Speaking Remarks: The Art of Decision-Making

Administrative Appeals Tribunal, 2018 National Conference

1. Ladies and gentlemen, first let me begin by saying how delighted I am to have received an invitation to participate in this national conference of the Administrative Appeals Tribunal. The invitation is very much appreciated, especially so because it comes from a colleague, the President, the Hon Justice David Thomas. I am conscious, of course, that not all invitations, historically viewed, are so much appreciated. Many years ago, the famous playwright, George Bernard Shaw, delivered an invitation to his old adversary, Winston Churchill, in the following terms:

Come to the opening night of my new play, and bring a friend – if you have one.

Churchill replied: I will come to the second night – if there is one^[1].

2. The topic to be addressed this morning is “The Art of Decision-Making”. In addressing remarks upon that topic, I am not so much concerned with the technique of writing decisions. Rather, I am concerned with the process of decision-making. How are decisions reached? That is not to say that the art of writing is unimportant to the art of decision-making, because I am an absolute adherent of the Chicago University view that writing is thinking. Writing is essential to decision-making. The very process of reaching a conclusion or forming a view almost always happens when complying with the discipline of having to reduce something to writing whether electronically or physically. Of course, impressions of the outcome are almost always formed in the course of a hearing or in the course of examining the relevant material. But the discipline of reaching the decision itself, the ultimate view about the factual and legal merits of the relevant controversy, is, in the main, reached in the process of writing. It is almost always a methodological mistake in deciding anything of consequence involving contested propositions to decide the answer and then go and write up the answer. Those who take that course would, no doubt, disagree but, first, they are wrong because such an approach is against all the serious learning, and second, they are likely to fall foul of one of the heuristic biases to be mentioned later.

3. That is not to say that decision-makers should “over-egg the pudding” or behave as “rabbits caught in the headlights” or turn themselves inside out agonising about what to do. Decisions must be made and we are decision-makers. Also, most decisions need to be made in constrained timeframes. However, it does mean that we all must have the capacity to “self-correct” as Lord Neuberger describes the process of decision-making. I will return to Lord Neuberger’s observations later in these remarks.

4. Today, I want to talk to you about some of the principles which govern how decisions are reached and especially the role of impressions and intuition and the need to take caution to avoid the biases inherent in intuitive or impressionistic decision-making.

5. I then want to talk to you about the notion that “writing is thinking” and I want to take you to a useful work written by three scholars from Chicago University. Anyone who goes to that university to undertake post-graduate work, including doctoral work, is required to read a book called The Craft of Research by Booth, Colomb and Williams. I have the work with me and I invite you to make a note of it^[2]. It is terrific. Its learning can be readily applied to decision-making generally not just the craft of (making decisions about) research.

6. I then want to talk to you about a heresy contained in some recent writing in relation to the implications to be drawn from the High Court’s decision in Minister for Immigration and Citizenship v Li^[3] in relation to judicial review of the exercise of a discretion by an administrative decision-maker.

7. Another worthy topic is reflections upon the role of jurisdictional error in the judicial review of administrative decision-making and, particularly, the constitutionalisation of that role in Australia by reason of Chapter III of the Constitution and the protective reach of that jurisdiction conferred on the Supreme Courts of the States, constitutionally, by and large as a result of some important observations of Justice Gummow and his influence on the jurisprudence in this area. As many of you in this room know, I have relatively recently given a paper on that topic to the Institute which is published^[4], and I am happy to provide it to the organisers electronically for distribution to you. Accordingly, I will not be speaking about that topic today.

8. And finally, today we have already discussed this morning the notion of addressing factual questions, making findings of fact, identifying the questions of law to be addressed, exercising discretions, bias and apprehension of bias and some aspects of jurisdictional error. I will touch on aspects of some of these things but I will do so in the context of addressing the art of decision-making itself.

9. Let me now turn to that topic and some aspects of the literature on cognitive thinking as it relates to the art of decision-making. In 2012, I was at the Folio Book Shop having a look at some of the new publications. I saw in the new publications the most recent edition in paperback form of Professor Daniel Kahneman’s^[5] work entitled Thinking, Fast and Slow which had already been an international bestseller in its earlier edition. The book has been described by one commentator as a “masterpiece” and “one of the greatest and most engaging collections of insights into the human mind” the reviewer had ever read^[6]. It seemed to me that because I am in the “decision-making business”, I ought to read this work. All of you are also in the decision-making business, and you should know about the research. I am not suggesting that every single decision is impacted upon by this qualitative research but it nevertheless has importance for most decision-making. Let me take you to it now.

