Episode post here. Transcription by Prexie Magallanes.
Matt Teichman:
Hello and welcome to Elucidations, an
unexpected philosophy podcast. I’m Matt Teichman.
Joseph Diller:
And I’m Joseph Diller.
Matt Teichman:
With us today is Oliver Traldi, postdoc at the
James Madison Program at Princeton
University, and visiting assistant professor at the University of
Tulsa Honors College. He is
here to discuss political expertise. Oliver Traldi, welcome.
Oliver Traldi
Thank you for having me.
Matt Teichman:
The first thing I always like to talk about,
when I’m talking about experts with people, is the difference between
actually being an expert—i.e. actually understanding stuff well, and
knowing things—versus being talked about as an expert, or
claiming to be an expert. At least when I talk about being an
expert, I’m talking about actually knowing stuff, as opposed to
claiming to be an expert. Is that also how you tend to use the term?
Oliver Traldi
I tend to use the term in both ways, which is
probably a little confusing. When I talk about political expertise,
I’m drawing on both a public conversation about experts and an
academic literature in epistemology about the nature of expertise. In
epistemology, when we talk about experts, we’re generally talking
about people with a lot of knowledge in some domain, people with a
good track record, people with true beliefs, or people who are
rational, or have some sort of understanding.
There are all these epistemic states that we consider valuable, and somebody who has them with a domain is what an epistemologist normally means by an expert. Then, in public conversation, when we talk about experts, we’re usually talking about a sort of expert class—maybe somebody with a certain kind of credential, or somebody who goes on some sort of talk show, and below their face it says ‘expert in this’ or ‘expert in that,’ or somebody puts ‘expert in something’ in their Twitter bio.
Matt Teichman:
They gave an exciting talk, and their startup
got seed funding.
Oliver Traldi
Yeah, exactly: somebody who did a TED talk, or
something like that. One of the questions is how these two categories
relate to one another, and as a matter of social engineering, how we
can make sure that there’s more overlap. Or, as a matter of
individual rationality, or individual decision making, how we can
decide which of the latter we can trust—which of the credentialed
experts we’re rational in treating as experts, in the epistemological
sense.
Matt Teichman:
But we’re allowing, then, that somebody could
actually be an expert, but not have any credentials. Maybe they’re a
hermit who has practiced something, and they’re really good at it.
Oliver Traldi
Yeah, absolutely. The notion of expertise
doesn’t have built into it any specific source of one’s beliefs. So
a mathematical savant could be an expert in mathematics.
Matt Teichman:
Right, yeah, Matt Damon, or whatever.
Oliver Traldi
I was thinking of Ramanujan.
Matt Teichman:
Yeah, the actual guy who—
Oliver Traldi
—yeah, who’s discussed in Good Will
Hunting. But yeah, he’s an example. The kind who sets the piece of
paper on fire after proving something that you could never do your
whole life.
Matt Teichman:
That part was an embellishment, but yeah.
Oliver Traldi
A technical term for that is a really mean
expert, in epistemology.
I think we would allow in epistemology that that’s an expert. But in a social discussion, or a discussion of the cultural nature of expertise, we would probably not call that an expert. They don’t have any credentials; they don’t have a platform. So when we think about expertise in politics, we’re trying to participate in both conversations at the same time.
Matt Teichman:
What would be the world’s simplest, most
straightforward example of somebody who’s an expert in something?
Oliver Traldi
I think the best examples are the—and it’s
also part of the case for expertise, you know. There are experts we
trust every day. Tomorrow, I’ll fly back to the East Coast. Obviously,
I’ve never flown a plane, and the pilot will probably be an expert
pilot, in the sense that they’ve trained to do it. They haven’t
crashed any planes yet.
Matt Teichman:
They didn’t just let me fly the plane.
Oliver Traldi
Yeah, exactly. Another one that I like is
dentists. You know, you go there and you say, stick these ridiculous
contraptions in my mouth, and put drugs in the insides of my cheeks,
and things like that, and here’s a thousand dollars. So you must have
some sort of trust that this person knows what they’re doing.
Matt Teichman:
Maybe even less high-stakes things, like
getting a haircut. You trust the person is not going to make your hair
look really stupid.
Oliver Traldi
Yeah, and it’s interesting where we get our
information about these things. For a haircut, in particular, I’m
always reading Google reviews. I want to make sure that if there’s
even one person who said, “Well, I looked like a real idiot coming out
of this place,” I’ll probably look for a place where they didn’t do
that to anybody. Although if it’s 1 out of 500, maybe I think I’ll
take the risk.
