Respected Experts Of Any Sort

From RespectedSoftwareExperts:

Respected Medical Experts can't even agree on whether meat and fat is good or bad for the body. Respected Physics Experts can't even agree on the angular momentum of the planets close enough to decide whether Velikovsky's theories work or not. Respected Architectural and Economics Experts disagree with each other as a matter of fad. Respected Experts in every field have been more wrong than right for as long as they have existed. I have no idea what a resolution to the "crisis of authority" could look like, but it scares me just thinking about it. -- AlistairCockburn
Um, Alistair, you might want to reconsider your comments about Velikovsky at RespectedSoftwareExperts. Unless you have a very strange definition of "respected physics experts". -- BenTilly?

I didn't write that Velikovsky is a respected physicist. It is when I try to read attack and defense of his ideas by allegedly respected physicists, it is among *them* that I notice immense disagreements. -- Alistair e p Credentials are easy to claim. Astronomers have very good estimates of the angular momentum of the Solar System, and can project past orbits with amazing accuracy, an extreme example being (needless to say, they don't find support for Velikovsky's theories) -- Ben

...ah, but you see, I can't distinguish one astronomer from another, and when two of them differ in print about the angular momentum of the Earth to 10**3 or 10**6, there's no way I can know which one's wrong - I can't do the calculation myself, and so I can't break either astronomer's calculation - - - that's the point I was making. The fact that you tell me I should trust astronomer A and not astronomer B doesn't help. -- Alistair

[Why can't you do the calculation (for the earth) yourself? Or, for the solar system, at least estimate for Jupiter, which should have the lion's share.]

Jupiter does have the lion's share, followed by Saturn. Those 2 have over 90% of the angular momentum of the observed solar system. The angular momentum of the Oort cloud is unknown and may be very large indeed. (Very long moment arm.) -- BenTilly

[The problem with this line of reasoning is that it implies that everyone must become able to do everything as well as the experts. The fact that any particular individual thinks he/she can compute the angular momentum of the earth is a RedHerring - we could have used mass of an electron, or attraction of two strings, or how to run hot water pipes from geothermal springs. How can I know if my answer is right if two experts disagree by huge amounts? If I am confident enough to know that my answer is right, then I am an expert and the question doesn't apply - just take the next subject and restart.]

Then your definition of "respected physics expert" was "writes well enough to look convincing to someone who knows nothing on the topic". Which is more about verbal fluency than expertise. In that case I try to either avoid venturing an opinion, or else qualify my ignorance up front. I say _try_ because I often find I have failed... (Relevant book: "WhyPeopleBelieveWeirdThings?" by Shermer.) -- Ben

Actually, I'd change that to "writes well enough to convince the people around him/her" and suggest that that's all there ever is. If I or anyone else opens up a journal and sees articles by two people with PhD in Physics, we trust the testing system that produced those PhDs?. That testing system consists of nothing but one person convincing the people around him.her, who are supposed to themselves be respected experts. The point I was making is that in no field at all - not even physics and certainly not even medicine - does the word "respected expert" connote "correct" -- Alistair

I didn't disagree with the basic point. (In fact you are being too kind to medicine IMHO. You may find interesting.) I disagreed with the example. But now you have opened up some interesting cans of worms. Here are several directions I see this conversation going.:

1. What makes an expert, expert?
2. What are the limitations on how much we should trust experts?
3. How can non-experts tell true experts apart from pseudo-experts?
4. What implications does this have for our profession?

I am not an expert in experts (:-), but here are my beliefs:

1. What makes an expert, expert? I don't disagree that convincing people is key. But experts need to convince people who know something about the subject at hand. People who, to paraphrase you, "Can do the calculations." In most subjects this takes a lot more than verbal fluency.
2. What are the limitations on how much we should trust experts? Well first of all we need to trust experts somewhat because they are more likely to be right than non-experts. However how much comes down to a matter of epistemology, "What do we know, and how do we know that we know it?" Experts in simple fields which readily examined in detail usually have concrete answers that we can trust to great precision. Experts in fields which are harder for us to have solid answers about generally have fuzzier answers. And even if experts are wrong regularly, most subjects develop knowledge cores that can be taken as "established fact", and have other areas that are less certain.
3. How can non-experts tell true experts apart from pseudo-experts? Generally they can't. Most of us check expertise by embedding what we know of the person in a value system. Do they have credentials? Do they sound fluent? Do they use buzzwords that ring a bell? Does anyone that I have heard of recommend them? If we know something about the topic, this is generally effective. If we don't...?
4. What implications does this have for our profession? A lot. To most people, the company with the best marketing campaign look more like experts than the real experts do. (That is ThePointOfMarketing.) This strongly assists the latest vendor selling snake-oil that obviously won't actually get the job done.

-- BenTilly

I ran off to find the definition of expert in the dictionary and all of them (that I checked) agree that an expert is someone who has acquired skills and/or knowledge. Most of them add the nuance that the skills/knowledge were acquired through experience. When we say that we trust somebody's expertise I find that more often than not what we are actually trusting is that person's experience. Trusting in somebody's skills is just not usually the way we do things (note that very few "respected experts" become so before they are 30).

The testing system Alistair refers to proves that a given person has the skills but says nothing about their experience. Indeed, an "inexperienced" university doctoral student could pass those same tests and could (hypothetically) do better than a 10-year veteran.

All of this leads me to state the following: no two experts can ever completely agree. Because they have different experiences they must have a different kind of expertise. Each one may be qualitatively and quantitatively equal to the other and yet they may be different. Each one's past has taught different lessons; how then can you ask them to agree on any but the simplest and most basic things? The less a discipline depends on measured experiences (medicine as compared to fine arts) the less experts will agree.

I feel the crisis of authority and was talking about it recently with a colleague. Unfortunately I am not yet a respected expert so I have no answers. Wait a couple of years and I may publish a book (HaHaOnlySerious).
Of course experts disagree on some issues; that's part of how scientific fields evolve. However, one expects that experts will agree on most of the issues which are accessible to a layman.
It all boils down to this: "N�o basta ser, � preciso parecer". --MariusAmadoAlves

View edit of January 24, 2008 or FindPage with title or text search