Minister for Immigration and Multicultural Affairs v Rajamanikkam

Minister for Immigration and Multicultural Affairs v Rajamanikkam - [2002] HCA 32 - HCA 2002 case summary — Zoe

[12]

GLEESON CJ. This appeal concerns the meaning and effect of provisions of the Migration Act 1958 (Cth) ("the Act") relating to judicial review of administrative decisions by the Federal Court of Australia on the ground that there was no evidence or other material to justify the making of the decision in question. Those provisions have since been repealed. Even so, the case is of general importance, because they are to the same effect as provisions in the Administrative Decisions (Judicial Review) Act 1977 (Cth) ("the ADJR Act").
The respondents, husband and wife, are citizens of Sri Lanka. They arrived in Australia in May 1996, and applied for protection visas under the Act. Their contention was that the first respondent was a person to whom Australia owed protection obligations on the ground that he was a refugee within the definition of the Refugees Convention and Refugees Protocol, having a well-founded fear of persecution in Sri Lanka. The applications were refused by a delegate of the appellant. The respondents appealed to the Refugee Review Tribunal ("the Tribunal"). The Tribunal affirmed the decision of the delegate. The respondents applied to the Federal Court for judicial review of the Tribunal's decision. They were successful, both at first instance before Einfeld J, and later before the Full Court (Kiefel, North and Mansfield JJ)[1].
The decision of the Full Court was based upon narrower grounds than that of Einfeld J. The present appeal concerns only the reasoning of the Full Court, and that Court's conclusion that the "no evidence" ground of judicial review of the Tribunal's decision had been made out.
The decision of the delegate was made pursuant to s 65 of the Act, which provides that if, after considering a visa application, the Minister is satisfied that the relevant criteria have been met, the visa is to be granted, and if the Minister is not so satisfied, the visa is to be refused. The delegate stood in the position of the Minister. The decision of the Tribunal, reviewing the decision of the delegate, was as follows:

[52]

The nature of a decision, and the process of reasoning that leads to it, will bear upon a contention that there was no evidence or other material to justify the decision, and upon a contention that the decision was based upon the existence of a particular fact. The circumstance that an applicant seeks to satisfy s 476(4)(b) for the purpose of making a case under s 476(1)(g) may also bear upon the level of particularity or generality at which a fact is identified. The higher the level of particularity at which a fact is identified, the harder it may be to demonstrate that the decision was based upon the existence of that fact, or that there was no evidence or other material to justify the decision, especially if the decision is arrived at because of a lack of satisfaction of the existence of a state of affairs.
The Tribunal's decision to refuse visas to the respondents was the consequence of the legal operation of s 65 of the Act upon the conclusion that the Tribunal was not satisfied that the first respondent satisfied the criteria for refugee status; relevantly, that he had a well-founded fear of persecution in Sri Lanka. The reasons why the Tribunal was not so satisfied involved a number of elements, as appears from the passages quoted above. The first respondent was not seen as fitting the profile of someone likely to attract adverse attention from the authorities. He was in receipt of a government pension. He had a long history of government service. He was an elderly, respected, medical practitioner. The Tribunal did not believe his story about being arrested in Trincomalee. That disbelief, in turn, was based on eight cumulative reasons, two of which were found to involve error. In each case, the error consisted in attributing to the first respondent a representation about a circumstantial fact.
The Tribunal's decision to refuse visas was not based upon a finding that the first respondent had not been arrested at Trincomalee in 1996. The Tribunal was not willing to accept the first respondent's evidence to that effect; but it went on to say that, even if his evidence about that had been believed, he could safely return to Point Pedro or elsewhere, and therefore his arrest was not a basis for a well-founded fear of persecution. The Tribunal considered that the first respondent was not of any interest to the LTTE. The view was principally formed by reference to his personal background and characteristics. I cannot accept, as did the Full Court, that the decision was based upon two of the eight reasons given by the Tribunal for doubting the credibility of the first respondent.
The Full Court said:

[55]

There is, however, a more fundamental problem confronting the respondents. The ground of review relied upon is that there was no evidence or other material to justify the making of the decision to refuse visas. The Full Court, in its reasoning, did not address this question separately, but explained why it found the case fell within s 476(4)(b). It did not explain why the case fell within s 476(1)(g).
In seeking to support the decision of the Full Court, counsel for the respondents submitted that s 476(4) is "both a limiting provision and a deeming provision". This submission must be rejected. It depends upon reading the opening words of sub-s (4) as if they read: "The ground specified in paragraph (1)(g) is not to be taken to have been made out unless but shall be taken to have been made out if." There is no warrant for such revision of the text of the statute. Having regard to the approach taken by the Full Court, it is understandable that the need to make such a submission was felt. But the language of sub-s (4) is plain. Sub-section (4) qualifies par (1)(g). It does not add to it. It limits it. It does not advance the argument to speak of sub-s (4) as elaborating par (1)(g). That merely distracts attention from the language of the statute, which involves no ambiguity.
The Act required the Tribunal to decide that visas should be refused if it was not satisfied that the first respondent satisfied the criteria for refugee status. The Tribunal's lack of satisfaction related to whether he could return to Sri Lanka without being persecuted. There was nothing in the evidence or other material that compelled a conclusion that the first respondent would be persecuted if he returned to Sri Lanka. There may be cases in which it could be said that there is no evidence or other material to warrant a lack of satisfaction that a person will be persecuted if returned to a particular country. It might depend upon the country, and the person. But this Tribunal was not satisfied that an elderly, respected, medical practitioner, who had many years of government service, and who was in receipt of a government pension, would be persecuted if he returned to Sri Lanka. It gave a number of reasons for that, which included, but were not limited to, reasons for not accepting him as a credible witness. Most of those reasons were plausible, and have not been shown to involve error. I find it impossible to conclude that there was no evidence or other material to justify the decision which was required by law in the event of such lack of satisfaction.
The only part of the reasoning of the Full Court that throws light on its approach to s 476(1)(g), as distinct from s 476(4)(b), is the Court's summary of the argument for the respondents as being that "there was no evidence or other material to justify the making of the decision … because the decision was based on the existence of particular facts that did not exist." (Emphasis added). That is a non sequitur. It may be that, in some cases, depending upon the nature of the decision in question, the process of reasoning by which it was reached, and the relevance to the decision of a particular finding of fact, it may be a short step from the proposition that the decision was based on the existence of a particular fact or particular facts that did not exist to the proposition that there was no evidence or other material to justify the making of the decision. But in other cases, of which the present is an example, that may be a large step. The Full Court did not explain how it took that step, or explicitly acknowledge the need to take it.