10. When you are asked the question, what you are thinking about, you can normally answer because you believe you know what goes on in your mind. That is, one conscious thought leads in an orderly way to another conscious thought. However, Daniel Kahneman and his colleague, Amos Tversky, have demonstrated in their research that that is not the only way the mind works. Nor indeed is it, as they show, the typical way it works. Most impressions and thoughts arise in your conscious experience without your knowing at all how they got there. You cannot trace how you came to quickly detect a hint of irritation in a person’s voice over the phone, or so quickly detect an unreceptive facial expression. You don’t know how the cognitive cues that led to those realisations in your mind, were or are engaged. You don’t know, often, how you came to avoid a threat of danger on the road before you became conscious of the threat and having avoided it. The mental calibrations that produce “impressions” and “intuitions” and the many “decisions” that we make, go on in silence in our minds. We are beginning to know much more about these cognitive cues and the true plasticity of the mind.

11. The work of Daniel Kahneman and Amos Tversky is about biases of intuition. Their work is seminal. Daniel Kahneman won the Nobel Prize for his work in this area. The Nobel Prize would have been awarded to Tversky along with Kahneman but for Tversky’s unfortunate death at 59 before the Nobel Prize was awarded. Apart from the obvious outstanding recognition that arises from having been awarded the Nobel Prize, The Economist, in its review of Kahneman’s book, Thinking, Fast and Slow, said this:

As Copernicus removed the Earth from the centre of the universe and Darwin knocked humans off their biological perch, Kahneman has shown that we are not the paragons of reason we assume ourselves to be.

12. So, Kahneman and Tversky’s work is about biases of intuition. However, Kahneman says that the focus on “error” arising out of these biases ought not to distort the reality that most of our judgements and actions are appropriate, “most of the time”. Human behaviour is such that as we navigate our lives, we normally allow ourselves to be guided by impressions and feelings. The confidence we have in our intuitive beliefs and preferences, is usually justified: but not always.

13. There is now an entire field of “decision research” which seeks to understand how we reach decisions and thus, the art of decision-making itself.

14. In this research, Kahneman and Tversky demonstrate that when decision-makers are required to decide a question, they will often have recourse to a “rule of thumb” otherwise known in the literature as a “simplifying heuristic” (or mental shortcut) which enables a decision-maker to make a difficult judgement. The work of Kahneman and Tversky that led to the Nobel Prize is essentially contained in a major paper entitled Judgement under Uncertainty: Heuristics and Biases. The paper, together with another important paper, is reproduced in Kahneman’s book, Thinking, Fast and Slow. I commend the book and the papers to you. I have the book with me for those who might wish to make a note about it^[7]. It may change your life. I will attach a photocopy of the front piece, to these remarks for distribution.

15. The book and the papers describe the simplifying shortcuts of intuitive thinking. It explains 20 biases as manifestations of these heuristics. It demonstrates the role of these heuristics or shortcuts in the art of decision-making and the extent to which they can distort rationality in decision-making.

16. One of these heuristics is called the “resemblance heuristic” which is used to make a prediction about a fact and often, and this is the real point, the mind’s deployment of the heuristic, used to determine the fact, ignores relevant statistical considerations which would ordinarily lead to a different result. For example, in introducing this topic generally, Daniel Kahneman asked a group, as part of a large research cohort, to consider the following proposition and the following question:

An individual has been described by a neighbour as follows: “Steve is very shy and withdrawn, invariably helpful but with little interest in the world of reality. A “meek and tidy” soul, he has a need for order and structure, and a passion for detail”. Is Steve more likely to be a librarian or a farmer?

17. As to the answers given in the survey group, Kahneman reports as follows:

The resemblance of Steve’s personality to that of a stereotypical librarian strikes everyone immediately, but equally relevant statistical considerations are almost always ignored. Did it occur to you that there are more than 20 male farmers for each male librarian in the United States? Because there are so many more farmers, it is almost certain that more “meek and tidy” souls will be found on tractors than at library information desks. However, we found that participants in our experiments ignored the relevant statistical facts and relied exclusively on resemblance [sometimes called association]. They used resemblance as a simplifying heuristic to make a difficult judgement. The reliance on the heuristic caused predictable biases (systematic errors) in their predictions.

