Our Reasoning
5 In respect of the first of the three possibilities canvassed by the Tribunal (that the applicant was Malou Timissi), we think that a fair reading of the Tribunal's reasoning process shows that it had, at that point, only reached the stage of setting out possibilities in the light of the report from the Document Examination Unit, but had not yet taken into account Mr Norton's report. (Mr Norton was the fingerprint expert.) The Tribunal's reasoning process can and should be seen as a progressive one. At that stage of its review of the evidence, the first possibility had not been dealt with and thus remained open for later consideration in the reasons. Later (at p 16 of its reasons) the Tribunal had regard to Mr Norton's report which had been submitted to the Tribunal after the hearing. The Tribunal found that, in the light of Mr Norton's report, it had to be accepted that the respondent was Faustin Epeabaka from the Congo.
6 From the above extracts it can be seen that the central foundation of his Honour's criticism of the Tribunal's reasoning as being "both illogical and self-contradictory" relates to the paragraph from the Tribunal's reasoning that we have marked "[A]" above. That is, his Honour concluded that once the Tribunal had resolved in favour of the applicant having given truthful evidence about his identity then no "credibility problems" remained in respect of that part of his evidence - see the paragraph from his Honour's reasons that we have marked "[B]" above.
7 We are unable to agree with his Honour's assessment of the Tribunal's reasoning. In our view, his Honour was mistaken in putting to one side the credibility problems that arose out of the eight matters outlined by the Tribunal at pp 13 - 15 of its reasons. These matters can be seen as relating to the two possibilities that the Tribunal was not prepared to rule out: namely, that the respondent might be Faustin Epeabaka and Malou Timissi, in that he might have used both of those identities, and that the respondent's account of his travel to Australia might be substantially false, in that he and Malou Timissi might have collaborated to arrange that journey. These possibilities were open on the material, put forward by the respondent, that showed that he had travelled to Australia with a passport and visa of Malou Timissi.
8 The eight matters which the Tribunal identified were that:
1. The respondent, an educated and intelligent man, did not take the elementary step of at least taking a photocopy of the passport before returning it, with the consequence that there was no means of checking whether the passport was that of another person.
2. It was implausible that the French authorities would have been willing to grant a long-term resident visa to the respondent so promptly after his arrival in France, and would have done so but for the fact that his passport was coincidentally about to expire.
3. It was implausible that the respondent could have taken Malou Timissi's passport without Timissi noticing its absence.
4. It was implausible that Timissi should have mentioned, over dinner conversation, that he had a visa for Australia, but not have mentioned when he was intending to visit Australia.
5. It was implausible that after Timissi recovered his passport he did not use the visa to visit Australia (up to 12 December 1996) despite the trouble he had gone to in obtaining the visa.
6. A letter from Roger Bokolo to the respondent referred to Timissi as being the respondent's Zairean friend and not as Bokolo's friend, as claimed by the respondent.
7. It was implausible that the respondent could have been able to use Timissi's passport, bearing the latter's photograph, and to forge Timissi's signature, so as to deceive the immigration authorities.
8. It was implausible that the Congolese authorities in Paris would, as claimed in Roger Bokolo's letter, have returned the respondent's passport to Congo. So far as they were aware, the respondent was still in France, and they wished to return him to Congo. But by returning the passport to Congo, instead of retaining it in Paris, they were making it more difficult for him to do this.
9 Once one has regard to these matters which (at p 13 of its reasons) the Tribunal described as reasons for questioning the applicant's account of the circumstances surrounding his use of Malou Timissi's passport and his travel to Australia, it seems to us that the two further possibilities which we have set out immediately above remained open to the Tribunal. Accordingly, in our opinion the learned primary Judge was incorrect in concluding that the Tribunal's reasoning was both "both illogical and self-contradictory". We should also note that the Tribunal identified several other issues in respect of which it doubted the respondent's credibility (see pp 17 - 19 of its reasons).
Standard of Proof
10 Counsel for the appellant contended that the primary Judge erred in law in saying, at page 3 of his reasons, that:
"It (the Tribunal) is more likely to arrive at the correct or preferable decision if its obligation is to determine the existence of facts in accordance with the civil standard except in respect of those matters where the nature of what must be decided makes this inappropriate."
11 At the outset it should be said that this passage in his Honour's judgment occurs in introductory paragraphs which lead into his Honour's thesis that the Tribunal fell into error of law because there had been "…a failure to rationally consider probative evidence". The decision of the primary Judge is not based upon the application of a legal test that the Tribunal was obliged to determine the existence of facts in accordance with the civil standard of proof. However, as the nature of the decision-making process which the Tribunal should follow was addressed at length by counsel for the appellant, it is appropriate that this Court deal with the submission that the primary Judge fell into error of law in his statements about the standard of proof.