More broadly, the problem of political knowledge is often a problem of trust. The broader question is who to trust. Somebody having expertise, that’s one of the ways to gain your trust. There are these different ways that we can establish expertise, if that makes sense.
Joseph Diller:
How do we judge a young and up-and-coming
expert to be an expert, when they don’t have the track record of
success?
Oliver Traldi
This is a good question. In a way, these
questions are almost as difficult as general questions about e.g. how
do we know anything? There’s a famous paper by Alvin Goldman that
gives some criteria other than track records: if you’re trying to
figure out which expert to trust, you should let them have some sort
of debate against one another. Which is a little difficult to do,
because, not being an expert, you might not know which sorts of
considerations should win out in the debate.
Matt Teichman:
I know this would never happen, but
hypothetically, you might imagine two experts talking past each
other?
Oliver Traldi
Yeah, exactly. A non-expert might even not be
in a good position to even understand when people disagree—when
their dispute is substantive versus verbal. Sometimes, even experts
themselves are not quite clear on whether they disagree with each
other. Certainly, I’ve seen cases of false consensus, where there
seems to be some sort of coalition in the academic literature, or in
the political world, but then you poke at it, and you realize actually
there’s like 50 different claims being made here by the different
people.
But also, I was at a talk last week by Cory Clark, who’s (I think) a social psychologist who (I think) works at Heterodox Academy now, and was at maybe Penn at some point. She works on adversarial collaborations in academia, and she said that the point of adversarial collaborations is: you get people who disagree with each other to work together. Try to, you know, see who’s convinced by the end, or something like that.
Matt Teichman:
Does that mean that they co-author a paper?
Oliver Traldi
Yeah, maybe they co-author a paper, or they do
a back and forth.
Matt Teichman:
Oh, like a debate in a journal.
Oliver Traldi
Yeah, yeah. You’re probably overestimating just
how locked in I was during this talk.
Some of the details, you know, a little of it was happening in my imagination. But one thing she mentioned was that sometimes these people who think they disagree with each other so much, you know, it’ll be just some tiny question about effect size in a study, or some very minor thesis. I think it can be very difficult even for the experts themselves to understand what’s at stake in their debates and their disagreements. That’s one of the many challenges of expertise.
I don’t know if that answered your question. Your question was about somebody who’s up and coming. They don’t have a track record, maybe they haven’t made predictions yet, or they made some predictions, but the time hasn’t hit them yet.
Joseph Diller:
Do you think that maybe we can look at other
track records that might be a useful reference, like good references
from advisors, or the prestige of the institution they went to, or the
fact that they passed some difficult exam, like the bar, or—
Oliver Traldi
—yeah, certainly that’s part of our
credentialing mechanisms. And you know, it’s probably rational to
trust these sorts of things, in lieu of other sorts of information. I
do think that a lot of what’s sometimes called
meta-expertise—degrees and awards, credentials,
recommendations—there can be some dangers of these things. If you
already have it in mind that you trust this first set of experts, the
older group who says, “Well, here’s this great student I had. Here’s
my letter of recommendation.” Well, if you trust that person, then to
a lesser extent, you’ll trust this new person, because you trust their
recommendation.
The hard case is: what induces you to trust the first set of experts to begin with? The hard case is when, maybe, if you have some more thoroughgoing skepticism about the system—you know, take me. I’m always complaining about the people who get MacArthur genius grants. I have this horrible view that they might not actually be geniuses.
Matt Teichman:
The very idea!
Oliver Traldi
I know, it’s crazy. This is one of my most
offensive views, actually. In that sort of circumstance, you might
say, well, I don’t know: I don’t trust the MacArthurs, I don’t trust
Harvard, I don’t trust the New York Times. Then you’re in a much more
difficult place.
Matt Teichman:
You might also just worry that the skill of
identifying talent in your domain is not necessarily the same skill as
yourself having talent in that domain. You might think that
identifying other people who are good in the same way that you are is
different from being good in the way that you are.
Oliver Traldi
I certainly think that’s right. Just thinking
in terms of academia, one thing I do sometimes notice is that you can
have somebody who’s a star by doing something really creative, and
then their star students are often people who are actually very
derivative, and they become stars with the professor by doing what
they want. So you can get into these odd sorts of situations where
you’re actually not looking for the sorts of traits that you yourself
have.