18. This example might seem overly simple but it is designed to illustrate a process.

19. Orthodoxy in the social sciences in the 1970s broadly accepted two ideas about human nature. The first is that people are generally rational and their thinking is normally sound, based on rational sequences. The second is that emotions such as fear, affection and hatred were thought to explain most of the occasions on which people depart from rationality. The research of Kahneman and Tversky challenged both of those propositions. The research identifies systematic errors in the thinking of normal people. The research traces these errors to the design of the machinery of cognition rather than to the corruption of rationality or thought by emotion. It should be noted, however, as Kahneman observes in his most recent work, that an important advance has now been made to the effect that emotion looms larger in our understanding of intuitive judgements and choices than was thought to be the case.

20. The importance of Kahneman and Tversky’s work is that the ideas of heuristics and biases have been used extensively in many fields to analyse errors in decision-making including medical diagnosis, legal judgement, intelligence analysis, philosophy, financial analysis, statistical analysis and military strategy.

21. There is no reason to believe that this research does not inform administrative decision-making.

22. There is one aspect of the research which causes me some concern (and no doubt my difficulty with it is entirely my own fault). It is to do with the role of experts. Kahneman and Tversky observe that the intuitions of experts are shown to be “more accurate” (and therefore less prone to systematic error) than those of non-experts. They say that the accurate intuitions of experts are better explained by the “effects of prolonged practice”. They conclude that a “richer and more balanced picture” in decision-making is one in which decision-making is, on the one hand, influenced by these 20 heuristic biases which they have identified as sources of intuitive judgements and choices, and, on the other hand, “skill”. They say that skill and heuristics are “alternative sources of intuitive judgements and choices”. They introduce their conclusion with these observations:

We have all heard such stories of expert intuition: the chess master who walks past a street game and announces “White mates in three” without stopping, or the physician who makes a complex diagnosis after a single glance at a patient. Expert intuition strikes us as magical, but it is not. Indeed, each of us performs feats of intuitive expertise many times each day. Our everyday intuitive abilities are no less marvellous than the striking insights of an experienced firefighter [an example which related to an earlier illustration of a highly experienced firefighter’s intuitive judgement, based on the pungency of the smell of the smoke above the seat of a fire, that the floor above the blaze was about to collapse thus urgently directing all firefighters to quit the building] or physician – only more common.

23. Those notions are then taken up further and result in these observations:

The psychology of accurate intuition involves no magic. Perhaps the best short statement of it is by the great Herbert Simon, who studied chess masters and showed that after thousands of hours of practise they come to see the pieces on the board differently from the rest of us. You can feel Simon’s impatience with the mythologising of expert intuition when he writes: “The situation has provided a cue; this cue has given the expert access to information stored in memory, and the information provides the answer. Intuition is nothing more and nothing less than recognition”.

24. My difficulty with this, so far as it concerns experts, is that it seems to me to deny any role for deep analysis, curiosity, inquiry, critical thinking and all of those features which distinguish great thinkers and great academics at major universities doing truly inquisitive research, let alone experts working in commercial environments trying to solve complex problems in a way that applies training and expertise to solve those complex problems. Expertise is not just a function of seeing things a thousand times, triggering a cue in memory and then instinctively recognising the cue. The notion that expertise is simply a cue, followed by a memory trigger, followed by recognition, followed by an answer, seems to me to misunderstand the true nature of expertise deployed in inquiry. I am sure that Kahneman and Tversky did not come to their Nobel Prize winning conclusions about the 20 dominating biases or heuristics by reason of a cue, a memory trigger and recognition. There is a lot of original thinking in the research and the modelling undertaken by them.

25. I will turn to some particular heuristic biases shortly but before I do so, I should mention the notion of “substitution” or the substitution heuristic. This is said to explain some forms of decision-making and particularly a disposition to immediately default to so-called analogues which provide the answer to the actual question in issue: almost always a very dangerous practice. The notion is that we have a disposition to not answer the actual question in issue but to default to a simplified analogue as a substitution for the real question. Professor Kahneman introduces the discussion in this way:

When confronted with a problem – choosing a chess move or deciding whether to invest in a stock, for example, – the machinery of intuitive thought does the best it can. If the individual has relevant expertise, he or she will recognise the situation, and the intuitive solution that comes to his or her mind is likely to be correct. This is what happens when a chess master looks at a complex position: the few moves that immediately occur to him are all strong. When the question is difficult and a skilled solution is not available, intuition still has a shot: an answer may come to mind quickly – but it is not an answer to the original question. The question that the executive faces [should investments be made in this particular stock] was difficult, but the answer to an easier and related question [do I like the look of this stock?] came readily to his mind and determined his choice. This is the essence of intuitive heuristics: when faced with a difficult question, we often answer an easier one instead, usually without noticing that substitution has occurred.

