis on the fact that the power conferred on a decision-making authority must be exercised according to law and to reason and within the limit of the subject matter, scope and purpose of the statute, is consistent with the proposition that a decision on factual matters essential to the making of a finding by a decision-maker (in this case a finding of corrupt conduct), can be reviewed on the basis that the reasoning which led to the decision was irrational or illogical irrespective of whether the same conclusion could be reached by a process of reasoning which did not suffer from the same defect.

His Honour considered this conclusion to be consistent also with broader Li jurisprudence including Minister for Immigration and Border Protection v Singh (2014) 231 FCR 437 and Minister for Immigration and Border Protection v Stretton (2016) 237 FCR 1, both decided in the Full Court of the Federal Court. Both cases, it may be noted, involved statutory discretions.

At the very least, these recent cases pose some questions. Why should an illogical process of reasoning, of any sort, be immune from judicial review for jurisdictional error unless it can be shown that there is no evidence capable of supporting the ultimate decision? And even if it should be (e.g. to guard against “merits review”), why should there be a higher standard at the jurisdictional threshold (i.e. the attainment of a state of “satisfaction”) than for the exercise of statutory discretions?