Arnold Kling  

Deception and Signaling

PRINT
Contest: What to Measure?... Obama on Leno...

In Be the Solution, Michael Strong writes (p. 66-69),


Are altruists occupationally prone to anger? Well,, it turns out that they are, in fact, biologically inclined to be angry and punitive toward those who they perceive to be not being helpful.

Evolutionary psychologists believe that status is correlated with perceived community altruism because in part it was in our evolutionary interest to prevent free riders...

we are willing to punish those who do not contribute to collective action even at a cost to us, another finding that is inconsistent with rational choice...

Thus, the very fact that we have moral impulses to support the public good is necessarily intertwined with the facdt that we have moral impulses to punish those who do not (and to punish those who do not punish those who do not, and so on)...

This instinct is especially harmful when used to punish those who are perceived not punishing free riders. This is the source of the bigotry against market economics among the do-gooders: It is believed that those who describe the positive outcomes of free enterprise are not doing their job to behave punitively toward free riders, and that therefore they, too, must be punished.

In other words, because economists do not want to punish rich people, altruists believe we must be punished.

But it's worse than that. You can signal that you are an altruist not by engaging in altruistic acts, but simply by expressing a desire to punish others. For example, by taking away AIG bonuses, you do a great deal to signal altruism, even though the actual social gains from taking the bonuses away are miniscule (the gains may even be negative).

This topic of signaling and deception is worth some extended remarks, below.

1. Tyler Cowen and Robin Hanson both think that signaling and deception are very important. If you listened to their bloggingheads and cannot remember anything about signaling and deception, then you need to listen again.

2. The most original idea that Robin Hanson has about health care expenditures is not that much spending is wasted--many other economists believe that. The original insight is his explanation for high spending, which is that we encourage the health care spending of others in order to show that we care. This hypothesis has the added bonus of helping to explain why so much of health care spending is paid for by "insurance."

3. I thought that the most fascinating parts of Tyler's Discover Your Inner Economist where the parts on self-deception. I wish that Tyler would do a whole book on deception and signaling. Alternatively, maybe I should do a book on Masonomics, and include a large chapter on the importance to Masonomists of deception and signaling.

4. I think that signaling and self-deception are in the back of Tyler's mind in his blogging heads with Peter Singer. (Note that Singer's debating technique is to begin by appearing to concede Tyler's points, and then to push back. I call this the "Yes, but" approach.)

5. A reason to focus on deception and signaling is that they appear to have great evolutionary survival value. Even stupid plants have evolved powerful tools for deception and signaling, looking tasty when it is helpful and looking unappetizing when that is helpful. Animals have many behaviors that are designed to deceive--think of animals that appear more fierce than they are, for example. But in all of nature, the most powerful tool for signaling and deception is the human brain.

It is plausible that a great deal of the evolution of the brain has been to make us better players of the game of deception. Think about that.

6. A lot of mating behavior involves signaling and deception.

7. A lot of political behavior involves signaling and deception.

8. Perhaps a lot of economic behavior (think of marketing and sales) involves signaling and deception.

9. As we become wealthier, perhaps signaling and deception increase. We do not have to focus as much on meeting basic needs, so we have more energy to devote to competition for status, which is mostly a matter of signaling and deception.

10. Michael Strong suggests that altruists may be particularly inclined to punish alleged free riders. But it could be that you signal altruism by showing an inclination to punish. You don't necessarily have to be an altruist if you want to take away AIG bonuses. You could be very selfish, but coming out against AIG bonuses is a cheap way to signal your altruism.

11. On the other hand, what is being signaled by those who are skeptical of taking away AIG bonuses?

12. I am not as comfortable as other Masonomists are with using signaling explanations. The opportunity for "just-so" stories strikes me as too great. "Counter-signaling" seems to me to take signaling completely out of the realm of testable theory and into the realm of nonfalsifiability. If a peacock growing a useless tail can be explained as a signal, then what would falsify the theory of signaling?