Matt Teichman:
When I was a PhD student in philosophy, 12
million years ago, all I ever heard from admissions committees was
that they were reading recommendation letters, and every single
candidate that came in was “the next David Lewis.” And I’m like, are
they really the next David Lewis? Or are they a carbon copy of a
carbon copy of a carbon copy of who knows how many generations off of
David Lewis, at best?
Oliver Traldi
Well, yeah: I think there’s a lot of reasons
not to think that there’s dozens of the next David Lewis every year.
Among other things, once you have a few next David Lewises, it’s sort
of like now there’s no next David Lewis, right? Part of it has to do
with being influential in a way that only a few people can. But yeah,
I think a lot of it also has to do with what you might think of as
normal science versus paradigm shifts. Somebody who has a genuinely
new idea, versus somebody who’s going with the flow of the way a
discipline is already set up. And in terms of experts and trust, I
think it’s probably much easier to get information that somebody is
doing a good job by the standards of their workplace, rather than the
question of: are those the right standards, in some deeper sense?
Matt Teichman:
So we’re raising all these worries for
political expertise, and now I’m starting to wonder: is there such a
thing as political expertise? One of the things we mentioned was:
maybe experts are the people who have good track records in their
area. I don’t know—the lead engineer at the Toyota factory, or
whatever—probably they have expertise in building cars that don’t
fall apart when you drive them for 10 feet. So I can trust that a car
that came off their assembly line is going to have that property. But
is there really any such thing as a track record in political
advisors, politicians, political philosophers? It just seems like the
‘building a car’ scenario involves controlled conditions that are
repeatable, whereas history never repeats itself exactly, and there
are so many different intervening factors. So I wonder: should this
push us to skepticism about the very idea that somebody can be a
political expert?
Oliver Traldi
One of my views is that a lot of different
types of questions can be political questions. I think of the domain
of the political as somewhat ill-defined, but mostly having to do with
public controversy, in some sense. So, given that we can have a public
controversy over absolutely anything, you could have experts on issues
that have become political—issues that have become
politicized—simply because you can have experts on anything. You can
have experts on climate science, which can become political. You can
have experts on geopolitics.
One of the difficulties in answering political questions is that most concrete political questions—the ones that might be, say, surveyed on a public opinion poll—are often questions about what we ought to do. They have a moral dimension, or at least a pragmatic, practical dimension. And so, one difficulty with political expertise just has to do with moral expertise. Moral epistemology, in general, is a difficult part of—well, whether it’s epistemology or ethics, it doesn’t really matter how we draw the lines—but it’s a difficult part of philosophy. The question of how do we know right from wrong seems different than knowing whether the sky is blue.
Matt Teichman:
Whether this car is going to fall apart, etc.
Oliver Traldi
Exactly, yeah. So in those cases where there’s
a clear moral element—well, we were talking before about the
difference between the knowledgeable people and the credentialed
people. One thing you see sometimes is credentials that have to do
with e.g. building cars, or flying planes, or being a dentist or, you
know, climate science, or epidemiology or whatever it is, some
credential being leveraged into the moral realm, somehow. You know, I
have this credential, and therefore, I know right from wrong. A
pretty slippery thing to do: I think Socrates would always yell at
people who did stuff like that.
Matt Teichman:
Yeah, there seems to be some intuition that in
that setting, the expert isn’t staying in their lane. They’re veering
into topics which they’ve maybe forgotten, or which they’ve lost sight
of the fact that they’re not actually experts in.
Oliver Traldi
I think that’s absolutely right. One of the
questions that I didn’t feel that I did at all a good job of answering
in my discussion of expertise was: what are the lanes? What does it
mean to stay in your lane? An expert is somebody who knows a lot, or
has some good standing on some domain of questions, or
propositions. So there’s this question of: what are those domains, and
how do you actually divide one domain from another? And even: how do
you know your own domain? How do you know what your own lane is?
Take us; we’re trained in philosophy. There are some philosophers who
argue against the notion of philosophical expertise.
Matt Teichman:
Maybe even some in Chicago. Who could say?
Oliver Traldi
Yeah, I think there are certainly some in
Chicago. And the debates about philosophical expertise are part of the
debates about experimental philosophy, which used to be a
controversial topic—I think it has now been accepted as something
people do, sometimes.
Matt Teichman:
This podcast is so old that it was a
controversial topic, when it started.