[63]

Provisions identical to ss 476(1)(g) and 476(4) of the Migration Act, as they stood at the relevant time, are to be found in s 5(1)(h) and s 5(3) of the Administrative Decisions (Judicial Review) Act 1977 (Cth) ("the AD(JR) Act"). However, the AD(JR) Act does not contain an exact counterpart to s 476(1)(e). Rather, s 5(1)(f) specifies as a ground of review under the AD(JR) Act that "the decision involved an error of law, whether or not the error appears on the record of the decision".
In Australian Broadcasting Tribunal v Bond[18], in the context of a discussion as to the extent to which the AD(JR) Act permits of the review of factual findings, Mason CJ discussed the relationship between s 5(1)(f) (the "error of law" ground) and s 5(1)(h) (the "no evidence" ground) of that Act. In the course of that discussion, his Honour pointed out that it had been accepted prior to the AD(JR) Act that "the making of findings and the drawing of inferences in the absence of evidence [was] an error of law"[19]. Because of that, his Honour thought the preferable view was that the "'error of law' [ground] in s 5(1)(f) [embraced] the 'no evidence' ground as it [had been] accepted and applied in Australia before the enactment of the AD(JR) Act and [that] the 'no evidence' ground in s 5(1)(h), as elucidated in s 5(3), [expanded] that ground of review in the applications for which pars (a) and (b) of s 5(3) make provision."[20] His Honour added that, as s 5(3)(a) of the AD(JR) Act requires only "an absence of evidence or material from which the decision-maker could reasonably be satisfied that the particular matter was established ... ss 5(1)(h) and 3(a) have the effect of overcoming to a limited extent and in a limited area the restrictions on the traditional 'no evidence' ground"[21].
As already indicated, the "error of law" ground in s 5(1)(f) of the AD(JR) Act is different from the "error of law" ground previously found in s 476(1)(e) of the Migration Act. The terms of s 476(1)(e) made it plain that the "error of law" ground specified in that paragraph extended only to "error[s] involving an incorrect interpretation of the applicable law or an incorrect application of the law to the facts as found by the person who made the decision". Those words are not apt to include the making of factual findings or the drawing of inferences in the absence of evidence. That being so, it is necessary to consider whether the views expressed by Mason CJ in Bond[22] in relation to s 5(1)(h) and s 5(3) of the AD(JR) Act are applicable to ss 476(1)(g) and 476(4) of the Migration Act.
In Bond, Mason CJ expressed the view that the effect of s 5(3) of the AD(JR) Act is "to limit severely the area of operation" of the "no evidence" ground in s 5(1)(h) of that Act[23]. His Honour did not make clear his reasons for that view. Nor did he indicate the meaning or operation which, in his view, s 5(1)(h) would have in the absence of s 5(3) of the AD(JR) Act. It seems to us that it can only be said that s 5(3) limits s 5(1)(h) if the view is taken that, standing alone, s 5(1)(h) is to be construed as bearing some more extensive meaning than that which is suggested by its terms - "no evidence or other material to justify the making of the decision".
The only basis upon which it is possible to give s 5(1)(h) of the AD(JR) Act, if it stood alone, a meaning more extensive than its terms would suggest is that, as the "error of law" ground in s 5(1)(f) of that Act covers the traditional "no evidence" ground, s 5(1)(h) must apply to factual findings which are not reasonably open or for which there is "no sufficient evidence"[24]. On that view, it may well be appropriate to say that s 5(3) of the AD(JR) Act limits the "no evidence" ground in s 5(1)(h) of that Act.
Whether or not the approach taken by Mason CJ in Bond is the correct approach to the "harmonization"[25] of ss 5(1)(f) and 5(1)(h) of the AD(JR) Act, it is not one that can be applied to s 476 of the Migration Act, as it stood at the relevant time. That is because the "error of law" ground in s 476(1)(e) clearly did not include the traditional "no evidence" ground and, thus, provided no basis for giving a meaning to the "no evidence" ground in s 476(1)(g) which was more extensive than its terms suggested. Once that is accepted, there is no reason to approach the words of s 476(1)(g) of the Migration Act on the basis that, although standing alone, they meant more than what they said, their operation was limited by s 476(4) of that Act. The better approach, in our view, is to treat the words of s 476(1)(g) as having introduced a new and discrete ground of review, with its precise content identified in s 476(4) of the Act.
We should add that, notwithstanding the difference between the terms of s 5(1)(f) of the AD(JR) Act and those of s 476(1)(e) of the Migration Act, we see no reason why the "no evidence" ground should not be approached in the same way in both Acts, namely, on the basis that it is a discrete ground of review, the precise content of which is identified by the succeeding paragraphs of sub-s (4). That is not to say that the "error of law" ground in s 5(1)(f) of the AD(JR) Act does not include the traditional "no evidence" ground applied before the enactment of that Act. Rather, it is simply to say that, under the AD(JR) Act, a single finding might involve a reviewable error that falls within both the "error of law" and the "no evidence" grounds of that Act.
So far as concerns s 5(3)(b) of the AD(JR) Act, which was precisely replicated in s 476(4)(b) of the Migration Act, Mason CJ observed parenthetically that that paragraph is directed to the "proof of the non-existence of a fact critical to the making of [a] decision"[26]. Clearly, the word "critical" is appropriate to a finding of the kind referred to in s 5(3)(a) and replicated in s 476(4)(a), namely, a finding as to a matter which is required to be established before the decision in issue can be reached. The question whether it is also appropriately applied to a finding of the kind referred to in s 476(4)(b) depends upon the meaning of the words "based ... on ... a particular fact".
The word "particular" in s 476(4)(b) is of significance. Had that paragraph been expressed in terms of a decision "based on the existence of a fact", it might have been apt to refer to any fact taken into account in the reasoning process leading to the decision in question. The word "particular" indicates that the paragraph is intended to have a more limited operation. And when regard is had to the requirement that the decision be "based ... on ... a particular fact", the paragraph, in our view, is to be understood as referring to a finding of fact without which the decision in question either could not or would not have been reached. In this sense, it is, in our view, appropriate to speak of a "fact critical to the making of the decision"[27].
Whether a decision could or could not have been reached without a particular factual finding may depend either on logic or on the law to be applied. To the extent that a decision could not have been reached without a particular factual finding because of the law to be applied, there may have been some overlap between pars (a) and (b) of s 476(4) although, of course, par (a) imposed a less stringent test.
Whether a decision would or would not have been made without a particular factual finding depends on indications to that effect in the decision, the reasons for decision or the decision making process. And unless it is possible to say on a proper analysis of the decision, the reasons for decision or the decision making process that, had a particular finding not been made, the decision in question would not have been reached, it is, in our view, impossible to say that the decision was based on that finding.