[emphasis added]

26. Again, this might be thought to be an overly simple example. However, Professor Kahneman is simply seeking to try and illustrate a disposition to default to substitution or analogising when seeking to find an answer to a complex problem. So often we hear a complex question or problem answered or explained by the decision-maker by saying, “it’s like etc …”. The solution is found in the analogue and then used to answer the actual problem under examination.

27. All of this, leads to the following qualification. Professor Kahneman says that the spontaneous search for an intuitive solution involves either a heuristic answer which comes to mind or an expert solution which comes to mind based upon a cue, followed by a memory trigger, followed by recognition. The qualification, however, is that these two methods sometimes fail and decision-makers find themselves “switching to a slower, more deliberate and effortful form of thinking”. This is the “slow thinking” of the title of the work. Kahneman puts it this way:

Fast thinking includes both variance of intuitive thought – the expert and the heuristic – as well as the entirely automatic mental activities of perception and memory, the operations that enable you to immediately know there is a lamp on your desk or retrieve the name of the capital of Russia. The distinction between fast and slow thinking has been explored by many psychologists over the last 25 years. … I describe mental life by the metaphor of two agents, called System 1 and System 2, which respectively produce fast and slow thinking. I speak of the features of intuitive and deliberate thought as if they were traits and dispositions of two characters in your mind. In the picture that emerges from the research, the intuitive System 1 is more influential than your experience tells you, and it is the secret author of many of the choices and judgements you make.

[emphasis added]

28. Much of the writing and research undertaken by Kahneman and Tversky and others is about the workings of System 1 and the mutual influences and intersections between it and System 2 – “a more deliberate and effortful form of thinking”.

29. I now want to identify four of the 20 heuristic biases which have attracted a lot of attention in decision-making which may have more direct applications to decisions made by administrative decision-makers, judges and lawyers generally. Three researchers sought to examine the application of this research to judges and lawyers. They are Professor Chris Guthrie from the Centre for the Study of Dispute Resolution at the University of Missouri. Professor Guthrie is a product of Stanford University and Harvard University. The second is Professor Jeffrey Rachlinski from the Cornell Law School and the third is a United States District Court Judge for the Central District of California, Judge Andrew Wistrich. Judge Wistrich is a product of the Berkeley Law School and the University of Chicago. These researchers undertook a study involving 167 Federal District Court judges. The result of their study is published in the Cornell Law Review^[8].

30. In this study, the researchers tested for the influence of an “anchoring” heuristic. The anchoring bias involves making estimates or decisions about the value of things (assets) based on irrelevant starting points. They also tested for a heuristic of “hindsight bias”. That bias involves perceiving past events to have been more predictable than they actually were. They also tested for the “representativeness” heuristic which involves ignoring important background statistical information in favour of “singular” information. They also tested for the heuristic described as “egocentric bias”. That bias is one in which individuals, especially individuals who perceive themselves to be experienced and perhaps very experienced, over-estimate their own abilities and reach decisions without an applied analytical method to the decision-making. In their paper, the researchers say this at p 784:

We found that each of these cognitive illusions [heuristic biases] influenced the decision-making processes of the judges in our study. Although the judges displayed less vulnerability to two of the illusions than other experts and laypersons, the results demonstrate that under certain circumstances judges rely on heuristics that can lead to systematically erroneous judgments. In short, our study provides empirical support for Jerome Frank’s assertion that “[w]hen all is said and done, we must face the fact that judges [decision-makers] are human”^[9]

[emphasis added]

31. I do not have the time this morning to go through each of these heuristics but I want to mention something about anchoring and hindsight bias. In the paper, the researchers examine the influence of anchors upon jurors. They say that legal scholars have long thought that anchors influence juries and in five separate studies researchers found that plaintiffs’ lawyers’ damages requests influenced the assessment by mock jurors of the appropriate amount of damage to award in civil suits. In one study, mock jurors awarded slightly more than $90,000 when the plaintiff’s lawyer requested $100,000 in damages but when the plaintiff’s lawyer requested $500,000 in damages in the very same case, mock jurors awarded nearly $300,000. The other studies produce the same results but in a more exaggerated form. The researchers suggest that the moral of these five anchoring studies seems to be “ask and ye shall receive” because in each study when more money was requested for damages by the plaintiff’s lawyer, the jurors awarded a greater sum. So pervasive has been the anchoring heuristic in the assessment of the value of assets and damages assessments that major reform has been called for in the United States. The researchers conducted a model or study with the 167 District Court judges to test the extent of the anchoring heuristic in decision-making. The judges were divided into two groups and each group was given a particular factual scenario. One group had a “no anchor condition” and the other group had an “anchor condition”. The difference between the two groups in determining the amount of damages was shown to be statistically significant. The measure of that statistical significance is explained in detail in the paper. It is not necessary for me to set it out here.