COMMENTS (25 to date)
eccdogg writes:

"I am not as comfortable as other Masonomists are with using signaling explanations. The opportunity for "just-so" stories strikes me as too great. "Counter-signaling" seems to me to take signaling completely out of the realm of testable theory and into the realm of nonfalsifiability. If a peacock growing a useless tail can be explained as a signal, then what would falsify the theory of signaling?"


This is the most important point in my opinion. Most of the evolutionary behavior theories that I read are interesting but seem to me to be a nice story that fits what we observe. But I can also come up with other stories that fit the observations as well and neither is usally falsifiable in any real sense.

Arnie writes:

"nonfalsifiability"? It isn't in my personal lexicon, and it isn't in my spell checker either.

I was enjoying this post until Arnold main caveat in point 12, which I couldn't follow.

Anyone care to help me out?

William writes:

Arnie -

Point 12 hinges on your understanding the concept of "just-so stories."

http://en.wikipedia.org/wiki/Just-so_story

Or google it.

Giedrius writes:

"If a peacock growing a useless tail can be explained as a signal, then what would falsify the theory of signaling?"

Maybe cutting off peacock's tail and examining the changes in its reproductive success?

David W. writes:

Nonfalsifiability is a reference to one interpretation of the scientific method - it states that a theory is only valuable if it excludes some things from being possible. If a theory is falsifiable, then there's something you could see that would prove the theory wrong or at least incomplete.

For instance, if your theory is that gravity near earth is 9.81 m/s^2, and you drop something that falls a lot faster, there's something wrong with your theory. If your theory is that things fall if they feel like it, and as fast as they feel like, then your theory can't be tested, because there's nothing you could see that would prove it wrong. More importantly, the theory isn't any use. You can use the first theory to predict ballistics and design artillery, but you can only use the second theory after the fact to say 'yep, that's what I thought would happen'. It's the difference between a reason and a rationalization.

Basically what I read Arnold as saying is that he's uncomfortable using signaling explanations for things people do, because they seem to him more like rationalizations than a theory that could predict behaviors in advance.

stephen writes:

I am growing to love signaling as an explanation for many causal relationships. Thanks g-mason! perhaps the abundance of "just so stories" is some bit of evidence the robustness of the general theory.

The more I try to approach situations with signaling in mind, the more I am struck by how obvious it is.

Wasn't it amazing how difficult it was for Singer to just come out and say "I support tax cuts for the rich to help the poor"? He just couldn't do it without thrice restating the same caveats that Tyler built into the question. Even when it comes to a partial solution to his most cherished cause he just can't show any support if it doesn't properly singnal what group he is affiliated with. Amazing!

fenn writes:

"Alternatively, maybe I should do a book on Masonomics, and include a large chapter on the importance to Masonomists of deception and signaling."

Yes, please.

Hugo Pottisch writes:

In the light of altruism...

I don't get the whole AIG discussion. For freaks sake - they are an insurance company. There is not enough money on the planet to compensate people who work for insurance companies.

Let's get real. You really really live only once. Somebody must take care of insurance. Who? I mean - no sane, smart, attractive people would want to do that? Masochists?

No - I do not care how many billions or trillions AIG and other insurers have wasted due to greed and incompetence. Every insurer that I have met so far - I pity. No matter how flashy his house and car... it is a sad sad life.

Nobody in the West is starving. Financial services are important civil services. They never dictade where we are going as do other industries (IT, Internet, Hollywood etc) - they just do boring boring support functions. bankers and especially insurers deserve to be the best paid workers - them and factory workers (same brains involved and also uniforms).

Again - in this light - are the rich insurers altruists or masochistic or both. And why do non-insurers care about the petty $160 million in bonuses - I would not swap place with them for that money and I am glad that there are some true altruist who would. We live only once - we need these people to sacrifice their lives for us and the least we can do is pay them!!!

HH writes:

"out of the realm of testable theory and into the realm of nonfalsifiability."