[65]

Dr Rajamanikkam claimed in his application for a protection visa that he had a well-founded fear of persecution for reasons of race, membership of a particular social group or political opinion if he were to return to Sri Lanka. It is not in issue that, if Dr Rajamanikkam's claim had been accepted by the Tribunal, he was entitled to a protection visa. Conversely, if the Tribunal was not so satisfied, he was not entitled to a protection visa[28]. Mrs Rajamanikkam made no separate claim and the success or failure of her application for a protection visa must follow on the success or failure of her husband's.
In support of his claim that he feared persecution if he were to return to Sri Lanka, Dr Rajamanikkam gave an account of events which, he said, he had experienced in Sri Lanka before coming to Australia. To put those events in context, it is necessary to note that Dr Rajamanikkam is a Sri Lankan of Tamil ethnicity who was born in the Jaffna area. After qualifying as a medical practitioner, he was employed by the Sri Lankan Department of Health in various areas of Sri Lanka before returning to the Jaffna area in 1971. From 1971 until 1976, he worked at Point Pedro, which is in the Jaffna area, and, in late 1976, he commenced his own medical practice there. He closed that practice in 1989 and, in 1992, commenced another practice in Trincomalee.
Dr Rajamanikkam claimed that, whilst at Point Pedro, he was harassed by members of the Liberation Tigers of Tamil Eelam ("LTTE") and repeatedly warned by the authorities against having contact with the LTTE. He claimed that the LTTE forced him to treat their members and, on occasions, kidnapped him for that purpose. On one occasion, he said, he was abducted by the LTTE and accused of supporting the Tamil United Liberation Front ("TULF"), a moderate Tamil organisation. He also claimed that when riding a bike in Point Pedro in 1984, he was kicked by members of the Sri Lankan Army.
In 1987, according to Dr Rajamanikkam, the Indian Peace Keeping Force ("IPKF") damaged his surgery and, in 1988, arrested and detained him for three days on suspicion of being an LTTE supporter. He said that he was released after he paid a bribe to a member of the Eelam Peoples' Revolutionary Liberation Front ("EPRLF") who, he said, was working with the IPKF.
Whilst in Trincomalee, according to his claims, Dr Rajamanikkam was forced by the LTTE to treat its members and, in 1995, was taken into and detained in the jungle for that purpose. He was able to leave, he said, on promising to pay SLR 50,000 and, also, to leave Sri Lanka permanently.
Dr Rajamanikkam's central or primary claim was that he feared persecution at the hands of Sri Lankan authorities who, by reason of his origins and activities, either would identify him as a supporter of the LTTE or had already identified him as such. In support of this claim, he gave an account of being arrested and detained by Sri Lankan authorities in 1996. He said in his application that he was arrested as a "Jaffna Tamil who had moved from Jaffna recently". On the basis of this statement, the Tribunal wrongly attributed to Dr Rajamanikkam the claim that, in 1996, he was new to Trincomalee.
According to Dr Rajamanikkam, he was identified during his detention in Trincomalee as an LTTE supporter by persons he believed to be members of EPRLF or of the Tamil Eelam Liberation Organisation, both of which organisations, it seems, supported the Sri Lankan government. He was released, he said, after paying a bribe to one of the persons concerned. Shortly after being released, he travelled to Colombo to obtain an Australian visa and then travelled to Australia.
The Tribunal rejected Dr Rajamanikkam's application for a protection visa on the ground that he had "concocted his primary claims" with respect to events in Trincomalee in 1996. Additionally, the Tribunal expressed the view that, if it was "wrong in [its] assessment that the applicant had no problems in Trincomalee in 1996," it was, nonetheless reasonable for Dr Rajamanikkam and his wife "to relocate to either Colombo, or ... Jaffna or Point Pedro or the Central area".
The issue of relocation to Point Pedro had been considered by the primary decision-maker whose decision was the subject of review by the Tribunal. It was also considered by the Tribunal. One question that arose concerning that issue was whether Point Pedro had been taken over by government forces. In its decision, the Tribunal attributed to Dr Rajamanikkam a statement to the Department that it had not been taken over when, in fact, he made no such statement.
The Tribunal's decision that Dr Rajamanikkam was not entitled to a protection visa was based on its conclusion that he had concocted his claim with respect to events in Trincomalee in 1996. Its conclusion in that regard was, in turn, based on its doubts as to his credibility. Those doubts came about because of what the Tribunal described as "a range of factors".
There were eight factors which led the Tribunal to doubt Dr Rajamanikkam's credibility. In general terms, those factors were:

[75]

Dr Nadasara Rajamanikkam and his wife Mrs Balambikai Rajamanikkam ("the respondents") are Sri Lankan citizens. They arrived in Australia in May 1996. In June of that year they applied for protection visas under the Act[32]. In March 1997 a delegate of the Minister refused their combined application. The respondents immediately applied for review of that decision by the Refugee Review Tribunal ("the Tribunal"). In September 1998 the Tribunal affirmed the decision of the delegate.
The respondents applied to the Federal Court for judicial review of the Tribunal's decision. In November 1999 the primary judge (Einfeld J) upheld their combined application[33]. The Minister for Immigration and Multicultural Affairs ("the Minister") appealed from that judgment to a Full Court of the Federal Court. In August 2000 that Court unanimously dismissed the appeal, although it did so on grounds more limited than those that had been upheld by the primary judge[34]. The Full Court found that two factual matters, relied on by the Tribunal to support its finding that the male respondent lacked credibility, were mistaken. In terms of s 476(4)(b) of the Act, they were facts which "did not exist". The error was held to invalidate the Tribunal's finding that the male respondent had "concocted his primary claims" as to the facts. This required confirmation of the orders of the primary judge. His orders had set aside the Tribunal's decision and remitted the matter to be heard again by the Tribunal, differently constituted.
The primary judge had reached similar conclusions with respect to other facts upon which the Tribunal member had based his decision. However, after special leave was granted to the Minister to appeal to this Court, no notice of contention was filed for the respondents seeking, upon grounds different from those upheld by the Full Court, to sustain the orders requiring redetermination of their combined application for review by the Tribunal. In particular, the respondents did not seek to restore the conclusions of the primary judge with respect to the other factual findings that he decided had been flawed for want of evidence. In this way, the present appeal has been confined to consideration of the legal effect, under s 476 of the Act as it previously stood, only of the concurrent findings of factual error in which the primary judge and the Full Court had agreed. As the argument developed, the entitlements of Mrs Rajamanikkam were treated as dependent upon the outcome of Dr Rajamanikkam's arguments. She sought a protection visa as a member of the same family unit as her husband[35].

[102]

For some time there has been a controversy in Australia as to the exact scope of the "no evidence" ground of review at common law[47]. It is not necessary in this appeal to settle this question for we are here concerned only with the application of a federal statute. Its language, in the then terms of the Act, was clearly derived from the substantially similar provisions of the ADJR Act[48]. The history of the latter has been usefully traced by Wilcox J in Television Capricornia Pty Ltd v Australian Broadcasting Tribunal[49]. I draw upon, but will not repeat, that exposition.
It is clear that the ADJR Act provisions, equivalent to ss 476(1)(g), (4)(a) and (4)(b) of the Act, were intended to reflect, in statutory form, the decision of the House of Lords in Secretary of State for Education and Science v Tameside Metropolitan Borough Council[50], contemporary with the drafting and enactment of the ADJR Act. When Tameside is read, the provenance in particular of s 5(3)(a) of the ADJR Act and of s 476(4)(a) of the Act become plain. That leaves s 476(4)(b) (and s 5(3)(b) of the ADJR Act) to be given meaning. But when those provisions are read against the pre-existing common law it is difficult to escape the conclusion that a purpose of those provisions was to extend and expand the availability of the pre-existing common law remedy of judicial review for want of evidence before the decision-maker.
In the Federal Court - which has the primary responsibility of applying the ADJR Act and (before it was amended) the equivalent provisions in s 476 of the Act in question in this appeal - there have been several attempts to explain how the primary ground for judicial review for "no evidence" (relevantly s 476(1)(g)) is to operate in relation to the elaboration of that ground (relevantly, s 476(4)(b))[51]. This Court was taken to these authorities. They are helpful and I have taken them into account in reaching my own view. However, I do not feel obliged to attempt to describe and distinguish the course that the authorities have taken. The Minister complained that the decision of the Full Court in the present matter represented a departure from the reasoning that had recommended itself to most of the judges of the Federal Court in earlier decisions. He asked for correction.
There is no binding authority of this Court on the relationship between ss 476(1)(g) and (4) of the Act (or of ss 5(1)(h) and (3) of the ADJR Act). A somewhat similar issue arose obliquely during argument in Australian Broadcasting Tribunal v Bond[52]. However, because the applicant for review under the ADJR Act in that case did not rely on a "no evidence" ground, the scope and meaning of s 5(1)(h) of the ADJR Act did not strictly fall for decision. In his reasons in Bond[53], Mason CJ expressly acknowledged that the ground of review in that paragraph had "no direct application here". Nevertheless, he went on to say[54]:

[107]

