2. Being an Expert and Distinguishing Experts from Frauds.

Around 2,500 years ago, the Greek philosopher Plato (in his dialogue, Charmides) wrote as follows: ‘If someone claims to be a doctor, how can I tell if he really is one?’.

That was a long time ago; but Plato's problem still confronts us. Given the complexity of the universe, the sophistication of our sciences, and the disputes that reign over difficult issues, people often feel – reasonably enough – that they are not as well informed as they need to be. Nor is it transparent who are the most knowledgeable and trustworthy people to consult with on this or that issue. People feel the need to consult experts who can answer their questions or execute various operations, such as repairing an automobile, performing surgery, or guiding one in making a financial investment. In each case one may ask oneself: Whom should I consult with, and whom can I trust the most? Suppose that one aims to launch a start-up company, but a tech specialist is required to run its digital arm. Who should be selected for that job? In each case, the layperson seeks someone with an appropriate kind of expertise. He or she is a novice, who candidly recognizes his own lack of the skills needed to meet with the challenge in question. However, the question then arises: how can such a layperson identify which of the available candidates would do the best job? Which should be chosen? (For present purposes, I ignore possible differences in the fees that potential experts might charge.) How can a layperson tell which of the numerous putative experts would really be good and which would turn out to be mere frauds or duffers?

Before proceeding, we should ask the following preliminary question: what does it take for someone to be an expert (in one or another field or domain)? We are not interested in enumerating all the different areas of expertise, which, of course, are legion. Rather, we are interested in the generic idea of what expertise consists in. What is required for a person to qualify as some kind of expert? Here are two different possible approaches:

1. (1) ‘Reputational approach’. An expert (in some specified field) is someone who has a reputation – at least in some circles – of being knowledgeable and/or skillful in the specified field.

2. (2) ‘Realist approach’. An expert (in some specified field) is someone who genuinely possesses appropriate knowledge and/or appropriate skills. In other words, he or she can correctly answer or resolve the kinds of questions or problems appropriate to that domain.

As between the two approaches to the nature of expertise – ‘reputationalism’ and ‘realism’ – I firmly side with realism. Being an expert does not require a reputation for possessing a high level of knowledge and skill. One can be an expert even if one keeps one's knowledge or skills quite private, rarely if ever displaying credentials or hanging out shingles to advertise one's skills or accomplishments. There can be genuine experts who have no clientele, following, or publicly established record.

We turn, then, to Realism. Realism is a bit tricky, however, because there are two elements, or two criteria, that need to be jointly satisfied for someone to qualify as an expert. The first criterion is a comparative criterion of expertise. This criterion requires of an expert that he/she exceed most other people in terms of knowing more correct answers to questions in the domain and knowing how to solve more problems in the domain than most people do. Meeting this comparative criterion, however, is not sufficient for being an expert. There is also an (admittedly vague) absolute criterion in addition to the initial, comparative criterion. If Terry holds false beliefs about most questions in the relevant domain, and false beliefs about how to solve problems in the domain, this is a very ‘weak’ performance that would disqualify him as an expert. And that disqualification will stand even if a large number of other people are even worse than Terry in terms of knowledge of the domain. In other words, one cannot qualify as an expert simply by there being others who have an even more inaccurate or misguided set of beliefs and problem-solving tendencies. The candidate must have a substantial amount of accurate knowledge and/or problem-solving ability to earn the status of expert.

This very broad characterization of expertise is intended to apply to a wide range of fields or domains, including natural sciences, medicine, law, automobile repair, finance, business management, music history, and so forth.

Although I highlight knowledge as a central component of being an expert, the types of knowledge in question here are not confined to narrowly ‘intellectual’ subjects that are devoted to what is or isn't a ‘fact’. Being an expert can also involve having the ability to produce actions or results, such as curing physical ailments or identifying profitable investment plans.

3. What methods can laypersons deploy to select the best, or most useful, experts?

Now we focus on the question of how to identify the best experts.

(A) Appeals to training and credentials.

Putative experts are commonly people who have received specialized training, which – in the successful cases – may lead to expertise. They may study at universities, law schools, medical schools, technological institutes, and other specialized instructional entities. Such institutions will often confer degrees, licenses, and/or certificates to display in their offices. Seekers of expertise might consult and compare these degrees, licenses, and certificates as potential indicators of the relative quality of the competing experts.

