One of my friends, Scott, is a toxicologist. He has a Ph.D. from a well-respected university. His job is to test water, soil, and air samples after fires or leaks at chemical plants. As far as I know, Scott is pretty good at his job. If he were to tell me, because of a recent chemical leak, the water in my neighbourhood has unacceptably high amounts of a harmful chemical and that I should filter my water before drinking it, I would take his advice. I would take it because he is an expert – he knows about such things. Further, I am not an expert. Not only do I not know about such things, I don’t have the time or training to check for myself (even if I cared enough to do so).

On the face of it, this story makes sense, and few would discourage me from taking Scott’s advice. But the world is more complicated than the story suggests. What if he weren’t my friend? What if, instead, I was sitting next to him at a bar and, after he convinced me of his credentials, he advised me to filter my water? Or what if I just saw him on television and he was telling the whole city to filter their water? It seems my relationship to him matters quite a bit for how much I trust him.

Further, what if I learned that this stranger’s salary was paid by my state’s Department of Health instead of the company that supposedly leaked the chemical, or vice versa? And would it make a difference if one political party had a stake in the company’s reputation or the town’s safety, so that they either publicly supported the toxicologist or rejected his advice? These questions suggest that the toxicologist’s social and political context matter to how much I trust him.

Maybe this isn’t true when we need experts for help with minor issues. If I need a mechanic to fix my car, an attorney to handle a legal matter, or a doctor to prescribe an antibiotic, I may not have many worries about trusting them. But when the stakes are high, figuring out who is an expert, whether to trust them, and how much to trust them are not simple matters. Whether it’s deciding whether to have a major surgery, what to do about climate change, or how to survive a pandemic – experts are a lot of trouble. Yet, as long as we’re stuck needing them (and we always will be), we have some work to do. So, let’s start at the beginning: Why would anyone trust an expert in the first place?

Should we trust experts because they know more?

Many people say, “We should trust experts because they know more than everyone else.” But is that right? It actually depends on what we mean by “know” and “everyone else.” If by “know,” we mean “can state true, well-supported propositions about,” then this is too narrow. This is a kind of knowledge that philosophers call “knowledge-that,” such as when you know that the earth is not flat, and you know that two plus two is four. Knowledge-that works for some kinds of expertise, such as scientific expertise and art history expertise, but not so much for others. For example, some people can show you how to do something so you can do it yourself, such as how to build a table or wire a house for electricity. And they might be able to teach you how to do it even if they don’t know a lot of true that-statements about woodworking or electrochemistry.

So, our conception of “know” should be expanded from knowledge-that to include “knowledge-how.” But even know-how doesn’t capture all the relevant aspects of performance. Some people with know-how can’t explain at all how they do what they do. As you’ve no doubt noticed with professional athletes, they can do a whole lot, but they have trouble talking about what they did (especially when interviewed after an event). The same is often true of musicians and artists. They have “tacit knowledge” of how they do what they do but cannot explain it to anyone else.

And finally, in many cases, every conception of “know” is too strong. Experts make a lot of guesses. They make judgment calls. They take risks. If you listen to a doctor rounding on a patient in an Intensive Care Unit (ICU), they may place a lot of orders for medications and treatments. But if you ask, “Do you know that these medications and treatments are right?” they often say, “We’ll see how the patient responds. If I’m right, we should see the patient improve. If I’m not, we will try something else.” The same goes for fields like teaching and psychotherapy. Teachers and therapists make predictions about how students and clients will respond to their skills, but it’s too strong to say they know what works and what doesn’t in any given case. This is not to say that doctors, teachers, and therapists aren’t experts. Quite the opposite. The point is that “knowing” isn’t always the best term for expertise.

Let’s call this broad sense of ability to do something in one’s field “competence,” and this includes propositional knowledge, explicit know-how, tacit know-how, and expert-level guesses. Someone who is an expert is competent in their field.

Should we trust experts because they are more competent than everyone else?

Not necessarily. Imagine the earth were hit by an asteroid leaving behind only the most competent toxicologists. In that scenario, my friend Scott wouldn’t be more competent than anyone else, yet he would still be an expert toxicologist. That’s because everyone who exists is as competent as they can be given the current state of toxicology science. So, whether someone is an expert depends more on the state of the field than how many people someone is more competent than.

Now, you might think: Okay, but being more competent than everyone else is at least sufficient for expertise, right? But that’s not true, either. A school teacher who knows basic first aid may be more competent at it than any of the students or other teachers, but they aren’t a medical expert. What if they were more competent at first aid than anyone in their neighbourhood, city, or state? Would that make them a medical expert? Not obviously. This is because we know just how far the field of medicine has come from basic first aid.

To be sure, they might be an expert in first aid (if it could be considered a stand-alone field). But even if first aid were its own field of expertise, it would probably be what’s called a “low performance ceiling” field. In other words, it would be a field that doesn’t require much training and in which it is pretty easy to become competent. Even then, being better than most everyone else is not sufficient for expertise. Consider this: The “history person” on your local bar trivia team may be more competent than anyone in the tri-county area, but they are still probably not an expert on history by the standards of the field of history. And someone who can fix a copy machine better than anyone in the office is competent but probably not an expert—at least not as expert as the copy machine repair person.

At this point, we have a working definition of “expert” and why we should – in general – trust them: An expert is competent according to the current state of practice in their field. And if they’re comparable to the best in the field, that’s a good reason to trust them. So far, so good. Except…

Shouldn’t I trust the history person on my bar trivia team? Even if they aren’t an expert historian, they surely know more than I do. Why does it matter if someone is competent according to the standards of some field when I just need someone better than I am? In other words, why should I care about expertise when having “more competence than I have” is good enough?