Oliver Traldi
Well, no, that’s good. That’s when I was in my
masters program at Tufts. There was a lot of debate about experimental
philosophy—somewhat less at Notre Dame, as you can imagine. But
yeah—the question of where we draw these lines, and what we expect
somebody to be able to do, or be able to know, based on doing one
thing. Take the dentist. Does the dentist know about other health
problems? Does the dentist also function as a kind of doctor? It
might depend on what training they’ve had.
Matt Teichman:
Should they give you nutrition advice?
Oliver Traldi
Exactly. Maybe they tell you to eat something
because it’s good for your teeth, but it’s actually bad for your
digestion, or something. And then, if you push that too far, it turns
out the only experts are people who know everything. It’s a pretty
high bar. But one of the questions about political expertise is,
maybe, that metaphor of the body that I accidentally drew there, in
which you can have unintended consequences, and a lot of things relate
to each other in complicated ways.
Maybe society is also like that. Society is very complex. It’s hard to make predictions about what effect something will have, a long time away, in society. So I think some of the doubts about political expertise have to do with the question of whether you can have a full enough view of society to make long-term predictions—to make predictions across a whole series of different aspects of society, if that makes sense.
Joseph Diller:
Do you think we follow political experts
simply because A, we agree with them, and B, they likely said it
better than us? That is, they are smarter, and have better reasoning
skills, vocabulary, and tacit knowledge?
Oliver Traldi
I was thinking about this on the train on the
way over here, on the Red Line. A compliment I sometimes get is: you
know, I’ve had thoughts vaguely like this, but I would never have
thought to put them together in this way. It’s a nice compliment, but
it’s more of a compliment about one’s writing ability than one’s
knowledge, per se. It sort of suggests that anybody could have these
thoughts, but you composed them well.
Probably not a surprise. You guys know me from Twitter, and there are some vague references in my book to relationships these ideas have to online debates. One thing you see in online debates is that people come in already having made their decisions about all the issues they’re debating about. They look for an expert. They go expert shopping, in the phrase of Gabriele Contessa, who maybe is at Northwestern? Or that could be completely wrong. Gabriele Contessa has this idea of expert shopping, where if you have experts who disagree with one another, basically you can always find the expert who supports your ideas. And if you do that, you’re not really deferring to the expert. You’re starting with your own view, then equipping yourself with them.
Matt Teichman:
It seems like it’s just a post hoc
rationalization of the things you already believe. You’re not looking
to the experts to learn something; you’re just looking to justify
antecedently held beliefs.
Oliver Traldi
I think that’s exactly right. And certainly,
there’s a worry about political thinking, in general, that most of it
is post hoc rationalization. You see a lot of different forms of
this. Jonathan Haidt’s theory of moral foundations, social
intuitionism applied to the political realm, is an aspect of
this. He’s a Humean, of sorts. He doesn’t say reason is the slave of
the passions, but he says the reason is the public relations firm of
the intuitions, or the spin doctor, or the press secretary, of the
intuitions. So depending on whether you think intuitions are rational
themselves, you might think that that’s post hoc rationalization.
Certainly, Hugo Mercier and Dan Sperber have a theory of rationality
on which it’s all there to persuade. You know, they call it an
argumentative theory, but it’s argument in the—
Matt Teichman:
—it’s like we evolved it in order to be able
to stand our ground, or something like that?
Oliver Traldi
Exactly. And in order to be able to advocate
for our interests. They call it the argumentative theory, but it’s the
kind of argument you have over the dinner table, not the kind of
argument you put into a philosophy paper. The disputational (or
persuasive) theory of reasoning. This would be of a piece with
searching for experts, and the culture of expertise being employed
post hoc, without our opinions really changing upon encountering the
experts.
Joseph Diller:
When searching for experts, are there
sometimes meta experts novices can follow in judging between experts?
Oliver Traldi
Yeah. I think this goes back to the question
of—
Matt Teichman:
—it feels a little bit like the letter of
recommendation case, doesn’t it?
Oliver Traldi
It’s basically an example of that. There’s a
whole variety of different types of meta experts you have. You might
say, well, this philosopher is a meta expert on other philosophers
because they can recommend philosophers. You might think that there
could be other kinds of meta experts that are, like, well, this sort
of person can tell me: psychology is a trustworthy discipline, but the
whole discipline of philosophy is nonsense. Or somebody might have a
broader view. Things like college rankings are a sort of meta
expertise. Or within philosophy, journal rankings. Like Philosophical
Review is the
most—
Matt Teichman:
—the Nobel committee. The Fields Medal
committee, or whatever.