32. The second heuristic of some significance is hindsight bias. Hindsight vision is said to result in decision-makers overstating their own ability to have made a future prediction about the likelihood of a past event occurring. Psychologists are the source of the term. It occurs because knowing an outcome causes people to “update their beliefs about the world” with the result that people then rely on these “new beliefs” to generate estimates of what was predictable. However, the psychologists say that this ignores the change in a person’s beliefs brought about by learning the actual outcome. Because courts usually evaluate events after the fact, they are particularly vulnerable to hindsight bias. The researchers illustrate the phenomenon in this way at p 800:

Few judgments in ordinary life require people to assess the predictability of past outcomes, but such judgments are pervasive in the law. Several studies have demonstrated that the hindsight bias influences judgments of legal liability. Kim Kamin and Jeffrey Rachlinski, for example, compared foresight decisions regarding whether to take a precaution against flooding with comparable hindsight evaluations of whether the failure to take this precaution was negligent. They instructed participants judging in foresight to recommend the precaution if they believed that the flood was more than 10% likely to occur in any given year (which was based on a cost-benefit comparison of the precaution and the damage that a flood would likely cause). The researchers told the participants judging in hindsight that the precaution had not been taken and that a flood causing $1 million in damage had occurred. They instructed these participants to find the defendant liable for the flood if the likelihood of the flood, from the perspective of the defendant before the fact, was greater than 10% in any given year. Although both sets of participants reviewed identical information about the likelihood of a flood, the participants reached different conclusions about appropriate defendant behaviour. Only 24% of foresight participants concluded that the likelihood of a flood justified taking the precaution, while 57% of the hindsight participants concluded that the flood was so likely that the failure to take the precaution was negligent. Because of the hindsight bias, the decision to refrain from taking the precaution seemed reasonable to most participants ex ante, but it seemed unreasonable to most participants ex post.

[emphasis added]

33. The researchers say that all the literature shows that “hindsight bias” is “ne of the most robust cognitive illusions” and although “experience reduces the effect of the hindsight bias somewhat, it does not eliminate it”^[10]. The researchers conducting the study with 167 United States judges found that knowing the outcome “significantly affected judges’ assessments”. When predicting the likelihood of something after the fact, decision-makers cannot help but rely on facts that were unavailable before the fact^[11].

34. The work of these researchers is taken further in another analysis published in the Cornell Law Review at Vol 93, p 1. Another useful article which takes up the work of these researchers and the work of Kahneman and Tversky is an article entitled Judging, Fast and Slow: Using Decision-Making Theory to Explore Judicial Fact Determination by Emma Cunliffe, an Associate Professor in the University of British Columbia, Faculty of Law. It can be found at the International Journal of Evidence and Proof (2014) 18 E&P 139. The principles discussed in this article readily apply to any decision-maker, including administrative decision-makers, deciding questions of fact.

35. All of us in this room seek to engage System 2 decision-making which seeks to bring deliberation to determinations of competing propositions. We are concerned with contentious questions which involve assessments of the strength of the evidence. We are committed to avoiding arbitrary or capricious decision-making. Many of you will be familiar with the observation of Lord Chancellor Eldon when he said this of the Court of Chancery (in the context of capricious and arbitrary decision-making):

The doctrines of this Court ought to be as well settled and made as uniform almost as those of the common law, laying down fixed principles, but taking care that they are to be applied according to the circumstances of each case. I cannot agree that the doctrines of this Court are to be changed with every succeeding judge. Nothing would inflict on me greater pain, in quitting this place, than the recollection that I had done any thing to justify the reproach that the equity of this Court varies like the Chancellor’s foot^[12].

[emphasis added]

36. All of us, however, are committed to testing the strength and weaknesses of propositions through reflection. It may well be that decision-makers start with a System 1 intuitive decision but we seek to measure and balance intuitive decisions by bringing to bear System 2 decision-making. By doing so, we seek to “override”, in the language of the literature, one or more of the 20 heuristic biases now well accepted as influences upon cognitive thinking. In particular, we seek to avoid anchoring, hindsight bias, over-reliance upon estimates of our own ability and the resemblance heuristic.

