From Edge confab on What Scientific Concepts Should Be Retired? the estimable Jonathan Haidt comes out against Occam's Razor:
Jonathan Haidt
Social Psychologist; Professor, New York University Stern School of Business; Author, The Righteous Mind
[Anti-] The Pursuit of Parsimony
There are many things in life that are good to have yet bad to pursue too vigorously. Money, love, and sex, for example. I'd like to add parsimony to that list.
William of Ockham was a 14th-century English logician who said that "entities must not be multiplied beyond necessity." That principle—now known as "Occam's Razor"—has been used for centuries by scientists and philosophers as a tool to adjudicate among competing theories. Parsimony means frugality or stinginess, and scientists should be "stingy" when building theories; they should use as little material as possible. If two theories really do exactly as good a job of explaining the empirical evidence, then you should pick the simpler theory. If Copernicus and Ptolemy can both explain the movements of the heavens, including the occasional backwards motion of some planets, then go with Copernicus's far more parsimonious model.
Occam's razor is a great tool when used as originally designed. Unfortunately, many scientists have turned this simple tool into a fetish object. They pursue simple explanations of complex phenomena as though parsimony were an end in itself, rather than a tool to be used in the pursuit of truth.
The worship of parsimony is understandable in the natural sciences, where it sometimes does happen that a single law or principle, or a very simple theory, explains a vast and diverse set of observations.
But in the social sciences, the overzealous pursuit of parsimony has been a disaster. Since the 18th century, some intellectuals have striven to do for the social world what Newton did for the physical world. Utilitarians, the French philosophes, and other utopian dreamers longed for a social order based on rational principles and a scientific understanding of human behavior. Auguste Comte, one of the founders of sociology, originally called his new discipline "social physics."
And what do we have to show for 250 years of pursuit? We have a series of time-wasting failures and ideological battles. Human behavior cannot all be explained by positive and negative reinforcement (contra the behaviorists). Nor is it all about sex, money, class, power, self-esteem, or even self-interest, to name some of the major explanatory idols worshipped in the 20th century.
In my own field—moral psychology—we've suffered from the same overzealous pursuit of parsimony. Lawrence Kohlberg said morality was all about justice. Others say it's compassion. Others say morality is all about forming coalitions, or preventing harm to victims. But in fact morality is complicated, pluralistic, and culturally variable. Human beings are products of evolution, so the psychological foundations of morality are innate (as I and many others have argued at Edge.org in recent years.) But there are many of these foundations, and they are just the beginning of the story. You must still explain how morality develops in such variable ways around the world, and even among siblings within a single family.
The social sciences are hard because human beings differ fundamentally from inanimate objects. People insist upon making or finding meaning in things. They do it collectively, creating baroque cultural landscapes that can't be explained parsimoniously, and they do it individually, creating their own unique symbolic worlds nested within their broader cultures. As the anthropologist Clifford Geertz put it: "Man is an animal suspended in webs of significance that he himself has spun." This is why it's so hard to predict what any individual will do. This is why there are almost no equations in psychology or sociology. This is why there will never be a Newton in the social sciences.
Let's retire the pursuit of parsimony from the social sciences. Parsimony is beautiful when we find it, but the pursuit of parsimony is sometimes an obstacle to the pursuit of truth.
Let me suggest a distinction: Let's retire the pursuit of constructing one-size-fits-all Grand Unified Theories of All Human Behavior.
But let's very much retain the use of Occam's Razor for pointing out problems with social scientists' ideas.
For example, here's a graph of social mobility, of a child rising from a family in the bottom quintile of the national income distribution to the top quintile, made by award-winning Harvard economist Raj Chetty that has been widely publicized, along with Chetty's convoluted theories for why it has so many counter-intuitive findings, such as that the American Dream Is Alive in the dusty, dying small towns of the Great Plains but the American Dream Is Dead in bustling Charlotte, NC and the greater Atlanta metroplex.
My model of the map is a lot simpler:
1. Poor blacks (and poor Indians on big reservations) regress toward a lower mean income than poor whites regress toward. Regression toward the mean is a hugely useful tool in any analysis covering two or more generations. That's exactly the problem that Galton invented the concept to deal with in his 1869 book Hereditary Genius. To ignore regression toward the mean in this kind of analysis should get the authors laughed at, but instead it gets them praised.