It may be accepted that the meaning and intended operation of the provisions of s 476(4)(b) of the Act in relation to s 476(1)(g) are not entirely clear. Indeed, they are unclear. It can also be accepted that the Minister presented a number of persuasive arguments for the interpretation that he advanced. I will collect them to show that I have given them due consideration.
First, there is the language and structure of s 476 itself. Upon one view, s 476(1) is the primary grant of the facility of application to the Federal Court. Sub-section (4) is a subordinate provision, intended to elaborate, but not to alter, the scope of the primary ground as stated in s 476(1)(g). Upon this basis, the starting point for a "no evidence" application is to ask whether the requirement of par (g) of sub-s (1) has been fulfilled. The Minister argued that, in the present case, it had not. There was evidence or other material to justify the making of the "decision" in question. This was the "decision" that the Tribunal was not satisfied that Dr Rajamanikkam was entitled to a protection visa. So far as that "decision" was concerned, the mere fact that it was based, in part, upon an assessment of his credibility and that two of eight grounds advanced for reaching that assessment were knocked away, did not mean that there was "no evidence". There remained the six other grounds for the Tribunal's credibility finding, undisturbed by judicial review. Therefore, so the Minister argued, the respondents did not get to first base. The respondents acknowledged, properly, that the requirements of ss 476(1)(g) and (4)(b) were cumulative in their requirements. According to the Minister, if there was some evidence to sustain (that is, "justify") the "decision" of the Tribunal, that was the end of the matter. Judicial review was unavailable.
Secondly, the Minister submitted that it was essential to read the two related provisions together. Clearly, that is correct. Insofar as any interpretation of par (b) of s 476(4) might otherwise appear to give rise to a ground of review on the basis of some mistake in the assessment of a particular item of evidence, the language of that paragraph, and the context in which it appeared, would lead to the contrary conclusion. As to the language, it was suggested that the reference to basing "the decision on the existence of a particular fact" was inapposite to a case where the "decision" in question involved a satisfaction as to the absence of entitlements rather than the affirmative existence of a "particular fact". It is this line of reasoning that has persuaded several judges in the Federal Court that s 476(1)(g) read with s 476(4)(b) has no application where the "decision" in question is to the effect that facts do not exist. The Minister argued that, in this case the relevant "decision" was of such a character. It was that the Tribunal, reviewing the decision of the delegate who stands in the position of the Minister, was not satisfied that Australia has protection obligations to the particular claimant[59]. In this sense, it was argued, the notion of the decision being based on the existence of a particular fact (as s 476(4)(b) contemplates) was inapplicable. The "decision" in question involved the conclusion that the relevant particular fact or facts did not exist - rather than that they existed[60]. One judge of the Federal Court has even described the contrary interpretation as "extremely strained"[61].
Thirdly, the Minister pointed to the context of the provisions in question. They appear in legislation (both the ADJR Act and the Act as it then stood) providing for judicial review. This is a limited form of review for errors of law. It does not involve a full merits review of factual finding. Respectful of their limited role in judicial review proceedings, courts need to resist attempts to press their functions into what amount, in effect, to factual scrutiny of the evidence and the way in which the administrative decision-maker, charged with the responsibility of making the decision, has carried out that task. This Court was reminded of what it had said in Attorney-General (NSW) v Quin[62], and repeated recently in Minister for Immigration and Multicultural Affairs v Yusuf[63].
The Minister argued that, if s 476(4)(b) were given the "expansive" construction upheld by the Full Court and urged by the respondents in this appeal, it would effectively open up every affected administrative decision to scrutiny against the broad criterion of whether some factual error or omission had arguably occurred in reaching a particular conclusion about matters of detailed evidence. It was submitted that this was not the proper province of judicial review and that the detailed judicial analysis of the evidence in the present case illustrated that this could not have been the purpose of s 476(4)(b) of the Act.
Further, it was argued that the conjunction of ss 476(4) and (1)(g) made it clear that the former was intended to narrow, and not expand, the operation of the latter. This was made obvious by the introductory language of s 476(4) by which, presumably whatever else was arguable in an appeal to the generality of the language of s 476(1)(g), that provision - narrow as it already was - "is not to be taken to have been made out" unless one or other of the requirements of pars (a) or (b) of s 476(4) were also satisfied.
According to the Minister, the interpretation of s 476(4)(b), accepted in this case by the Full Court, would permit a provision obviously intended to be restrictive to expand beyond all recognition the "no evidence" ground provided by s 476(1)(g). The added qualification would be distorted into an over-reach that was fundamentally incompatible with the limited role of judicial review envisaged in s 476(1). Moreover, it was inconsistent with the strictly limited enlargement of the former common law position provided in the companion provisions of s 476(4)(a), which gave effect to what the House of Lords had held in Tameside. In short, in the context, the Minister suggested that par (b) of s 476(4) was an unlikely means to work such a revolution in the traditional limits of judicial review for absence of evidence to justify an administrative decision.

[108]