37. The problem confronting most people in this room (and judges) is that often decisions need to be made under time-pressured conditions that contrive to emphasise reliance upon cognitive shortcuts. We must all strive to identify the question properly, understand the statutory context of the decisions to be reached, understand each of the integers which give rise to subsidiary questions which must be addressed and ensure that we bring qualifying reflection and deliberation to decision-making so as to “override” or balance out intuitive responses.

38. Decision-making is often characterised as a contest between the formalists and the realists. The formalists take the position that all decision-making is reached by applying the legal or statutory regime to the facts in a logical and deliberative way. The realists give much greater emphasis to the role of intuition. They say that the art of decision-making is an intuitive process used to reach a conclusion which is then later rationalised using deliberative reasoning. For the realists, decision-makers decide things by “feeling” and not by judgment; by relying on “hunches” and not by reference to rational, critical thinking. The realists say that decision-makers use “deliberative faculties” not only to justify that intuition to themselves but to make it “pass muster”^[13].

39. Recently, Lord Neuberger, in delivering the 2015 F A Mann Lecture had some things to say about the relationship between System 1 and System 2 decision-making (although not using those terms). Lord Neuberger, of course, was speaking in terms of the role of an appellate judge but the principles described in that address provide insight into the art of decision-making in this context. Lord Neuberger said this^[14]:

… there is nothing wrong with judges starting off with the fair and common sense answer and seeing if they can get there –provided they are intellectually honest and legally principled: at [10].

However, [we need to recognise that] there may be more than one arguably right answer, and, having got what you think is the most satisfactory result intellectually, you may then have to ask yourself whether it is the commercially sensible, practical and morally acceptable result – the iterative process at work: at [11].

It is almost inevitable that a judge will form an initial view, either because of his opinion of the legal principles or because of his view of the fair and sensible outcome – or both. However, having formed that view, the judge will then consider the various legal and policy arguments which have been raised, and often some of these arguments will cast doubt on his or her initial view. The reasons for departing from that initial view may be sufficiently telling to justify a change of mind, but it may well transpire that consideration of other arguments causes a modification of that change or even a reversion to the initial view: at [12].

I would suggest that this ability to self-correct is one of the strongest characteristics judges can possess and is likely to help to lead to a just result. The reality is that, in many cases, it is possible to reach more than one conclusion on the facts, which raises the question of what we even mean by the “right” answer. And the more difficult the case, the more true that is, and so it is scarcely surprising that one not infrequently sees sharp differences of opinion between judges in appellate courts: at [13].

[emphasis added]

40. I now want to say something about apprehended bias and guarding against apprehended bias (and indeed, of course, actual bias) in decision-making. This morning, you have heard some remarks about that topic. Let me give you the observations of Lord Hope of Craighead expressed in the House of Lords (before, obviously enough, the establishment of the UK Supreme Court) in Helow v Secretary of State for the Home Department and Another^[15] in identifying the characteristics of the “fair-minded lay observer”.

41. The facts of the case were these. The party that petitioned the court was a Palestinian refugee living in Lebanon. She arrived in the United Kingdom in 2001 and claimed asylum on the ground that she feared that if she were returned to Lebanon she would be attacked by Lebanese and Israeli agents on account of her Palestinian ethnicity and political opinions. She claimed, in particular, that she had been living in the Sabra/Shatila refugee camp when it was attacked in 1982; had later appeared on television to blame the Lebanese and Israeli authorities for the ensuing massacre; and had assisted Belgian lawyers to gather evidence to support a criminal prosecution against the then Israeli Defence Minister. Her claim for refugee status was refused by the Secretary of State for the Home Department and his decision was upheld by an “adjudicator” sitting at Glasgow. The Immigration Appeal Tribunal refused the claimant leave to appeal whereupon she presented a petition to the relevant court for judicial review of the decision under the applicable legislation. The petition for judicial review was dismissed by the “Lord Ordinary”, that is, Lady Cosgrove, the judicial officer known as the Lord Ordinary. The claimant’s advisers then discovered that the Lord Ordinary was a member of the International Association of Jewish Lawyers and Jurists whose magazine circulated to all members. The magazine had carried a number of extreme pro-Israeli articles and pronouncements in particular by the President of that association. The claimant then presented a petition to the court seeking to set aside Lady Cosgrove’s dismissal of the claimant’s judicial review application on the ground that “a fair-minded and informed observer would have concluded that there was a real possibility that the Lord Ordinary was biased by reason of her membership of an association actively antipathetic to the interests with which the claimant was identified”. The House of Lords dismissed the appeal and made some observations about the characteristics of the fair-minded and informed observer.