2. Chetty hasn't dealt well with the complications caused by diverging land prices, related changes in income, and migration (e.g., according to Chetty's map, the American Dream Is Alive in West Virginia, but that's most likely an artifact of so many Mountaineers moving to places like Charlotte).
In summary, while I have zero intention of building a giant overly-simplistic ideology to predict everything (conversely, my most ambitious philosophical hopes are to construct a simple explanation of why we can predict so few things of interest to us, like the Super Bowl; personally, I think the Seahawk defense is going to expose the fact that Peyton Manning is a middle aged guy and ... oh, where was I? Okay, back to boring stuff that nobody is really interested in), Occam's Razor remains the single most useful tool for slashing down faddish new conventional wisdom, like Chetty's bogus analyses.
Metaphorically, perhaps it's better to think of Occam's razor more as a weapon for chopping down than as a tool for building up. I suspect Dr. Haidt and I could agree on that
Posts
59 comments:
Steve, You're being discussed quite a bit on Rod Dreher's blog today.
The parsimony concept has been oft abused by pseudo-sophisticated writers in political economy; the various "What's The Matter With [x]" tracts come to mind, and practically everything by David Brooks (Bobos, red/blue discord, Target vs. Kmart, Patio Man, blah blah blah). Though I suppose Patio Man in the sprinkler city may be a kind of failed parody of Huntington's Davos Man.
SS: Peyton Manning is a middle aged guy...
37 is middle aged?
SS: Peyton Manning is a middle aged guy
37 is middle aged?
Haidt is right - but what he is objecting to isn't parsimony but reductionism (it's all about sex, power, etc).
The project to make the social sciences like physics will never be retired precisely because it never yields certain results; it is a fertile field for people to play our their dreams and will remain so until it begins to yield certainty, which is never, of course.
And physics of course isn't "correct" or "reality" - which we can never know - it simply allows us to control things through theoretical approximations that have frequently proven wrong or inadequate even as they were useful. Mathematics adds not one iota to our understanding of a phenomena, but it does give us power.
And of course, human nature is "all about" power.
What Haidt is against is leaving out bits of evidence so as to get to a simpler theory.
"37 is middle aged?"
Not if you are, say, a symphony conductor. And if Manning can somehow continue to almost never be sacked, he may come through the Super Bowl feeling fine and fettle. Then again, Seattle's defensive priority is probably to make sure Manning feels his age all of a sudden.
Based on how Haidt defines 'occam's razor', I would have to agree with him.
When we are dealing with a complex phenomenon and don't yet have the answers, we should not jump to conclusions with a single theory as the explanation/solution for everything.
So, with stuff we really don't know much about--or how things may eventually turn out--, let's use occam's probe than occam's razor.
It's like we didn't know how Russia would be like after the collapse of communism. Some thought 'free market liberalization' would be a panacea that would solve all problems. Not so. Likewise, we can't just plant democracy in the Middle East(especially as the West itself is becoming more oligarchic and less democratic).
But there's another meaning of occam's razor. The way most of us mean it, it applies to cases where there's overwhelming evidence that supports a certain observation or statement. For instance, blacks are better at sports. So, the occam's razor explanation for black domination in football would be 'they are faster and stronger'. This doesn't mean all blacks are strong or fast nor that non-blacks don't have strong and fast people. It means there's more such people among blacks, and blacks outperform others in sports.
On this issue, if someone goes against occam's razor to avoid charges of 'racism' and instead argues that blacks are good at football because they historically picked a lot of cotton, that's just occam's snake oil.
So, no occam's razor where the evidence is not yet available. But occam's razor is the best bet when there's an abundance of evidence that a lot of people are simply afraid to sift through out of fear of what they might find(or fear of what might happen to them if they were to find it and talk about it).
Occam's Razor has to go because using it requires that people pay attention and notice things, and also requires that people come to politically incorrect conclusions.
In football? Hell, it's late middle age!
A ticking bomb.
If you don't know which wire to cut, don't jump to conclusions--bad occam's razor--and ask around if someone knows.
But if you do know what wire to cut--good occam's razor--, just cut it and stop wasting time.
There probably ought to be a rule in social sciences that first, you compare your map to a map of the percentage of the black population. Then you discuss how it's different.