The foregoing are substantial arguments. Various reflections of them have found favour in particular decisions of the Federal Court concerning the problem that is now before this Court. However, for a number of reasons I consider that the preferable construction of the legislation, that is at once consistent with the language and purpose of the Act as it stood at the relevant time and with the decision of the Full Court in the present case, supports the submissions for the respondents.
First, it is clear from an examination of the pre-existing common law, that the purpose of enacting at least the paragraphs of the ADJR Act in question here (and hence of ss 476(1)(g) and (4)(b)) of the Act was not merely to state the existing law but to enlarge its ambit of operation. So much was stated in Bond[64] by Mason CJ. Indeed it is plain on the face of the provisions. The ADJR Act was a major example of reforming legislation. Although the decision of this Court in the present case has significance for the present parties and other transitional cases still affected, the substantial impact of this decision is likely to be upon future cases arising under the "no evidence" provisions of the ADJR Act. There is no reason to give such provisions a narrow reading.
Secondly, as I would read the interaction of s 476(1)(g) with s 476(4), the latter is not a qualification of the application of the "primary" requirement for judicial review stated, as such, in s 476(1)(g), so much as a statement of the content of that application, that is, an exposition of the particular circumstances in which, for these statutory purposes, a "no evidence" ground is taken to apply. Viewed in this light - which appears to be the way Mason CJ treated the ADJR Act equivalent ground in Bond[65] - the statutory "no evidence" ground of judicial review is both wider and more specific than was the case with "no evidence" grounds for judicial review at common law. This does not read s 476(1)(g) out of the Act. It simply gives that paragraph particular content. So much is made clear by the words of s 476(4) themselves. The ground specified in par (g) of s 476(1) is not to be taken to have been made out unless the requirements of par (a) or (b) of s 476(4) are satisfied. So far as operative effect is concerned, therefore, s 476(1)(g) has none unless one of the two points in s 476(4) is satisfied.
Thirdly, once this conclusion is reached, there is no reason to narrow unduly the two grounds mentioned in s 476(4), as the Minister's construction would do. In particular, there is no reason to confine the availability of judicial review by reference to the common law notions that preceded the passage of the Act (and earlier, the ADJR Act). On the contrary, the correct approach to applying such provisions is to give effect to their purposes as apparent in their language. Once this is accepted, there is nothing offensive to the provision of relief by way of judicial review in interpreting s 476(4)(b) of the Act to cover a decision based on a "particular fact" so long as that fact is not some inconsequential or minor fact or item of evidence on the way to reaching the "decision" in question.
It does not involve an undue intrusion of the judiciary into the fact-finding processes of an administrative decision-maker to conclude that where, say, a critical fact is found to exist without any foundation in evidence or other material, the judge, supervising the administrative decision-making process for legal error, will set aside such a flawed decision and require it to be made again, freed from such an error. As Deane J indicated in Bond[66], where a reviewable decision made under a statute by a repository of statutory power is not supported by some probative material properly before the decision-maker, the "decision", resting on such a foundation, will be invalid. In earlier times, at common law, this would have been explained by resort to the fiction that the Parliament could not have "intended" that a conclusion reached, resting on a mistaken view as to the existence of a critical fact, would be a "decision" of the kind for which it had provided. The judge ordering review is confined to a proper and limited role. He or she does not substitute a decision on the facts or an opinion on the merits of the evidence for that made by the repository entrusted with such functions. All that is done is to set aside the flawed "decision" and to require that a true "decision" be reached without a disqualifying basis resting on the existence of a particular fact which the applicant for relief can show did not exist.
Fourthly, the concerns of the Minister that this approach to s 476(4)(b) would open such decisions to undue scrutiny of evidentiary (and thus factual) findings can be met satisfactorily, so long as that paragraph may properly be confined to the non-existence of a fact "critical to the making of the decision". Those words appeared first in the reasons of Mason CJ in Bond[67]. Relevantly, those reasons had the added authority of the concurrence of Brennan J[68] and of Deane J[69]. As I have pointed out, Bond strictly concerned the ADJR Act but not s 5(1)(h), which is the provision equivalent to s 476(1)(g) of the Act here in question. Nevertheless, the confinement of provisions such as s 476(4)(b) to cases where the "particular fact" in question is "critical" to the decision impugned has been accepted in a series of decisions in the Federal Court. Such decisions can be traced to the reasons of Black CJ (with the concurrence of Spender J and Gummow J) in Curragh Queensland Mining Ltd v Daniel[70].
In Curragh, Black CJ sought a textual source for the idea propounded by Mason CJ in Bond - so as to put outside the ambit of judicial review factual findings and evidentiary conclusions at a lower level of importance to the "decision" in question but to admit as proper to judicial review findings and conclusions that could be shown to be "critical" in the sense used by Mason CJ in Bond. If the applicant for judicial review could affirmatively establish that such a "critical" finding was based on facts that could be shown not to exist the "decision" would qualify for review.
According to Black CJ in Curragh, the textual source for this distinction was to be found in the verb "based". A decision will be "based … on the existence of a particular fact" where a finding of that fact is critical to the decision. Where a decision is found to be based on the existence of more than one fact, Black CJ employed a metaphor to explain the circumstances in which judicial review might be granted in such a case. A decision would be "based" on a particular fact even where that fact is but "[a] small factual link in a chain of reasoning … and there are no parallel links"[71]. In the summary of his reasoning, Black CJ said[72]:

[112]

The decision of the Full Federal Court in Curragh, in the context of the corresponding provisions of the ADJR Act, has stood for a decade. It has been applied many times by single judges[75] and by Full Courts of the Federal Court.[76] In earlier proceedings in the Federal Court, the Minister suggested that it was erroneous or inapplicable, at least where the statutory power has not been exercised. So far this argument has not enjoyed success in the Federal Court[77]. In cases similar to the present, the Full Court of that Court has declined to limit, or qualify, Curragh[78].
I acknowledge that metaphors such as "links in a chain" and synonyms such as "critical" can present difficulties in practice, at least in borderline cases[79]. However, such difficulties have not proved insurmountable in the decade since Curragh was decided. On the contrary, the judges of the Federal Court have experienced no apparent difficulty in applying the principle established in that case. After all, it is substantially the one foreshadowed by Mason CJ and Deane J in Bond. In most instances, judges have rejected the argument that the particular fact in issue, said not to exist, was "critical to the making of the decision" in question.[80] Sometimes, whilst accepting that the impugned finding of fact was "critical" they have rejected the contention that the applicant (who bears the burden of doing so) has shown that the relevant fact does not exist[81]. On rare occasions, of which this is one, the Federal Court has found that the decision-maker has "based" the decision on the existence of a particular fact, classified as "critical", and the person affected has proved that such "fact did not exist", warranting judicial review under s 476(1)(g) of the Act, as elaborated by s 476(4)(b). In my opinion, this approach represents a sensible and entirely appropriate application of the statutory provision. It gives content to the provision consistently with its terms. The construction urged for the Minister would severely curtail the operation of s 476(4)(b). There is no reason to do this. The Minister has now procured from the Parliament amendment to delete the provision in cases of this kind. Why, then, would this Court now impose a construction of the statutory language, different from that which the Federal Court has long treated as settled?