Oliver Traldi
Yeah, exactly. So, awards, prizes,
rankings. You know, I’m a bit of a critic of some of these things, but
I also use them. I think most people have to admit that we use these
things, when they exist. It’s certainly not irrational to use these
things—it’s just that we can get evidence that lowers our opinion of
them, and we can also see, just like with other humans, meta experts
and institutions like these can have blind spots, or can have patterns
of making mistakes. It might be that the prize for best philosopher
always goes to an ethicist, and you say, well, they’re probably a good
ethicist, but I’m going to keep my own opinion on how maybe the best
philosopher is an epistemologist, or a metaphysician, or something.
You can learn a way to take the judgment of an expert (or a meta expert) as evidence without completely deferring to it, as you learn the patterns of their successes and failures. And I think this is what we all do with each other. It’s the way we ought to be with experts.
Matt Teichman:
This is something I wanted to ask about when
we started talking about experts. What’s the correct response to
hearing the opinion of an expert? Or maybe you’re even talking
directly to the expert. Is it just mindless assent to everything? Is
it: they’re the expert, therefore I must be wrong? Or is some kind of
equilibrium give-and-take the healthier response to expert discourse?
Oliver Traldi
The first question—answering your question
with questions myself—it’s certainly important to distinguish—
Matt Teichman:
—room full of philosophers, who ever heard
of that happening?
Oliver Traldi
Yeah. Certainly the first thing to do is to
distinguish between the case where we’re confronted with the opinion
of somebody who we take as an expert in the epistemological
sense—they know a lot of stuff—and the case where we’re confronted
with an expert in the social sense, where they have a lot of
credentials.
Matt Teichman:
I was thinking more of the first, but we can
do both.
Oliver Traldi
Well, let’s talk about the first. That’s, in a
way, the more intellectually interesting one. The next question to ask
is—we talked about the lanes, the domains—how squarely in the
expert’s domain is this question? Of course, that can be hard to know
without being an expert yourself. One way to answer it is to think
about: do I know whether other experts have the same view? This might
almost go without saying, but if you hear an expert in say, chemistry,
say something about chemistry, and you’ve heard another chemist say
the opposite, then you’re not supposed to go willy-nilly just
believing whichever expert you just heard. Rather, you’re supposed to
balance out the experts. Maybe you should just say, well, this is
obviously some controversy, and I’ll await more evidence, or I’ll come
to my own opinion about it.
Matt Teichman:
Or maybe you failed to understand what they
were saying, and actually, they weren’t contradicting each other. You
go through this whole deliberative process to figure out what was up.
Oliver Traldi
Yeah. In this situation, there’s all sorts of
possibilities when you confront expert disagreement. When you confront
expert consensus, I do think that simply deferring is a live option. I
don’t think you necessarily must continue, so to speak, thinking for
yourself. Again, just thinking concretely about all the different
things people can be experts in, there’s lots of questions on which we
would all defer instantly. You know, if I met up with a group of art
historians, and they told me Rembrandt was born in this place, in this
year, I would be like, okay, that’s probably right.
Matt Teichman:
We wouldn’t say: let’s go fact check this
basic thing they would have learned when they first started their
study.
Oliver Traldi
Exactly. So it’s a place where it’s hard for me
to think: what evidence could I have that they don’t? It’s easy for me
to think: what evidence would they have that I don’t? And it’s hard
for me to see how my own cogitation could help out on the issue. It’s
not like my imagination will help in figuring out when Rembrandt was
born.
Matt Teichman:
At the same time, though, I think a lot of
people have the intuition that this is just a concession to the
finitude of human life. If we were mortal, if we had infinite
resources and time, maybe what we should do is just actually learn
everything ourselves. Maybe we only listen to the experts because
we’re rushed, because this is what makes sense in the moment, and
because it’s not that high-stakes a question, so we make this
concession. But there’s something that feels intuitively a bit icky
about just accepting something because someone said it.
Oliver Traldi
Well, in a way, this is a good case for
undermining that view, at least as a universal view. So if I want to
learn about Rembrandt, there’s not actually a real, first-personal way
to do so. There’s not really a way to avoid the social side of life,
in learning about Rembrandt. I could read a book, but that book was
written by a person. I could ask an art historian in person. That’s
obviously a personal interaction. I could go to Rembrandt’s house, but
I’m going there because somebody told me it’s Rembrandt’s house. So
there’s no way of really escaping the problem of trust.