12 Counsel for the appellant contends that the impugned statement reflects error as it is inconsistent with the reasons for judgment of Brennan CJ, Toohey, McHugh and Gummow JJ in Minister for Immigration & Ethnic Affairs v Wu Shan Liang (1996) 185 CLR 259 ("Wu Shan Liang") at 282 - 283. Their Honours there referred to the terms "balance of probabilities" and "evidence" as being of little assistance in the context of administrative decision-making. They referred to a passage in Fernandez v Government of Singapore [1971] 1 WLR 987 at 993 - 994 where the House of Lords described the application of the term "balance of probabilities" as inappropriate to ascertaining what, if it happens at all, can only happen in the future. Their Honours concluded (at 283):
"We would adopt that reasoning as applicable in the present case. The term 'balance of probabilities' is apt to mislead in the context of s 22AA of the Migration Act [the predecessor of the relevant provisions in this matter] even if it be used in reference to 'what has already happened' [a direct quote from Fernandez]."
13 The impugned sentence from the judgment of the primary Judge must be understood in the context of statements that preceded it. The full paragraph in which the sentence appears reads:
"When deciding a case the Tribunal must have regard to what is an appropriate standard of persuasion. In Sodeman v R (1936) 55 CLR 192 at 216 Dixon J said that the common law only knew of two such standards, that applicable to criminal cases, beyond a reasonable doubt, and that applicable to civil cases, the preponderance of probability. However, Dixon J pointed out that 'questions of fact vary greatly in nature and, in some cases, greater care in scrutinising the evidence is proper than in others, and a greater clearness of proof may be properly looked for'. In Liang the High Court observed that the decision-making processes that are applicable to civil litigation, such as notions of burden of proof and the like, are not always applicable to administrative decision-making: see 185 CLR 282. In some contexts, such as when the Tribunal is seeking to determine what might happen in the future or even what has already happened, the use of the term burden of proof might be misleading. But when the Tribunal is required, as a step in the process of arriving at its decision, to determine whether a fact does or does not exist generally the civil standard should be held to apply to its decision-making with due regard being paid to serious issues: compare Re Letts v Secretary to the Department of Social Security (1984) 7 ALD 1 at 4. Unless the Tribunal is required to apply some standard of proof it is not easy to see how the Tribunal should direct itself in determining whether the evidence before it permits it to make a particular finding of fact. On one view the Tribunal could approach the matter solely by reference to 'natural justice and common sense' (see McDonald at 9) but this does not give a sufficiently clear guide to the Tribunal in my opinion. It is more likely to arrive at the correct or preferable decision if its obligation is to determine the existence of facts in accordance with the civil standard except in respect of those matters where the nature of what must be decided makes this inappropriate."
14 The primary Judge was not advancing the civil standard as one to be applied without exception in migration cases. Rather, his Honour was advocating the use of the civil standard as a guide likely to produce the correct or preferable decision "except in respect of those matters where the nature of what must be decided makes this inappropriate". This is an important qualification.
15 It is commonplace for asylum seekers to support their claim for refugee status with assertions about hostile treatment they have received from a government in the country of their nationality from which they have fled. The circumstances of their departure, and the nature of the allegations they make against that government, may make it difficult for them to establish their allegations to the degree of satisfaction which would be expected of an ordinary litigant pursuing a civil claim in an Australian court. The difficulties of proof which may beset asylum seekers are recognised in paragraphs 195 to 204 of the "Handbook on Procedures and Criteria for Determining Refugee Status" published by the office of the United Nations High Commission for Refugees (re-edited, Geneva, January 1992). These considerations were recognised by Gaudron J in Chan v Minister for Immigration & Ethnic Affairs (1989) 169 CLR 379 at 413 where her Honour said:
"The humanitarian purpose of the Convention, the fact that questions of refugee status will usually fall for executive or administrative decision and in circumstances which will often not permit of the precise ascertainment of the facts as they exist in the country of nationality, serve, I think, to curb enthusiasm for judicial specification of the content of the expression 'well-founded fear' as it is used in the Convention. Perhaps all that can usefully be said is that a decision-maker should evaluate the mental and emotional state of the applicant and the objective circumstances so far as they are capable of ascertainment, give proper weight to any credible account of those circumstances given by the applicant and reach an honest and reasonable decision by reference to broad principles which are generally accepted within the international community." (Emphasis added.)