The social mobility map - Hispanics seem to be pretty socially mobile, compared to blacks. But maybe that's an artifact of their recent increase in numbers. New York, the Bay Area, Boston, Houston, seem to do pretty well. That's something interesting - why not Chicago? Why not Miami?
Who was it that said that all procedural arguments are insincere? Haidt doesn't like Occam's Razor because it doesn't produce the results he likes in this case. In some other situation where it gets the results he likes, he would be the first to wield it. Just like the NY Times is for the filibuster when Republicans control the Senate and against it when they are not.
There have been a lot of football games this season where Manning's physical experience of them hasn't been all that different from a symphony conductor's -- Peyton waves his arms around a lot and almost nobody knocks him down. He was only sacked 18 times in 16 regular season games. Obviously, he's not very mobile, so this a tribute to his mastery of his job.
If the Super Bowl is like that too, then the Broncos will likely win.
Then again, I presume the Seahawks are taking their 13 days of planning to try to make Manning feel less like a young conductor and more like an old football player at the end of the Super Bowl.
We shall see.
"Hispanics seem to be pretty socially mobile, compared to blacks. "
A lot of that is the income boost from Spanish-speaking first generation to English-speaking second generation.
In New Mexico, where there are lots of Hispanics but few immigrants, social mobility is lousy, as the map shows.
Haidt makes good sense on the subject of theories, and I don't think he's using the term 'occam's razor' in the way we mean it.
For Haidt, a good example of 'occam's razor' would be 'inequality would go away if we spend a lot of money on pre-k education'.
It focuses on a single theory to explain and solve all problems.
I think Haidt is more concerned with the problems of theorizing and proposing solutions than with observations of fact.
On the issue of morality, Haidt is right. Too many scientists try to explain everything with a single paradigm--evolutionary theory, etc. He calls this 'occam's razor', but it's not how we mean it.
For us, occam's razor simply means EVIDENCE IS ABUNDANT, IT'S EVERYWHERE, BUT SOME REFUSE TO SEE IT BECAUSE OF CURRENT ORTHODOXY.
(I wonder if Chua steered clear of the pitfalls of occam's razor because hers is a triple-razor).
Occam's Razor, properly understood, must reflect the Einstein apocrypha-quote: "Everything should be made as simple as possible, but no simpler.”
Emphasis here on the "but no simpler".
Occam's Razor chooses the hypothesis with the smallest number of assumptions, but only under the condition that all of the competing hypotheses fully explain the phenomena in question.
In still other words, it's not in general true that
Occam's Razor = Reductionism
So leave Occam's Razor alone!
Steve, two senior JP Morgan executives committed suicide on the same day. Coincidence or there's more to the story?
If the two financial guys had been wearing miniskirts when they killed themselves, would Senator Elizabeth Warren be the subject of a Two Minutes Hate?
Occam's dildo.
It's what SWPL use to feel good about themselves. It doesn't have to be true as long as it FEELS true within their PC bubble.
Maybe what Haidt is talking about should be called Occam's (magic)wand or Occam's glass key.
"Steve, two senior JP Morgan executives committed suicide on the same day. Coincidence or there's more to the story?"
Uh-oh, maybe Drudge was right about 'have an exit plan'.
http://www.freerepublic.com/focus/f-chat/3116452/posts
I'd like to see Occam's razor used to sever permanently the word "social" from the word "science."
the success of rap.
occam's beat.
just enough groove and beat to make you feel badass, thuggish, and sexual.
So much of modern discourse on race is like Zeno's paradox.
We must pretend that we cannot catch up to the turtle. There's always something and then something and then something and then something that prevents us from bridging the gap.
Candid_Observer is close.
Occam's Razor only means that when all explanations are unproven, yet you have to take action, choose the explanation with the fewest moving parts.
That's all.
It isn't a proof of anything. It isn't the end of investigation. You still go on investigating until you have a proven explanation.
It's simply a wager - and the principle that simple explanations should be investigated first. If the simple ones don't hold up, then you look into more complicated ones..
Haidt's idea, by contrast, could be called "the Rube Goldberg principle." That is, begin out of the blue with the most complicated explanation, and proceed from there.
That principle throws all into confusion: it's perfect for obfuscation; it's a kind of squid ink.
But what could induce Haidt to suggest it?
He says: "In my own field - moral psychology [...]"