42. Lord Hope said this^[16]:

1. My Lords, the fair-minded and informed observer is a relative newcomer among the select group of personalities who inhabit our legal village and are available to be called upon when a problem arises that needs to be solved objectively. Like the reasonable man whose attributes have been explored so often in the context of the law of negligence, the fair-minded observer is a creature of fiction. Gender-neutral (as this is a case where the complainer and the person complained about are both women, I shall avoid using the word “he”), she has attributes which many of us might struggle to attain to.

2. The observer who is fair-minded is the sort of person who always reserves judgment on every point until she has seen and fully understood both sides of the argument. She is not unduly sensitive or suspicious … Her approach must not be confused with that of the person who has brought the complaint. The “real possibility” test ensures that there is this measure of detachment. The assumptions that the complainer makes are not to be attributed to the observer unless they can be justified objectively. But she is not complacent either. She knows that fairness requires that a judge must be, and must be seen to be, unbiased. She knows that judges, like anybody else, have their weaknesses. She will not shrink from the conclusion, if it can be justified objectively, that things that they have said or done or associations that they have formed may make it difficult for them to judge the case before them impartially.

3. Then there is the attribute that the observer is “informed”. It makes the point that, before she takes a balanced approach to any information she is given, she will take the trouble to inform herself on all matters that are relevant. She is the sort of person who takes the trouble to read the text of an article as well as the headlines. She is able to put whatever she has read or seen into its overall social, political or geographical context. She is fair-minded, so she will appreciate that the context forms an important part of the material which she must consider before passing judgment.

[emphasis added]

43. The leading Australian authority on the principles to be applied in determining apprehended bias are the principles set out by Gleeson CJ, McHugh, Gummow and Hayne JJ in Ebner v Official Trustee in Bankruptcy (2000) 205 CLR 337, especially at [6] and [7]. For a particularly interesting illustration of the principle, see Expectation Pty Ltd v PRD Realty Pty Ltd and Anor (No 2) (2006) 151 FCR 160.

44. I now want to say something more about the notion that writing is thinking. I am a great believer in the view that writing is thinking. In the course of writing, we tend to see larger patterns in the material we have read. In the course of writing, the decision-maker tends to arrange and rearrange material in ways which provide insights and enable the discovery of new implications, connections and relativities. As the Chicago researchers say, one of the virtues of the discipline of writing is that it enables the decision-maker to “get your thoughts out of your head and on to paper, where you’ll see what you really can think” ^[17] [original emphasis]. The Chicago University authors say this:

Just about all of us, students and professionals alike, believe our ideas are more compelling in the dark of our minds than they turn out to be in the cold light of print. You can’t know how good your [analysis is] until you separate [it] from the swift and muddy flow of thought and fix [your views] in an organised form that you – and your readers – can [see]. In short, we write to … understand better, and evaluate what we think more objectively.

[emphasis added]

45. As decision-makers, you will be conscious that when you write your reasons, you are, in a real sense, writing for others. You are writing for the immediate parties affected by the decision but you are also writing for a wider audience. That may be true generally or it may be true in the more limited sense that your decision may be scrutinised to determine whether there is error and whether jurisdictional error, in particular, has arisen. The principled matter, however, is this. As the Chicago authors say: “When you write for others, you demand more of yourself than when you write for yourself alone” [emphasis added]. They make this observation:

You will understand your own work better when you try to anticipate your readers’ inevitable and critical questions: How have you evaluated your evidence? Why do you think it’s relevant? What ideas have you considered but rejected? [original emphasis]

46. They also say this:

All researchers, including the three of us, can recall moments when in writing to meet their readers’ expectations, they found a flaw or blunder in their thinking or even discovered a new insight that escaped them in a first draft written for themselves. You can do that only when you imagine and then meet the needs and expectations of informed and careful readers.

47. They then usefully say this:

Writing is, finally, thinking with and for your readers. When you write for others, you disentangle your ideas so that you can explore, expand, combine and understand more fully. Thinking for others is more careful, more sustained, more insightful – in short, more thoughtful – than just about any other kind of thinking.