Trust is my favorite frame for this social epistemology stuff. There’s no way of really escaping the frame of trust, except by saying, well, I can’t see it myself, so I’m going to just withhold judgment on anything that I can’t see with my own eyes. In fact, it’s not clear that what we see with our own eyes couldn’t also be affected or challenged by other people as well. Even our direct sensory information could probably be challenged.
Matt Teichman:
Right. They’re not totally infallible. We do
experience hallucinations.
Oliver Traldi
Exactly. We experience hallucinations. You
know, some substances make us see things that aren’t there.
Matt Teichman:
Memory is fallible.
Oliver Traldi
Exactly. Memory is fallible. We can get really,
really tired. We’re always looking to other people to check our
views. And so, in the history of epistemology, this started to be
emphasized by epistemologists about 25 or 30 years ago, who basically
asserted that, to so much knowledge there’s this social element. Most
of what we know comes from other people. Almost all of what we know
could be undermined by other people. That’s certainly right.
And the thing that I mean to add is: before going to other people, you also need to make the decision yourself about who to trust—about who to actually go to, when you get more information. So there’s a sense in which you’re always on your own, but you’re never on your own. And in a way, that’s part of the difficulty of thinking through how to form the right opinions about some of these issues, or whether there’s even as such a thing as the right opinion.
Joseph Diller:
Can novices ever be seen to be thinking for
themselves, when listening to experts?
Oliver Traldi
Yeah, certainly a novice listening to an
expert, like I said, in a way, can’t escape thinking for
themselves. The novice is thinking for themselves about: how did I
pick this expert? How did I recognize them as an expert? What
information did I use in determining that they were an expert? We
talked about lanes and domains before. How did I decide to split the
world up into chemistry, and physics, and philosophy, and psychology,
and dentistry, and flying? And how did I decide to put this person in
one of those categories, rather than another?
Matt Teichman:
It seems like a paradigm case of this is the
teaching situation, especially in a one-on-one type of class, where
what a lot of experts are hoping to get out of it is some fresh,
left-field questions they never heard before.
Oliver Traldi
I think that’s completely right. As
philosophers, we might have a somewhat different view than people in
other disciplines. Philosophy is maybe unique, or close to unique, in
that in a 101 course, we really start with the most unsolved
questions, rather than the most solved questions. That is what best
communicates what we’re about, as philosophers. And of course, in
doing so, we probably communicate a lot of things we’re certain about,
as well. Maybe we’re certain about what is the right intellectual
method to go about, in investigating some of these questions, or what
is important intellectually, or what is important philosophically. But
philosophy is in a unique place, in that way. 101 classes are so
interesting.
One of the most important events relating to expertise in the past 10 or 15 years is the replication crisis in academic publishing—at least some people think psychology the most affected discipline—and it was the case that there were results that were taught in Psych 101 courses that were undermined; that turned out not to replicate. Replication is when you take an experiment, and you say we’re just going to do it again, see if the same thing happens again. Or the same kind of thing; that’ll confirm that the first paper was correct. You know, if you—
Matt Teichman:
—assuming the background conditions are the
same.
Oliver Traldi
Exactly, yeah. So if you’re Isaac Newton, and
you see an apple fall from a tree, and you say, oh wow, that’s crazy,
things tend to fall back to the earth, but then the next day, you see
an apple break off of a tree, and it’s just sitting there in the sky;
it levitates, so to speak. Then you would say, oh wow, my whole theory
of things falling to earth—that theory didn’t replicate. So if
things keep going different ways, that undermines your theory. And it
turned out that that was the case for a lot of results that have
become doctrine in psychology.
Matt Teichman:
Would it be fair to say this in large part
affected results in what you could call the social sciences?
Oliver Traldi
Absolutely, and I think it’s not completely
clear why that is. It wasn’t exclusive to the social sciences. There
were some medical trials, and things like that. Maybe even some
results in physics? I never quite understood the arguments around
these things, but you know, I’m not an expert. There were some things
where some results had statistical artifacts that weren’t controlled
for, and there also may have been some cases of outright fraud that
weren’t caught in other disciplines. But yeah, the—
Matt Teichman:
—that’s actually why I brought car
manufacturing up earlier, because it turns out that that is one of the
empirical/physical/material phenomena we have the most data on,
because we’re constantly making cars. And even other very well studied
physical phenomena, we’re not necessarily gathering that volume of
data on them.
Oliver Traldi
Yeah. So certainly, the more data the
better. Car manufacturing is an interesting case, because I imagine a
lot of the research into it is done not in the academy, not at
universities. And so, it’s just interesting to think about—
Matt Teichman:
—I feel like the role of engineers, overall,
is a bit underestimated, in the philosophy of science.