But how much information can a layperson hope to extract by inspecting such licenses or certificates? A layperson may know very little about the relative strengths of the different schools or institutions that trained the putative experts. Moreover, if two competing experts have been trained at one and the same institution, it might still be unclear (to the layperson) who is the cleverest or most competently skilled trainee.

Patients’ evaluations of doctors might be rather unhelpful for different reasons, especially when the patients lack medical knowledge. For example, patients may base their evaluation of a doctors’ expertise more on the doctor's ‘bed-side manner’ than on the level of success relative to the severity of the symptoms. Good bedside manner will not, in general, be a reliable indicator of medical expertise.

(B) Appeal to the numbers.

Confronted with limited information and questionable reliability of a putative expert's knowledge and skills, a layperson might check to determine if the putative expert he initially encounters (expert #1) advocates a similar diagnosis or plan of action as other putative experts advocate. If many of them agree with the opinion of expert #1, whereas only a few concur with expert #2, shouldn't it be obvious that #1 should be selected rather than #2? This would be a case of ‘trusting the numbers’. If most trained people agree that #1 is best, isn't #1 the better choice for the layperson? Unfortunately, matters are not so simple.

How so? Two groups of experts may be products of two different training schools. Suppose each group slavishly follows his/her own training school. It just so happens, however, that one school has more students, and therefore produces more trainees. This is what accounts for the numerical difference in response to the question at hand. However, a larger number does not prove that this group is more trustworthy than its competitor. Appealing to a larger number has merit only if and when the different consultants are independent of one another (as will be explained shortly). But independence does not hold in the present case in question.

The significance of independence can be illustrated by a different example. Suppose that Jones is a juror in a trial. He listens to witness #1 testify that he (#1) personally saw Schmitt commit the crime. This lends some initial support to Schmitt's being guilty. Next, Jones hears witness #2 also testify to Schmitt's guilt. But witness #2 did not personally witness the crime; he only heard about it from witness #1. Does the additional testimony from witness #2 strengthen Jones's evidence for Schmitt's guilt? No. Witness #2's evidence is not independent of witness #1's evidence; rather, it completely relies on witness #1. Therefore, witness #2's evidence adds no additional evidential weight (for juror Jones) with respect to Schmitt's guilt.Footnote ²

(C) Comparing performances in a debate.

An entirely different method for trying to identify a superior expert is to attend, or listen, to a debate between competing experts. Preferably, the debate would be on the very subject matter on which the layperson seeks advice, for example, how best to treat a certain kind of cancer. Again, however, it is unlikely that an otherwise untrained layperson could profit much from hearing such a debate.

The problem is that such a debate is likely to be rife with technical terminology and lines of reasoning that might be familiar to experts but obscure to laypersons. The language in such a discussion would be highly esoteric, where ‘esoteric’ entails being beyond the ken of a typical layperson. It is therefore doubtful that a layperson could glean a great deal of usable, reliable information by watching or listening to such a debate.

(D) Track records

I suggest that the most helpful method – if it is available – is to check the past track records of the candidate experts’ success rates. How often did his or her advice generate an improved result, outcome, or state of affairs? How well did the putative experts prove themselves in the test cases? A track record consisting of a large percentage of successes and few if any failures would be a prime indicator of a candidate expert's quality.

Is it feasible, however, for laypersons to verify the past track records of multiple competing experts? This will depend, of course, on the kinds of tasks or problems that the experts in question have tried to tackle and the answers, or solutions, that they proposed. Are laypersons in a position to assess the accuracy or adequacy of the putative experts’ solutions? Consider another medical example. Suppose a layperson learns that five years ago, Dr. Jones was confronted with the following problem: ‘Patient Johnson suffers from symptoms X, Y, and Z. What would be the best way to treat his condition?’ Suppose that the layperson intends to use this information to assess Dr. Jones's qualifications. Can he/she find out what treatment Jones delivered? That surely seems possible. But will our layperson be able to tell whether Dr. Jones’ treatment was medically sound? That result is problematic, given the layperson's lack of medical training. Of course, the layperson may learn that patient Johnson eventually recuperated from his condition. Still, this would not provide sufficient guidance concerning the quality of Dr. Jones's diagnosis or the treatment of patient Johnson. (The patient may have recuperated all on his/her own.)