[emphasis added]

48. In undertaking the writing process, it goes without saying that you need to identify the evidence carefully which leads to rational supportable conclusions. You need to report the evidence accurately. You need to be appropriately precise. You need to be conscious of the distinction between claims, a reason proffered for a claim and evidence which supports a claim. It is not necessary for me in this audience to expand upon the claim, reason and evidence analysis. It is enough for present purposes to say that I agree entirely with the remarks of the Chicago researchers which I have quoted to you.

49. I now want to say something about a heresy which is contained in some recent writing about the Li decision^[18]. The High Court in a number of decisions, but most affirmatively in the Li decision, has held that all powers conferred by statute are to be exercised reasonably unless the Parliament, by the enactment, has made it plain that the power may be exercised otherwise. In Li, Hayne, Kiefel and Bell JJ put it this way: “The legislature is taken to intend that a discretionary power, statutorily conferred, will be exercised reasonably^[19]. Li, of course, concerns the exercise of a discretion. The facts in Li are described in the footnote below^[20]. In Li, the discretion was one of whether the particular proceeding before the Tribunal ought to be adjourned or not. The Tribunal refused to adjourn the proceeding. The High Court affirmed the decision of the Full Court of the Federal Court that the exercise of the discretion had miscarried and held that the decision to refuse the adjournment lacked “an evident and intelligible justification” and was thus “unreasonable”^[21].

50. The heresy is the notion that because the exercise of each statutorily conferred discretion or power must be exercised reasonably, as a presumption of law, the question of whether the power has been exercised reasonably is itself a question of law which engages the court’s supervisory powers in an open-ended examination of executive or administrative decision-making, on the facts.

51. The true position is this. There are two different contexts within which the concept of legal unreasonableness is engaged. First, it can be a conclusion reached by a supervising court after identifying an underlying jurisdictional error in the decision-making process. Second, it can also be “outcome focused” without necessarily identifying an underlying jurisdictional error. The latter occurs when the decision has the character of a “choice” made by the decision-maker. The plurality in Li described this second area as giving rise to an inference to be drawn that although the power has been exercised, it seems to “lack an evident and intelligible justification” and thus the “choice” is shown to be arbitrary, capricious or without “common sense”.

52. In circumstances where no reasons are given for the exercise of a power (and particularly a discretionary decision), the supervising court will necessarily focus on the outcome of the exercise of a power in the factual context presented and then assess for itself its justification or intelligibility. The power, of course, must be exercised by the person in whom the Parliament reposes the power. It is not for the court to exercise the power but, nevertheless, the power must be exercised “according to law”. Of course, where there are reasons for the exercise of the power, the court will examine the reasoning process.

53. An important thing to bear in mind about challenges to administrative decisions on the ground of unreasonableness is this. Unlike some other grounds for review of the exercise of power, the reasoning process in review for legal unreasonableness will inevitably be fact-dependent. That is not to say that the importance of the supervising court maintaining an approach which does not involve the substitution of its own judgment for that of the decision-maker, is diminished in any way. It simply recognises that any analysis which involves the notion that a decision “lacks an evident and intelligible justification” involves scrutiny of the factual circumstances in which the power came to be exercised. In that sense, there is a necessary factual inquiry. However, the factual inquiry is bound up in an examination of whether there is an evident or intelligible justification for the exercise of the power or not.

54. So, legal unreasonableness might arise as a conclusion consequent upon a supervising court identifying an underlying jurisdictional error in the sense that we all understand the term “jurisdictional error”. You have already heard something about that concept today so I will not examine that notion except to mention again the paper I am happy to circulate as I mentioned at the outset of these remarks. Legal unreasonableness can also arise when the court cannot identify how the decision was arrived at and thus an inference might be drawn that there is a lack of an evident and intelligible justification for the decision which suggests legal unreasonableness even though no underlying jurisdictional error has been identified. A contention that a decision is said to lack an evident and intelligible justification, will necessarily engage an analysis of the facts. However, recent writing to the effect that because the legislature is taken to intend that a power statutorily conferred will be exercised reasonably, a basis arises for a supervisory court deciding whether the decision was reasonably reached by examining, for itself, the facts going to decision-making and then substituting its own view about the merits of those facts, is simply not correct. Rather, what must be decided is whether there is underlying jurisdictional error or, alternatively, whether an evident and intelligible justification for the decision is lacking such that the decision could be said to be “legally unreasonable”. As to these matters also see Minister for Immigration and Border Protection v Singh [2014] FCAFC 1 where the Full Court identifies the correct approach.

55. Thank you for your attendance. I have kept you too long. I hope you enjoy your conference.

The Hon Justice Andrew Greenwood

Federal Court of Australia

29 May 2018