16 Whilst a decision-maker must have regard to the difficulties of proof which confront an asylum seeker, findings about events in the past which have affected the asylum seeker will be necessary in many if not most cases. In Minister for Immigration & Ethnic Affairs v Guo (1997) 144 ALR 567 Brennan CJ, Dawson, Toohey, Gaudron, McHugh and Gummow JJ said at 578 - 579:
"The course of the future is not predictable, but the degree of probability that an event will occur is often, perhaps usually, assessable. Past events are not a certain guide to the future, but in many areas of life proof that events have occurred often provides a reliable basis for determining the probability - high or low - of their recurrence. The extent to which past events are a guide to the future depends on the degree of probability that they have occurred, the regularity with which and the conditions under which they have or probably have occurred and the likelihood that the introduction of new or other events may distort the cycle of regularity. In many cases, when the past has been evaluated, the probability that an event will occur may border on certainty. In other cases, the probability that an event will occur may be so low that, for practical purposes, it can be safely disregarded. In between these extremes, there are varying degrees of probability as to whether an event will or will not occur. But unless a person or tribunal attempts to determine what is likely to occur in the future in relation to a relevant field of inquiry, that person or tribunal has no rational basis for determining the chance of an event in that field occurring in the future.
Determining whether there is a real chance that something will occur requires an estimation of the likelihood that one or more events will give rise to the occurrence of that thing. In many, if not most cases, determining what is likely to occur in the future will require findings as to what has occurred in the past because what has occurred in the past is likely to be the most reliable guide as to what will happen in the future. It is therefore ordinarily an integral part of the process of making a determination concerning the chance of something occurring in the future that conclusions are formed concerning past events."
17 As the primary Judge in this case observed, unless the Tribunal is required to apply some standard of proof it is not easy to see how the Tribunal should direct itself in determining whether the evidence before it permits it to make a particular finding of fact. In Wu Shan Liang at 294 Kirby J said:
"There is no suggestion in Chan that this Court intended that the evaluation of past facts (as distinct from the speculation on future possibilities) would be based otherwise than on likelihood Chan (1989) CLR 379 at 422. The process of determination involves the delegate's making findings as to primary facts, identifying the inferences which may properly be drawn from the primary facts, as so found, and then applying those facts and inferences to an assessment of the 'real chances' affecting the treatment of the applicant if he or she were to be returned to China."
18 Findings of fact based on likelihood will usually be findings made on the balance of probabilities arising from the available information before the decision-maker. However, when dealing with the claims of an asylum seeker, the available evidence might not imbue findings so made with the degree of confidence that justify the conclusion that an asylum seeker does not have a well-founded fear of being persecuted. It is for this reason that the civil standard cannot be universally applied to the fact finding process in claims of this kind. It is necessary to recognise the risk of error in adopting such a fact finding process, and to make allowance for it. The manner in which this can be done was explained by Kirby J in Wu Shan Liang at 293:
"It is not an error of law for such a decision-maker to test the material provided by the criterion of what is considered to be objectively shown, so long as, in the end, he or she performs the function of speculation about the 'real chance' of persecution required by Chan.
Secondly, the decision-maker must not, by a process of factual findings on particular elements of the material which is provided, foreclose reasonable speculation upon the chances of persecution emerging from a consideration of the whole of the material. Evaluation of chance, as required by Chan cannot be reduced to scientific precision. That is why it is necessary, notwithstanding particular findings, for the decision-maker in the end to return to the question: 'What if I am wrong?' Guo Wei Rong v Minister for Immigration and Ethnic Affairs (1996) 135 ALR 421 at 441, per Einfeld J. Otherwise, by eliminating facts on the way to the final conclusion, based upon what seems 'likely' or 'entitled to greater weight', the decision-maker may be left with nothing upon which to conduct the speculation necessary to the evaluation of the facts taken as a whole, in so far as they are said to give rise to a 'real chance' of persecution."
19 In our opinion, the primary Judge in the present case did not fall into error in expressing the view that the Tribunal was more likely to arrive at the correct or preferable decision by determining the existence of fact in accordance with the civil standard "except in respect of those matters where the nature of what must be decided makes this inappropriate". The statement was intended as a general one, and the circumstances of the case under consideration did not make it necessary to expand upon the important qualification embodied in the statement.
Whether Illogicality Constitutes Error of Law?