Oh.
"Moral psychology" - aka "Bullsh-t Hot Air Studies"
Does it matter that Jonathan Haidt is a jew and William of Ockham was a franciscan friar?
Could it be that William's insight derived from the christian virtues such as temperance instead of parsimony?
Occam's Incomprehensible Post Modernist Thesis, Peer Reviewed with Copious Citations
Occam's Blinders
Occam's Head in the Sand
Occam's Ball Gag and Collar
Occam's Colonscope
Occam's Lobotomy
For 21st century, it should be updated to Occam's Laser.
But PC goes for Occam's Taser. Stun and knock down anyone who dares to violates the taboo.
And there's hokum's razor too, which is most of what not-so-sharp Edge has been peddling lately.
Oxfam's Fundraiser, the way to end poverty all over the world.
No poor country by 2036, says Bill Gates.
Kafkian logic is the opposite of Occam’s razor. Haidt’s article is a Kafkian wet dream of multiple unrelated facts that confuses NOT enlightens.
He is striving to prove something that is untrue – namely that the blacks main problem is white racism. Not true today – it was true – but not today. Today the blacks main problem is their family structure that was destroyed by Haidt’s liberal NY predecessors. No father in the home is the big problem for blacks – end of story.
By design or ignorance Haidt is engaged in debilitating power seeking propaganda. Without question he is hurting black people with this article.
Occam's Haidtjacket
Isn't intellectual parsimony how the Flynn effect was explained by that boy-genius a few months ago?
If physicists and chemists started directing their hefty intellects towards human behavior, guys like Haidt would be relegated to the sidelines as data collectors and research assistants or just enthusiasts. Goodbye status.
Luckily for him, a politically correct atmosphere keeps the brightest people away from fields where inconvenient truths lead to ostracism.
Elephant in the Room: politicized science should be retired.
To me, occam's razor is simply applying the general explanation to a specific case, rather than assuming a special explanation. So why does one race score higher on IQ tests than another? The same general reason as anyone outscores anyone. Because they're smarter. Why is one race smarter than another? The same general reason any contemporary American is smarter than any other. Genetics, since IQ has an 80% heritability by mature adulthood.
Or what's causing the flynn effect? The same thing that's caused any other variable (height, brain size, birth weight) to increase over the 20th century. Nutition.
Now the people who deny that the flynn effect reflects a genuine biological increase in intelligence are violating occam's razor.
Doesn't occam's razor imply obama's a genius? The simplest explanation for how he was able to do so well at harvard, write such a literary masterpiece and become the most powerful black in human history is that he's a genius.
Other common internet explanations invoke multiple theories that combine affirmative action, ghost writers, conspiracies by Jews and/or the CIA etc, and thus violate occam's razor. So maybe we should rename it obama's razor, since it proves Obama has a razor sharp intellect.
If physicists and chemists started directing their hefty intellects towards human behavior, guys like Haidt would be relegated to the sidelines as data collectors and research assistants or just enthusiasts. Goodbye status.
It's always fascinating when people with super genius intellect suddenly decide to compete in fields that are cognitively beneath them. A good example is business. The best business men are alpha males that have IQ's in say the 130s, and by the late 20th century they made if they were lucky, several billion bucks.
Then along came a scrawny four eyed beta nerd businessman like bill gates, with an IQ perhaps as high as 170, and by the late 1990s he's worth $100 billion...so rich the jealous Clinton admin starts persecuting him.
The alpha males of the world are very lucky that most super genius beta nerds are happy playing video games all day, because when they do decide to compete for real success, it's an absolute massacre.
Occam's Razor only means that when all explanations are unproven, yet you have to take action, choose the explanation with the fewest moving parts.
Yes. Be willing to consider other explanations, but start with the simplest explanation and work from there. If I can't find my wallet, the simplest explanation is that I've mislaid it, so I look in the places I've been recently. It's also possible that someone snuck into my house without my seeing him and stole it without taking anything else -- but that requires many more unlikely things to have happened together, so I'm not going to grab the phone and call the cops right away.
His definition of Occam's Razor is just plain wrong. That's embarrassing.