Oliver Traldi
I think that’s right. I do think that
philosophers of science sometimes—you know, we like really clear
divides in philosophy, divides between things like theory and
application. John Rawls had this theory of reflective
equilibrium. John Rawls’s theory of reflective equilibrium is that we
are always going back and forth between what are our plausible
principles, or what are our well-supported theories, and what are our
judgments about individual cases, or what results do we get in
experiments.
So if you believe in a theory strongly enough, you might think, well, this experiment must be wrong, because it contradicts my theory. And that wouldn’t necessarily be irrational, but you would probably want to run the experiment again. You wouldn’t want to just discount it. But yeah, I think certainly engineers, and people who are working on specific instances of things, are participating in the same process of—I don’t know, all the terms for this I don’t like—but knowledge production, or knowledge creation, or whatever. Or information gathering. Data generation.
Matt Teichman:
Observation, as it’s sometimes called.
Oliver Traldi
Yeah.
Joseph Diller:
So back to you political experts, how do you
think we should respond to the fact that political experts are more
often wrong than right in their predictions? Should we take this to be
a crisis, or should we just have low standards for political experts?
Oliver Traldi
For me personally, at least, the main reason
not to take anybody as a political expert has to do with just—you
know, no a priori argument about how society is complex, or about
morality, or whatever—but just the fact that everybody who claims to
be one seems to make some incredible mistakes the next day.
Matt Teichman:
There’s a lot of mutual contradiction as well,
between experts.
Oliver Traldi
Yeah. There’s something general to say about
like political cognition, which is that—
Joseph Diller:
Is it just more of a tribal activity?
Oliver Traldi
Yeah, exactly. Again, it depends on who we’re
looking to be the experts. But if the people we take to be political
experts are themselves political actors, then they’re always going to
be having some other sort of motivation—always trying to present
themselves in the best light, maybe trying to win some other election,
or get some other appointment.
Matt Teichman:
It seems like the terrain is ripe for bad
incentives, in that scenario.
Oliver Traldi
Yeah, it seems like the incentives are always
going to just be really, really horrible. Here’s an example of a claim
about expertise. In toothpaste ads, it’ll often say e.g. three out of
four dentists recommend this toothpaste. Now, it might be that if it
was actually the case that 75% of dentists recommended the same
toothpaste—
Matt Teichman:
—it really seems high!
Oliver Traldi
Yeah, that seems really high, right? There’s a
lot of toothpaste out there.
Matt Teichman:
I mean, there are a lot of cranky people in
the field. You’re going to get all those people to agree?
Oliver Traldi
Yeah, it seems really unlikely that you’re
going to get all these dentists to agree. So it just seems like when
you see that, people are savvy to this. People are not always good at
assessing information, but they can be quite good at assessing other
people’s motivations and incentives. And actually, people are quite
cynical. So I think most people, when they see that, they don’t think,
oh, wow, you know, 75% of dentists, that’s really good. They think,
well, they probably found the dentists who would agree with them, and
put them in the ad, or they paid them to agree with them, or
something.
Matt Teichman:
They kicked all the ones who didn’t think
Colgate was the best toothpaste out of the American Dental
Association.
Oliver Traldi
Exactly. Yeah. Maybe it’s all a scam, or a
racket. My dad loves using the word “racket”. He thinks cell phone
companies make the charging ports really bad, so that you have to keep
buying more chargers, which isn’t so hard to believe. I think a lot of
people have this view about people who are trying to gain power in
politics, through claims of expertise.
Part of the problem is that in what’s sometimes called the current information ecosystem, or the current attention economy, if somebody wants their opinions to be heard—if an expert even wants somebody to be able to find them as an expert—they need to engage in a lot of self-promotion, and a lot of alliances, and finding the right platform, and things like that. And those are activities that could have a lot of other motivations, like attention, and money, and talking to women who spend a lot of time online, and things like that. It’s really hard to find a way to become recognized as an expert that couldn’t also be explained through some much more self-interested motivations. I don’t want to say nefarious motivations; a lot of them are just the motivations we all have.
We do this sort of reasoning about people in politics a lot of the time. There are certain things that are factored in. If you know what somebody’s brand is—a lot of expertise these days is about branding—you might say, well, of course, this person is going to take this line on this thing, and that person is going to take that sort of attitude towards it. This person will make this prediction; that person will make that prediction. If you read a lot of pundits, if you read them for a long time, you can become a little ChatGPT in your head, and just produce some of the analysis that the pundits that you read all the time would.