20 We have already concluded that the learned primary Judge misunderstood the Tribunal's reasoning, and that such reasoning was not attended by the error that his Honour found, and so it is not strictly necessary for us to consider this point. We do, however, think it desirable that we make some brief observations about it. As appears from the final sentence of the passage we have set out from his Honour's reasons, he took the view that the Tribunal's decision could be set aside if its adoption of an illogical and self-contradictory approach was an error of law in respect of which relief could be obtained under the Migration Act 1958 (Cth). His Honour held that it was, as it amounted to a failure by the Tribunal "to rationally consider the probative evidence that was before it" (see page 12 of his Honour's reasons). His Honour's conclusion that this failure was an error of law was reached principally in reliance on the decision of Brennan J in Pochi v Minister for Immigration & Ethnic Affairs (1979) 2 ALD 33 at 41 - 42, and on the following passage in the reasons for judgment of Deane J (with whom Evatt J agreed) on appeal ((1980) 4 ALD 139 ("Pochi") at 155 - 156):
"In my view, the Tribunal was bound, as a matter of law, to act on the basis that any conduct alleged against Pochi which was relied upon as a basis for sustaining the deportation order should be established, on the balance of probability, to its satisfaction by some rationally probative evidence and not merely raised before it as a matter of suspicion or speculation, or left, on the material before it, in the situation where the Tribunal considered that, while the conduct may have occurred, it was unable to conclude that it was more likely than not that it had. It seems to me that this conclusion follows, as a matter of law, from the authorities referred to and the reasoning advanced by the Tribunal to establish the proposition as a general principle to be observed by it as a matter of administrative practice."
21 The learned primary Judge, after referring to Australian Broadcasting Tribunal v Bond (1990) 170 CLR 321 ("Bond") at 357; R v. District Court; Ex parte White (1966) 116 CLR 644 at 654; Roads Corporation v Dacakis [1995] 2 VR 508 ("Dacakis") at 520 and Waterford v The Commonwealth (1987) 163 CLR 54 at 77, said that he regarded the cited passage from Pochi as binding on him "in the absence of a decision of the High Court that is inconsistent with it", and he concluded that no such decision existed.
22 The present point arises for discussion, of course, in the context of judicial review, and the general principles that limit the scope of judicial review need to be borne firmly in mind. In relation to findings of fact and related questions of illogicality in reasoning, the judgment of Mason CJ in Bond at 355 - 360 provides authoritative guidance. After reviewing the authorities, Mason CJ (with whom on this point Brennan, Toohey and Gaudron JJ agreed) said (at 356):
"Thus, at common law, according to Australian authorities, want of logic is not synonymous with error of law. So long as there is some basis for an inference - in other words, the particular inference is reasonably open - even if that inference appears to have been drawn as a result of illogical reasoning, there is no place for judicial review because no error of law has taken place."
23 In Dacakis at 517 - 520, Batt J examined in detail, and with particular reference to the judgment of Mason CJ in Bond, the question whether want of logic in drawing an inference of fact would of itself constitute an error of law, and concluded that by itself it will not. We agree with Batt J that the position in Australia is as stated by Mason CJ in Bond, and not what might be seen as the broader position articulated by Deane J in Pochi and relied upon by the learned primary Judge in this case (see also Minister for Immigration & Ethnic Affairs v Teo (1995) 57 FCR 194 ("Teo") at 199 - 200).
24 It is nevertheless important to bear in mind the observations of Phillips JA in Powley v Crimes Compensation Tribunal (1996) 11 VAR 146 at 156 - 157 about the use of the word "reasonably" in this context which can, as his Honour suggested in the following passage, be a distraction:
"The word 'reasonable' is used in this context, I suggest, just to emphasise that, when judging what was open and what was not open below, we are speaking of rational tribunals acting according to law, not irrational ones acting arbitrarily. The danger of using the word 'reasonably' lies in its being taken to suggest that a finding of fact may be overturned, on an appeal which is limited to the question of law, simply because that finding is regarded as 'unreasonable'. That is not the law as I understand it, at least in Australia. A finding of fact will be overturned on an appeal on a question of law only if that finding was not open." (Our emphasis.)
25 Moreover, as Batt J pointed out in Dacakis (at 520), although want of logic in drawing an inference will not of itself constitute error of law, it may sound a warning note to put one on inquiry whether there was indeed any basis for the inference drawn. Likewise, want of logic might in some cases also sound a warning note to put one on inquiry whether there was only a purported, and not a real, exercise of the functions entrusted to the decision-maker, as to which see Teo and the cases there cited.
26 We agree with the learned primary Judge that a failure rationally to consider probative evidence is not the same kind of error as making a simple mistake of fact but, in our view, on the current state of authorities in Australia, that difference does not of itself allow for the elevation of such a failure to a mistake of law.