Another way of looking at is, the greater the number of facts a theory can explain, the more justified it is by occam's razor. Dinesh D'souza would often argue that his anticolonial theory explains more about Obama than any competing theory (I.e. He's a secret Muslim, he's a socialist, he's incompetent etc)
Anyone with half a chance flees southern West Virginia and eastern Kentucky at the first opportunity--it's been that way almost forever. Nature abhors a vacuum, so them as is left naturally rise to the top. All of my grandparents were WV-born. No prizes for guessing how many aunts, uncles or cousins I have there still.
Steve, former 49'er QB Jeff Garcia said today that the Seahawks probably won't be able to pressure Manning much, so he favors the Broncos. Bet accordingly. On the other hand, well-known OR-wielder Randy Moss viewed him as gay.
Haidt might be right in that Occam's Razor can't be used in dissecting human behavior, but not because it will yield conclusions that aren't true or are incomplete, but because it will yield results we can't bear.
His definition of Occam's Razor is just plain wrong. That's embarrassing.
What do you expect from a guy who is totally ignorant in his own field.
In my own field—moral psychology—we've suffered from the same overzealous pursuit of parsimony. Lawrence Kohlberg said morality was all about justice. Others say it's compassion. Others say morality is all about forming coalitions, or preventing harm to victims.
It's about self-interest -- the self-interest of a group achieved through the mediation of personal conflicts of interest between and among group members. Science has answered this question; it's time to read David Sloan Wilson and move on.
Steve needs to think through the Chetty analysis more clearly. Places with high income mobility need not have either (a) high income disparity or (b) high absolute levels of median prosperity. You're conflating mobility with prosperity/achieving the American dream which is false. You could move from the 10th to the 90th percentile in income in the dust belt, but that doesn't mean much, nor does Chetty claim that it means much.
Occam's fist.
The negro way.
Anonymous said...
"If physicists and chemists started directing their hefty intellects towards human behavior, guys like Haidt would be relegated to the sidelines as data collectors and research assistants or just enthusiasts. Goodbye status."
Plenty have tried. None has made much progress.
Simple logic should tell you that if one could walk in and pick off the (supposedly) low-hanging fruit in the social sciences, plenty of physicists would have done so. If it's so easy, why hasn't it been solved?
It's the science version of "if only our [U.S.] best [black] athletes played soccer…"
Different skillsets. In basketball, and especially in football in the low-skill positions, it can be enough to be unusually fast/tall/long/strong to be successful. In soccer, it is never enough, though one must be fast to even have a chance. But it takes at least a decade or more of skill refinement. Nobody just "crashes" elite soccer (or even a mediocre pro league such as MLS) from another sport. But you can find former basketball players in the NFL, and (very tall) former soccer or volleyball players in the elite ranks of basketball.
"Steve, You're being discussed quite a bit on Rod Dreher's blog today."
Rod Dreher's comment box is a stain on the face of humanity.
Based on the abundant evidence for Oswald acting alone and on the utter lack of evidence for conspiracy, the Warren Commission applied Occam's razor, and to this day millions of Americans Haidt it.
Simple logic should tell you that if one could walk in and pick off the (supposedly) low-hanging fruit in the social sciences, plenty of physicists would have done so. If it's so easy, why hasn't it been solved?
That is a very good point. But I'd wager it is much more likely that a physics undergrad eventually makes a breakthrough in biology than the reverse.
1. I don't know if it's in the Chetty study or not, but a first thing to worry about is that lots of people would rather stay in their poor home town than move to New York and double their income. That's not true of people in Newark, but it is of people on Indian reservations and the rural South--- at least, of those from families that have stuck it out this long.
2. Thus, it woudl be interesting to look at the study after eliminating people who moved out of their state.
3. The regression ot he mean idea seems like it ought to be testable. Some tenured economist ought to take it up.
I saw Peyton Manning talking to the press yesterday on one of the sports channels. It sure sounded like he was implying that the Super Bowl would be his last game.
Regarding aging and athletic performance, my first aches and pains came at age 40; and I had to drastically cut back running, and strength-training plateaued. I'm still consistently winning my age group for local 5K races though.
i'm guessing chetty's conclusions were simply GIGO. threw a bunch of unrelated variables into an equation, churned the numbers, and got junk out.
Steve said . . .
Jonathan Haidt comes out against Occam's Razor
No he didn't. And, no, it does not matter that he said he did.
He is an academic who was asked a question. He answered "Now let me talk about my work and attack my enemies."