I do not mean to argue that appeals to past track-records are a totally fruitless method for addressing our problem. As indicated earlier, I think that tracking a putative expert's track record is a very good method to use, at least when used properly. Indeed, it may be the best method available. Here is an example (Goldman, Reference Goldman2001). Oscar is untutored in matters of astronomy. He hears Sidra claim that there will be a solar eclipse visible from London precisely one year hence. By itself this does not provide much (if any) evidence that Sidra has considerable expertise in astronomy; because as of yet Oscar has no clue as to whether Sidra's prediction will be right. However, suppose Oscar waits a year and then personally observes a solar eclipse in London, precisely as Sidra predicted. This would definitely supply evidence of Sidra's expertise (assuming she did not merely borrow this prediction from some astronomer friend of hers). The point is that this might be a legitimate example of how a novice can acquire genuine support for someone else's expertise, without having transformed himself into an expert.

To take simpler examples, laypersons can sometimes successfully identify cases in which their doctor, their mechanic, their investment advisor, etc. gives them just the sort of treatment, advice, etc. that palpably succeeds in generating sought-after results. To summarize, I do not claim that getting good evidence for various individuals’ expertise is wholly beyond the grasp of laypersons. However, such evidence may be rather difficult to acquire – which is not to deny the value of utilizing such evidence whenever this is feasible.

4. Paying Attention to an Expert's Personal Interests.

An important additional factor to consider, of course, is the self-interest an expert might have in recommending one way of addressing a problem rather than another, for instance, by getting ‘kickbacks’ from choosing one investment outfit rather than another. In general, which expert to select for a given task will not depend exclusively on how knowledgeable or skillful the expert is, but also on how trustworthy the expert is in offering helpful advice to the layperson rather than advice that merely advances the expert's own self-interest. Other things being equal, it may be reasonable to avoid the guidance of an expert who clearly has a significant self-interest. Of course, it may sometimes be true that all potential experts stand to benefit from being employed for the same task. In such cases, the interest factor may not play a significant role. But where extremes of self-interest are significant, such information should not be ignored.

5. Mistakes to avoid in Assessing Levels of Expertise.

Psychologists and other scientists of cognition have identified a wide assortment of ‘biases’ or ‘errors’ in human thinking and reasoning. Some of these would apply to decisions concerning whom to trust as an expert.

(A) Overly confident expert.

Listeners are often blown away when they hear speakers assert things with considerable force and self-confidence. Listeners seem to assume that speakers’ self-confidence is a reliable sign of authoritative knowledge. In fact, however, assertiveness and swagger are not very reliable indicators of truth possession (or expertise). The psychologist Philip Tetlock, for example, reported an experimental finding that experts who are most often accurate tend to be cautious, quiet, and somewhat more boring than their more self-confident and emphatic counterparts (Tetlock, Reference Tetlock2006). Thus, if you want to identify a genuine expert, you may well be better off choosing a cautious character rather than a super-confident one.

(B) Confirmation bias.

Another belief-forming bias has been reported by the psychologists Lord, Ross, and Lepper (Reference Lord, Ross and Lepper1979). They write: ‘People tend to take at face value any evidence that confirms their prior beliefs, while subjecting evidence that disconfirms prior beliefs to intense critical evaluation’. Similarly, Ditto and Lopez (Reference Ditto and Lopez1992) report: ‘Less information is required and less cognitive processing is devoted to reach conclusions we antecedently favor as opposed to conclusions we antecedently disfavor’.

(C) Conforming with the group.

Another common error people tend to make is trusting others when they have no reason to do so. Psychologist Solomon Asch (Reference Asch and Guetzkow1951) devised an experiment in which groups of eight participants were shown a card with one line on the left and three lines of different lengths on the right. Their task was to identify which line on the right was of the same length as the line on the left. Meanwhile, seven members of the group were confederates of the researcher, instructed to choose the same wrong line. The subjects, answering last, could then either agree with the rest of the group – thereby picking the wrong line – or pick the right line. A significant number of subjects chose to go against the evidence of their own senses and instead conform with what the other group members said. This highlights the pitfalls of following, or conforming with, ‘the crowd’ (O'Connor and Weatherall, Reference O'Connor and Weatherall2019, p. 81). As this example suggests, ‘blind’ conformity with others is a dubious basis for choosing an expert.