I had bad news for you. Much of economics and evolutionary science is like this. Anything that lacks controlled experiments is going to, at some point, appear nonfalsifiable. The reason we still have socialist tendencies in parts of the world, for example, is that you can't run a controlled study of a country in parallel universes with different policies. Sure, we come close: the Koreas and US states are great natural experiments, but because they're in fact different we can't truly "falsify" any theory. ["South Korea had X but North did not, so it's really not the same."]

We should also not confuse something we don't know how to falsify with something that can't be falsified. For example, there's a theory that twins develop slightly differently in the womb, which affects their postnatal development. I doubt that we'll have the technological ability to test this theory for a long time, if ever. But it's at least theoretically falsifiable, as I imagine the peacock's tail is.

Hugo Pottisch writes:

PS: Biologically speaking there is not altruism and selfishness.

All social, but also loner, animals have evolved to sometimes endure some short-term discomfort for later long-term benefits. Sometimes not. That is all and there ain't much more to it.

Moral maze

what is a small sentence for Darwin - the Objectivits and some economist but also esoteric left-wingers turn into mystic books and philosophies... why - if their intellect is so easy to please. I bet they prefer the tredmill to running outside as well?

my issue with singer is legal and not philosophical. if the group indeed needed more rights than the individual - we would not need any laws at all.. ever.

but might does not make right and therefore it is the libertarian idea that actually demands laws and hence justice and hence a monopoly of power and hence a state of some form. Libertarians demand that it is limited - but they are the only ones who need it. It is ironic that the issue of "self-defense and guns"... never mind.

Dave writes:
This is the source of the bigotry against market economics among the do-gooders: It is believed that those who describe the positive outcomes of free enterprise are not doing their job to behave punitively toward free riders, and that therefore they, too, must be punished.

That's quite a claim. Very hard for me to believe. It is much simpler (and to me, more likely) to assume that anti-market people believe that free market policies lead to bad outcomes, e.g., negative externalities, fraud, irrational consumption behavior, and happiness-reducing levels of inequality. Which of course they often do.

Also, I am personally pro-market, pro-altruism, and anti-free-riding, and I see no contradiction between those stances--they (can) all promote general welfare. This goes against the altruists-punish-economists idea.

In general ethical and political terms, it would seem the book authors are on the same page as me, perhaps even with the same propensity for moderately tenuous speculation to support their positions. :-)

Dave writes:

Oh, I forgot to mention that all this talk of "signaling and deception" is weird to me. (Point 12 FTW.) Is that what we used to call "communication"? If so, I agree, it's valuable and ubiquitous, and social behavior can't exist without it, but I don't expect anyone didn't know that.

Some of the related claims are overblown, too. For example, the peacock tail as fitness indicator is questionable in the light of recent research.

El Presidente writes:

Hugo Pottisch,

. . . [I]t is the libertarian idea that actually demands laws and hence justice and hence a monopoly of power and hence a state of some form. Libertarians demand that it is limited - but they are the only ones who need it.

Monarchists don't demand a state of some form? I'm perplexed. I would suggest that the need for state authority is an embarrasing fetish of libertarians that they only admit to one another after a few drinks behind closed doors with shades drawn and lights dimmed, but I'm not sure they are alone in adopting a philosophy that suggests states have legitimate roles. They're just repressed, I suppose. :-)

My reading of history and philosophy shows evidence of the presence of states across time, place, and culture explained various ways and for various reasons. Some are compelling, and some are not. Proffering the notion that this is all about self-deception and signaling seems to me like entering a long drama after intermission and trying to explain the intent of the writer based entirely on the second half of the show. Frequently, the precursor to a desire for security is the perception of insecurity. Where did that originate? My guess is it emerged in the first act.

I'm with Arnold in thinking that this is a sketchy psychological explanation. It might be true, but what argument can be made for it? I don't see anything other than conjecture to recommend it.

liberty writes:

"we are willing to punish those who do not contribute to collective action even at a cost to us, another finding that is inconsistent with rational choice..."

But consistent with group selection?