You want to retire something from science, retire academics talking about their own work in reply to EVERY FUCKING QUESTION.
This is Haidt speech and I find it offensive.
Stop the Haidt!
37 is old for a football player. not just middle aged. old. there might not be a single defensive starter in the entire league who is 37. the only guys who make it to 40 are quarterbacks, kickers, and punters.
"Nobody just "crashes" elite soccer (or even a mediocre pro league such as MLS) from another sport. But you can find former basketball players in the NFL, and (very tall) former soccer or volleyball players in the elite ranks of basketball."
accurate observations. nobody will just come into soccer and play at the highest level right off the bat. this does happen in football and basketball though. they are lower particpation rate, lower skill sports.
this is usually the way you can 'level check' a sport in the real world, without knowing about participation rate. if a great athlete can come in and get to the top in 3 or 4 years, the sport is either not that big, or the skill level required in the sport is not that high. these 2 variables interact. you can have a high participation rate, low skill level sport like track & field, where it's hard to crack into the world class in just a few years. or you can have a low participation rate, high skill level sport like tennis, where it is equally hard to crack into the world class quickly.
how long it would take on average to crack into the world class, if you started tomorrow, is a good way to evaluate the level at which a sport is operating. if it would take more than 3 or 4 years for most guys, the sport is not some minor sport that nobody plays, and either has a participation rate high enough that lots of good athletes are already doing it, or has a skill level requirement that takes years to match, or some combination. a sport becomes more than a minor sport, if it would still take a great athlete 6 or 7 years of regular play to get near the top. freak athletes can do it faster, but they're very rare. they wouldn't be freaks if they were common.
note here, tennis a low participation rate sport only by way of comparison. it's not a low participation rate sport generally speaking. a great athlete cannot wash out of football or baseball or soccer, pick up a racquet, and start winning million dollar purses in 3 or 4 years of whacking the ball around.
i have wondered the same about scientists. how much talent is general ability and how much is field specific. how would guys from one field do in another. i'm kind of in the middle there. i do think that if the smart guys from physics went into other fields, you'd see some serious breakthroughs. i consider physicists to be the smartest people in general. smarter than mathematicians. they have more general purpose problem solving ability. not as good as mathematicians at esoteric realms of math (although physicists are great at math), but more ability to break new, unfamiliar problems which nobody has encountered before. it's impressive to watch a high level physicist approach a new problem and systematically crack it over time with dazzling new insights you never would have come up with. and, i would say, mathematicians rarely would have come up with.
with mathematicians, their range of problem solving ability is more narrow. it's critical to our way of life in some instances, but it's a less broad, and usually a less useful ability. also, mathematicians often demonstrate surprising lack of contact with the real world. they can really be out of touch, and you'd never trust them to be 'normal', or make any important decisions. you'd probably have to make a mathetmatician prove he can be trusted to be a normal, responsible adult who is aware of what's going on and has a good grasp on how the world works, before you'd put him in charge of anything important.
it's not true that physicists would be good at all fields equally. it's not like they have great aptitude for every field by virtue of their physics ability. i agree with that reservation. the main thing you have to wonder today, is how much brainpower is being wasted by going into finance, instead of hard science and engineering.
Cochran is an example of a physicist who transitioned to evolutionary theorist.
Occam would also suggest that we not believe in catastrophic global warming until there is good evidence of it.
As with the "social sciences" the objection seems to be that "we want to kudos of calling ourselves scientists without sticking to the basic rules that have allowed science to work.
with mathematicians, their range of problem solving ability is more narrow. it's critical to our way of life in some instances, but it's a less broad, and usually a less useful ability. also, mathematicians often demonstrate surprising lack of contact with the real world. they can really be out of touch, and you'd never trust them to be 'normal', or make any important decisions. you'd probably have to make a mathetmatician prove he can be trusted to be a normal, responsible adult who is aware of what's going on and has a good grasp on how the world works, before you'd put him in charge of anything important.
Autism?
it's not true that physicists would be good at all fields equally. it's not like they have great aptitude for every field by virtue of their physics ability. i agree with that reservation. the main thing you have to wonder today, is how much brainpower is being wasted by going into finance, instead of hard science and engineering.
Smart people often go where the money is, because that's the smart thing to do.
Post a Comment