Should we trust experts because they lower our risk of failure?

Now we’re getting somewhere. For every field, people fall along a continuum of competence. Your middle school science teacher knows a lot more than the average person about basic science. Students and average folks should – all things being equal – trust them about basic science. But this knowledge is very general, covering all sorts of fields, like chemistry, physics, biology, etc. It is not specialised knowledge in any one field. A professional chemist is more competent than your middle school science teacher. They not only know the basics, they also know the extra stuff they specialised in. And they know more of the latest findings in chemistry – their understanding is up-to-date. Further still, they practice chemistry every day, so they not only have first-hand experience with much of what they know, but they understand it better than less competent people because they practice. Your middle school science teacher does not have this kind of competence.

So, imagine you have a basic chemistry question. Whom should you trust: your middle school chemistry teacher or the chemist? It might be that either would get it right. What’s interesting is that the chemist is more likely to have a richer understanding of why the answer is right. Your science teacher might have a fuzzy memory of learning it from a class or a textbook. But the chemist has current, working competence in chemistry. That suggests that, even if you should trust your middle school science teacher when there is no chemist around, the chemist is more likely to be right for good reasons. This means you have a lower risk of failing to get what you need when you go to an expert over someone who is just minimally competent.

This becomes clearer as the stakes go up. Imagine the question is not about chemistry in general, but about a specific medicine, and whether that medicine will interact poorly with another that you’re taking. For this question, your middle school science teacher isn’t even in the running. The chemist may be helpful – they could probably look up the chemical makeup of the drugs and understand a few scholarly articles on it – but they might not know enough about human bodies and illnesses to help you safely. What you really want in that case is a specialist, namely, a pharmacologist. The chemist and the pharmacologist might both give you the right answer, but the pharmacologist is more likely to be right for the right reasons. Trusting the expert lowers your risk of getting poor information. This is why expertise matters: As the stakes go up, getting the right answer for the right reasons is more important than just getting the right answers because it lowers your risk of failure when you really need something you can’t do for yourself.

This is a good reason to trust your family doctor more than medical websites. While medical websites may say true things, those things may not apply in your case, given your medical history or the medicines you’re taking, or what risks you’re willing to take with side effects or other complications. Whether a treatment will work for you, given your particular values and risk factors, can only be determined by someone who knows how to take general knowledge and understand how it will affect your – very specific – medical needs.

So, why is expertise controversial?

If all this is right, then why isn’t trusting experts always the obvious choice?

It turns out that – surprise – experts are also humans. And humans don’t always use their competencies as well as they could. An expert carpenter might still do slipshod work. Expert researchers might fudge data to get a big grant. Expert physicians might try to speak authoritatively outside their field of expertise (politics, climate science, etc.).

The real trouble with experts is not with expertise itself, but how experts behave. Experts don’t always use their expertise in ways that are helpful to non-experts.

· Sometimes they have a conflict of interest or a political motivation.

· Experts sometimes speak outside their field – a phenomenon called “epistemic trespassing.”

· They may try to tell you things that are not “tell-able,” as when a very successful person tries to give step-by-step advice on how to navigate all the obstacles they handled intuitively.

· Experts may offer advice that doesn’t meet your needs.

In addition to these concerns, some issues get tangled up with political or religious perspectives that make it difficult to know when an expert is really acting like an expert (and trying to help you) or when experts are speaking from their personal political allegiances. This is especially concerning when the issue is time-sensitive, as in the case of pandemic diseases like COVID-19.

Further still, sometimes whole fields of expertise give the wrong answer in a terrible way. Consider the field of medicine. Medical researchers have exploited people of colour, obstetricians have ignored medical decisions from women in labour, pharmaceutical corporations have conspired to increase addiction, and trans patients are routinely stigmatised or refused care. There are lots of reasons to be sceptical about experts. But it’s important to note that those reasons have nothing to do with expertise. The trouble comes because of the power experts have to put people in compromising positions and to use their positions in ways that harm others.

What is the future for experts?

It should be clear that we can’t live without trusting experts. The world is far too complicated. You’re relying on thousands of experts right now just to be able to access the words you’re reading online. But the more we learn about the human side of expertise, the more reasons we have to be cautious.

A step toward improving expert reliability is to add checks on expert judgment. Rather than having a single physician make medical decisions, these days medicine is often a team-based affair. Nurses, social workers, ethicists, and others are involved to make sure critical aspects of the patient’s life, social situation, preferences, and values aren’t overlooked. Similarly, for government-funded research, Institutional Review Boards (IRBs) evaluate studies for ethical concerns about how participants are informed and treated.

These aren’t fool proof measures, to be sure. But consider all the ways you can use the basic model of checks-and-balances to review and weigh expert judgments. Try to get insights from multiple, independent experts from a variety of closely related fields. See if they explain why they arrive at their conclusions. Test those conclusions against advice given in past cases that were similar. Choosing and using experts wisely is not easy. But with the vast amount of misinformation and disinformation floating about, we can’t escape the responsibility of doing the hard work of deciding whom to trust and how much.

Jamie Watson is an Associate Clinical Ethicist in the Center for Bioethics at the Cleveland Clinic and author of Expertise: A Philosophical Introduction (Bloomsbury 2020) and A History and Philosophy of Expertise (Bloomsbury 2021). His interests also include the epistemology of the doctor-patient relationship and ethical issues related to health literacy.