[114]

When the foregoing approach is applied to the facts of the present appeal, it follows that the judgment of the Full Court should be confirmed. The Full Court took as its guide the reasons of Black CJ in Curragh.[82] Indeed their Honours recorded that, for the purposes of this case, counsel for the Minister had agreed with counsel for the respondents that what had been said in Curragh, in the context of the provisions of the ADJR Act, applied aptly to the relationship between ss 476(1)(g) and (4)(b) of the Act. The Full Court correctly stated that whether or not a finding as to the credibility of an applicant for a protection visa was a finding of a "particular fact" would depend on the circumstances of the particular case.[83] And that an applicant would still have to surmount the two hurdles established by the terms of s 476 of the Act as it stood at the relevant time:

The decision-maker must have "based" the decision in the existence of the particular fact rendering it, as it is sometimes described, "critical" in the circumstances; and
The applicant must be able to show that the fact did not exist, that is, that there was no evidence or other material concerning the fact before the decision-maker "to justify the making of the decision".[84]

Correctly, the Full Court pointed out[85] that this was not a case where the Tribunal had based its conclusions as to the credibility of Dr Rajamanikkam simply on the impression that he had made as a witness. Different considerations would apply in such a case. Instead, commendably in my view, the Tribunal itemised the grounds on which it had found him to be lacking in credibility. Careful analysis had shown that two of these grounds were incorrect in fact. Correctly too, the Full Court noted the wording in s 476(4)[86]. In using the wording "particular fact", it is suggested that something more "particular" and "factual" than the decision on the ultimate fact in issue was contemplated by the terms of s 476(4)[87]. Yet it still had to be a "particular fact" upon which the decision-maker "based the decision" and the consideration that that fact did not exist had to be such as to deprive the "decision" of the Tribunal of evidence or other material justifying its making[88].
In the circumstances of this case the Full Court considered that these preconditions for the statutory "no evidence" ground had been made out in respect of the two facts that it found did not exist. In my opinion, that conclusion is fully supported by the critical importance of the finding that the Tribunal made about the want of credibility in Dr Rajamanikkam's evidence and the way in which that evidence explained the significance for its reasoning of the identified "factors" (by inference each of the "factors" operating together). Following the decisions at two levels in the Federal Court, this Court should not, in my opinion, perform its own analysis in order to substitute its own version of the facts on which the Tribunal's decision was based[89]. We are here to correct any error of the Federal Court. In doing so we should accept the concurrent findings upon which that Court approached its legal duty.
Against the background of my analysis, it is worth repeating, with emphasis, what the Tribunal said[90]:

[116]

Such credibility was critical to the combined application for protection visas. Without credibility, doubt would be cast on the version of events he gave of the circumstances immediately prior to his departure from Sri Lanka and his arrival in Australia. These circumstances were, in turn, critical to the existence of the "fear" and to establishing its grounds in terms that could attract the operation of the Convention definition of "refugee" contained in the Act. In the event, the Full Court refrained for the most part from using the non-statutory synonym "critical". There was no mention of the metaphorical "links in the chain"[91]. However, it is clear that the Full Court conceived itself to be applying what had been said in Curragh, although by reference to the terms of s 476(4)(b) itself. The Full Court asked itself whether there were other factors independent of the credibility of Dr Rajamanikkam that would sustain the decision of the Tribunal[92] as "based on" a separate factual foundation. Given the critical importance of the issue of credibility to the success of the combined application and the way in which the Tribunal had explained that the eight factors had to be "considered together", it is unsurprising that the Full Court concluded as it did.
The members of this Court might not think that the "critical" factual decision was "based on" the flawed ("non-existent") fact nominated by the Tribunal[93]. But that, in my view, is a factual assessment. It is inconsistent with the list of facts provided by the Tribunal itself as the basis of its decision. All that the respondents ask is that the Tribunal's decision be made again without any weight at all being given to the facts found to have been non-existent. In judicial review, limited to requiring that the process of decision-making not be legally flawed, this Court's task is not to make its own factual assessments. It is to uphold the respondent's right to have such decisions made lawfully by the repository of the fact-finding power, relevantly the Tribunal. Doing so, the Tribunal might well come to the same conclusion in this case. But at least it will then have done so without a "critical" disqualifying flaw in its reasoning. The Full Court recognised this distinction. This Court should maintain it because it lies at the heart of the proper relationship of courts to administrative decision-makers in conducting judicial review.

[118]

The first respondent who is a medical practitioner lived for some years at Point Pedro near Jaffna in north Sri Lanka. The second respondent is his wife. In 1992 the respondents moved to Trincomalee but retained their residence in Point Pedro.
The first respondent and his wife arrived in Australia on 24 May 1996. On 26 June 1996 they lodged applications for protection visas. On 20 March 1997 the applications were refused by a delegate of the Minister. On 1 April 1997, the respondents sought a review of that refusal by the Refugee Review Tribunal ("the Tribunal"). On 29 September 1998 the Tribunal affirmed the decision of the delegate.
An application for review of that decision was filed in the Federal Court. On 19 November 1999 the Court, Einfeld J, upheld the application, set aside the decision of the Tribunal, and remitted the matter for hearing by a different member of the Tribunal.
The Minister appealed against the decision of the Federal Court to the Full Court of the Federal Court. On 3 August 2000 the Full Court (Kiefel, North and Mansfield JJ) unanimously dismissed the appeal. They did so on the basis that the Tribunal purported to rely on two facts critical to its decision which "did not exist" (s 476(4)(b) of the Act). Those two "facts" were the subject of two findings out of some eight in total made by the Tribunal, held to be false, and elements of a concoction on the first respondent's part to enable him to obtain residence in this country.
The substance of the first respondent's claim was that he had been arrested by Sri Lankan government forces in Trincomalee in April 1996 and that he had been released from custody only by paying a bribe and offering to leave the district. During the proceedings before the Tribunal the first respondent was asked why he could not return to Point Pedro or Jaffna. He answered that he was harassed in Point Pedro by the Liberation Tigers of Tamil Eelam ("LTTE") and that the LTTE was still operational. The Tribunal then suggested to him that he would be able to recover his property which he claimed he had handed over with the title deeds to it, to the LTTE before leaving Trincomalee, a claim inconsistent with another assertion by him that he had given his property to his children. The Tribunal pursued the question of the respondent's right to his property at Point Pedro. The first respondent then said that the house might have been bombed, and that his friends had told him that the indications were that it had been. He conceded however that his children would still own the house, and that, according to Sri Lankan law it remained in the children's names.
The Tribunal made a number of findings adverse to the first respondent: that his evidence about his entitlement to, and receipt of a pension was inconsistent and deliberately misleading; his evidence of the details of his arrest and subsequent release in April 1996, specifically as to his place of confinement, who were actually present there, his interrogation and his access to people who could negotiate his release, was simply not credible; and indeed, was fabricated. Nor did the Tribunal accept the first respondent's account of his movements in Sri Lanka, particularly his assertion that he had moved to Jaffna recently when in fact he had lived in Trincomalee for four years. Other matters in respect of which the Tribunal thought the first respondent to have been less than frank were as to the fact and time of payment of the bribe to secure his release, the timing of his visit to Colombo to obtain a visa, and his arrest after he had informed the LTTE that he was leaving the country only six weeks before his departure.
Passages in the first respondent's evidence to the Tribunal which capture both some of its uncertainties and the sense of some of the contradictions in it are as follows:

[184]

Sub-section (1) requires that there be no evidence or other material to justify the making of the decision. Its provisions raise a high hurdle. Here there were eight grounds for the Tribunal's finding on credit, which was, I would accept, critical. Only two of these were erroneously made. The appellant submits that six of its findings adverse to the first respondent were based on evidence or other material well justifying the conclusion that the respondents' claims were unacceptable, and therefore the decision. Against that, the respondents put that the Tribunal relied, for its ultimate finding of a concoction on the part of the first respondent, upon the compendium of the eight matters: that if one, or, as here, two must be subtracted from the compendium, then there was no evidence to justify the making of a critical part of the decision, that is, of a concoction.
With respect to s 476(4)(b), the appellant submits that even if the respondents made good their argument as to the effect of s 476(1)(g) they cannot bring the Tribunal's decision within s 476(4); that this sub-section is concerned with facts which are particular, in the sense of being facts which the Tribunal must find, facts essential to the exercise of a jurisdiction in favour of an applicant such as, the holding of a fear, that the fear is well founded, and it is of persecution for a Convention reason; and not facts such as whether an applicant was untruthful as to a matter of detail, or as to a matter which was not an ultimate one for the conclusion that a Convention ground has been made out.
It is important to distinguish between the exercise by the Federal Court of its jurisdiction to review a decision under the Act, and the exercise of appellate power on an appeal by way of rehearing. In the course of the latter, subject only to the conventional caution exercised by appellate courts with respect to findings on credit and of fact, the appellate court is entitled to reverse a decision of the court or tribunal appealed from on matters of fact as well as law. And, I have pointed out[104] on other occasions, even if there is evidence available to support a decision at first instance, error as to other factual matters may in some circumstances, infect that decision to the extent that it should be set aside. That is not the situation here. The decision of the Tribunal could only be set aside by the Federal Court if the respondents could bring it within the specific provisions of s 476(1)(g) and (4).
Sub-section (4) was obviously not intended to expand the basis for review for which sub-s (1)(g) provided. This is clear from the opening words of the provision which employ the negative expression, "the ground … is not to be taken to have been made out unless …" [emphasis added]. A different interpretation might have been available had the sub-section been expressed in the affirmative to suggest that one, but perhaps not the only basis for the application of s 476(1)(g) would be, the making of a decision based on a non‑existent fact. Sub-section (4) stated:

[189]

The "decision" was not that the first respondent had not told the truth about two matters. It was not even that the first respondent's claims in support of his application were a concoction. That eight of them were not honest claims may have been a reason why the Tribunal regarded the first respondent's application as insufficient to establish a Convention reason, but that there had been a concoction (because of six or eight falsehoods) was not the decision of the Tribunal to which both sub-sections referred.
In order for the respondents to succeed however it was not necessary for the first respondent to show that a non-existent fact formed part of the decision. It would be enough if the decision were shown to be based on a non-existent fact. The Tribunal here did not find, as Courts or Tribunals sometimes do, that one or more, but not all of several facts or matters were sufficient to produce an adverse reaction to the first respondent's credibility. Quite specifically the Tribunal held that it was the cumulative effect of the first respondent's falsehoods that led it do so.
It is important nonetheless to give the words "based the decision" full weight. "Based" certainly implies in my opinion, something more than "had regard to" or "took into account". For s 476(4)(b) to apply the Tribunal must have used the non-existent fact as a base for, or, if a synonym be required, as a foundation for the decision.
It is also necessary to keep in mind the difference that may exist between an erroneous finding of fact and a non-existent fact. Not all of the former will necessarily involve reliance on the latter. An erroneous finding of fact will only involve the basing of a decision upon a non-existent fact if another condition is also satisfied, that an affirmative finding of a fact which does not exist has been made.
Another significant matter to be noted is the use of the emphatic word "particular" in s 476(4)(b). Its presence has three implications: that it is unlikely that the legislature intended a "non-existent fact" to embrace a negative finding by the Tribunal as to factual allegations by a claimant for refugee status; that the particular fact must be a central or critical one, or indeed a fact basic to the decision; and that the reference is to a fact and not an inference from it or other facts.
In Australian Broadcasting Tribunal v Bond[105] Mason CJ discussed ss 5(1)(h) and 5(3) of the ADJR Act. Those provisions, his Honour said, did not exclude the possibility of a "no evidence" ground of review arising as an "error of law" pursuant to s 5(1)(f) but that case did not raise the question which must be answered here. What his Honour did relevantly say was this[106]:

At a glance

Court

Decision date

Before

Catchwords

Source

Catchwords

Judgment (237 paragraphs)

Parties

Legislation Cited (5)

Cases Cited (24)