6. Choosing Between Bodies of Experts and a Skeptical Citizenry.

In the preceding discussion the focus has centered on the situation of a single person with a problem who needs the guidance of an expert but finds it difficult to decide which (putative) expert is the best or most reliable. In this section we consider a different kind of problem featuring experts. It is a case where there is a firm body of experts who agree that a certain type of action is desirable. However, there is also a body of citizens who question or deny that this type of action is desirable. They do not necessarily claim a scientific basis for their opinion, but nonetheless deny that the relevant type of action is desirable. What course of action should be adopted by the largely untutored citizenry?

It might seem implausible to be concerned with this type of problem, except for the fact that there are actual bodies of contrasting opinions that exemplify such disputes, even in the most educationally advanced countries, such as the United States and England. Which group is being unreasonable, we might ask, and how or why have they gone wrong? What is the source (or sources) of the error? The discussion that follows is based heavily on the reporting of Jan Hoffman (Reference Hoffman2019.)

In August 2019 the United Kingdom lost its World Health Organization designation as a country that has eliminated measles, because of outbreaks that year. The same thing was expected to occur in 2020 for the United States. In both cases, this happened because of ‘vaccine hesitancy’, that is, the practice of declining to follow standard medical vaccination guidelines. The World Health Organization lists vaccine hesitancy as one of the top threats to global health. This is a reversal of the public's attitude toward vaccination, which had previously followed the dominant position of medical science. How or why was there such a change on the part of the public? Assuming that the change was a ‘mistake’ on their part, what were the sources of the mistake?

Hoffman (Reference Hoffman2019) offers several explanations. One major contributor to the emergence of anti-vaccine sentiment was the infamous publication by Andrew Wakefield, a British gastroenterologist, of an article in Lancet that associated the M.M.R. vaccine with autism (Wakefield, 1998). That study was later discredited and withdrawn by the publisher. The rest of the present discussion, however, focuses on events in the U.S.A.

According to Jan Hoffman (citing Paul A. Offit), scientists now report that science has ‘lost its platform’: ‘Now, you simply declare your own truth’. She traces the constituents who make up the so-called ‘vaccine resistant’ come from disparate groups, including anti-government libertarians, apostles of the all-natural, and those who believe that doctors should not dictate medical decisions about children.

In a very different vein, Hoffman reports the sense among health experts that parents and even many doctors do not appreciate the severity of disease that immunizations have thwarted because they have not witnessed many actual cases of the disease. Thus, ‘vaccines are a victim of their own success’. Another strong strain in Hoffman's explanation of the rise of vaccine resistance is ‘groupthink parenting’, facilitated by social media.

A different kind of explanation offered is failure to understand numerical risk. ‘We pay more attention to numerators, such as “16 adverse events”, than we do to denominators, such as “per million vaccine doses”’.

These points might reasonably be classified as ‘mistakes’ that (certain) parents are making. But there are other factors that certainly should not be counted as mistakes. One such factor is the lack of trust in ‘Big Pharma’. Nobody can deny the humongous guilt of Big Pharma, especially with regard to Purdue's highly culpable role in the opioid catastrophe. Obviously, not every institution engaged in some sort of technical and profitable activity deserves uncritical obeisance. That is what makes our topic of expertise and responsiveness to expertise such a timely and significant topic for discussion and analysis in our time.

7. Selecting Educational Curricula and Governmental Policies

Most of the foregoing discussion focuses on a relatively narrow problem about expertise: namely, how laypersons can tell which of two self-styled experts is the better one to choose for guidance on the problem at hand? As already indicated, however, it would be a mistake to leave the impression that this is the only interesting and important question associated with expertise. Other issues have been raised by a variety of writers, primarily issues concerned with the role of expertise in public settings and venues. We shall illustrate these discussions in connection with two additional domains: (1) how should educational systems proceed in crafting curricula for public schools? and (2) what role should expertise play in guiding governmental policy- making?