Matt Teichman:
‘I’m sorry, I’m only a language model. I
wouldn’t know about such things.’ You start saying that to yourself.
Oliver Traldi
I think these sorts of things are part of the
challenge of political expertise. Once somebody involves themselves in
making policy recommendations, and trying to get positions, and trying
to have influence, then it becomes a little less clear what they
want. The most trustworthy experts are often the ones who are sort of
like: well, dinosaurs are the most interesting thing in the world, and
I just don’t care about anything else. I learned about dinosaurs and I
get no real credit for it; I don’t make much money. I’m just the
‘paleontologist’? Is that right?
Matt Teichman:
Yeah.
Oliver Traldi
I’m just the paleontologist at the local
university.
Matt Teichman:
I’ve liked them since I was three years old.
Oliver Traldi
Exactly. It’s just the thing that I’m into. And
if somebody has that sort of interest, it’s really easy to say, well,
of course somebody who is so into dinosaurs is not into having false
beliefs about dinosaurs. They’re really into having true beliefs
about dinosaurs. So the motivation is there, but we do a lot of
this. We do a lot of assessing people’s motivations. People start
talking to us, saying, here’s the truth about this, here’s the truth
about that. And we do a lot of saying, “Well, why are they even
talking to me?” Why did they start this conversation? What do they
have to gain from it? Sometimes we’re too cynical, and we ask that
question too much, and sometimes we’re insufficiently cynical, and we
don’t ask that question enough.
Joseph Diller:
Would you say a virtue like earnestness is
more valuable in the political arena than in something like the law
or, you know, in medicine?
Oliver Traldi
I think earnestness is a great virtue in all
walks of life. I would say earnestness is certainly in short supply,
in the political arena. That’s a little bit different than saying that
people lie a lot. There are mechanisms for punishing people if they
lie too much, and there are incentives to tell the truth. So
certainly, it’s important to have the right organizational
structures.
You know, if a reporter says there was a horrible event at UChicago today, and it turns out this event just never happened, it’s important that that reporter loses their job. If a reporter just reports something that never happened, there should be some sort of consequences for that. I assume at most mainstream publications, you can’t just make things up.
Matt Teichman:
And they’re expected to publish retractions,
etc.
Oliver Traldi
Exactly. So a correction or a retraction would
come out. Even, possibly depending on the nature of the story, they
could face a lawsuit. So there’s all these incentives—there’s all
these organizational structures—to keep people being trustworthy, or
truthful. But it’s a little bit different than somebody who just has a
native earnestness, or love of truth that somebody naive like
Aristotle might have thought everybody had.
Matt Teichman:
You have this sense, sometimes, when talking
to people, that you’re just having a conversation. Rather than a
conversation with some hustle happening simultaneously, in the
background.
Oliver Traldi
That’s exactly right. And there are ways to
tell the truth and still try to get something out of it. I go on a
certain number of first dates from dating apps, and certainly it’s not
like you’re going to make up a whole life story, but you’re also going
include—you know, you want to impress somebody—you’re going to
include the aspects that are impressive.
Matt Teichman:
Absolutely, it’s a job interview
situation. You want to make the next round.
Oliver Traldi
Yeah, exactly. But there’s levels, even in
that. I was applying for a post doc this year, before things got going
with the job I’m going to take at Tulsa, and my research interests
didn’t quite fit with the center that I was applying to. They said,
“Well, just say you’re going to research this and this and this.” I
said, but I’m not going to research any of those things.
They said: well, you can decide what you do when you get there, but if you want the job, you should say that you’re going to research these things. It took me a little aback, because those aren’t the circumstances under which I would say that I’m going to do something. I would usually only say that I’m going to do something if my plan involves doing it. So yeah, there’s different levels.
Matt Teichman:
Hot take! Hot take of the afternoon.
Oliver Traldi
I don’t know. there’s different levels, and I’m
not even saying that I’m a particularly earnest person. I think I act
out of self-interest a lot, and I can be a little slippery
sometimes. But yeah, earnestness is a really good quality, and it’s
definitely in shorter supply than the sort of incentives that keep
institutions in line, sometimes.
Matt Teichman:
Oliver Traldi, thank you so much for coming on
Elucidations. We’ll have to have you back again.
Oliver Traldi
Thank you so much for having me.
Elucidations isn't set up for blog comments currently, but if you have any thoughts or questions, please feel free to reach out on Twitter!