Are we, according to the facts in this book, not possible evolving a solution to public goods problems--to the tragedy of the commons? If we are, then perhaps Marx was right in a sense after all: one day human nature will have changed so that communism is possible.

But actually, he would still be wrong. Even if we all become altruists, and happily punish any breaches in altruism, we still need profit and loss, and decentralization via private property, for calculation. Otherwise, how would we know what to produce? Planning doesn't work, and that has nothing to do with altruism. Although, it would work somewhat better, perhaps.

kebko writes:

Dave,

Isn't that the point? Signaling leads people to overemphasize the negative outcomes of markets. There are positives & negatives. While in reality the positives heavily outweigh the negatives, perceptions are almost universally focused on the negatives. What reason could there be for this oddity? The speculation here is that it is signaling.

Tom Church writes:

Arnold,

I've been waiting for a Masonomics book for a while now. Please put one together!

AB writes:

Do we fight wars based on conflicting styles of altruism? I.E. collectivism vs. individualism, Free trade vs. protectionism, etc...

Arnie writes:

William and David. W. - You filled in the concept for me. Thank you very much.

Jayson Virissimo writes:

"Basically what I read Arnold as saying is that he's uncomfortable using signaling explanations for things people do, because they seem to him more like rationalizations than a theory that could predict behaviors in advance." -David W

Most of evolutionary biology and economics cannot predict in advance what will happen. They simply explain what has already happened using a few basic principles. When is the last time a biologist predicted the arrival of the new species and what its traits would be? They explain new species using natural selection and genetic drift only after the fact. It is barely falsifiable. Complex things like living organisms and markets are very messy and aren't as simple as the motion of bodies.

Patri Friedman writes:

I am fascinated by signaling, it is a fairly new topic to me and it seems there is a lot of meat there. It's the hot new area for EvPsych / behavioral econ.

Re #6, note that PUA, which while there are many bad things to be said about it is currently the only movement with serious scientific analysis of how men can get laid, is largely about signaling and deception. (Specifically, signaling that you are an alpha male)

Note that #7 and #9 are related. In #7, we signal in politics more than other areas because wrong political opinions are cheap. In #9, you point out that everything is getting cheaper because we are getting richer.

Robin Hanson writes:

I responded to this post here.

rhhardin writes:

think of animals that appear more fierce than they are, for example.

``[The Hog-nosed snake] could lead an ideal life if he had any sense. He is perfectly harmless, but he pretends to be a deadly African Puff Adder. Then he wonders why people kill him.''

Will Cuppy, _The Great Bustard_, p.103

fundamentalist writes:

I find it very hard to take evolutionary psychology seriously. Actually, I have a hard time taking psychology seriously, but add "evolutionary" to it and it becomes doubly hard. It seems to me that the stories these "scientists" make up are limited only by their imaginations. They are know different than mystery fiction writers.

And I think the importance of signalling and deception is overblown. It took decades of "science" to figure at that people are dishonest? This is one way in which the common man loses respect for science in general and economics particularly. The common man has known from the beginning of time that people are dishonest. Then an academic comes along and invents new words for it (such as signaling) and claims to have made deep insight into human nature.


Of course people are deceptive. Adam Smith recognized that. A major justification for state intervention is to prevent such deception, but Smith argued that the state makes deception worse, not better. The free market handles deception in marketing (signalling) by punishing the dishonest merchants.

Troy Camplin writes:

It has recently come out that cheaters/free riders are often the most likely to come out in favor of punishing other cheaters. Why? To reduce the competition. It doesn't benefit one to cheat if everyone is cheating, after all. So not only do the altruists want to punish cheaters, but other cheaters do as well. Which explains the outrage from our government officials. It also explains the existence of hypocrisy.

melancholyaeon writes:

Arnold,

I've had the privilege of being acquainted with Michael Strong for nearly 30 years now. He has always been one of the most intelligent and original people you could imagine. Even as a young man he was dazzling. Everything he says has always been of deep interest.

Comments for this entry have been closed
Return to top