Comments on “Upgrade your cargo cult for the win”
Adding new comments is disabled for now.
Comments are for the page: Upgrade your cargo cult for the win
Eugene Gendlin
Really excited for the next post! Have you watched gendlin’s 3 five minute videos on Thinking at the Edge on YouTube? Highly recommended!
Courage and ignorance
Great post!
I՚m surprised you didn՚t mention an over-reliance on formalism as an particular form of cargo-cultism, particularly in AI and economics. Maybe those aren՚t sciences enough to be cargo-cult science.
The kind of epistemic virtue you are promoting requires a certain intellectual fearlessness. I think most institutions and cultures are not set up to support this, because people are spending most of their energy competing for scarce resources (money or position). Which is why it is found mostly in people like Feynman who are so smart they don՚t have to worry about competition, or in scenes or fields that are new and interdisciplinary enough that the rules haven՚t yet ossified.
Here՚s something sort of relevant I posted awhile back: http://omniorthogonal.blogspot.com/2012/06/great-and-universal-ignorance.html The first step in learning anything is admitting that you don՚t already know it, and admitting ignorance is showing weakness, so officially discouraged by almost everything in our culture.
Dzogchen
”…and is widely considered the most reliable path to enlightenment.”
Made me smile. This might come across as slightly … partisan to some readers? I’m not sure what the more cautious equivalent would be - perhaps, ‘is widely regarded within central Asian Vajrayana…’?
Long
Hi David
This is a very long essay to read and digest.
I suppose the question I have at the end of reading is something along the lines of: “How normative are the activities of geniuses like Feynman?”
As you say Feynman was the leading theoretical physicist of his day (or indeed any day) - a profound thinker with few genuine peers at thinking. He was so fantastic that he never went to faculty meetings or served on any committees and still got tenure and a professorship.
He’s probably one in a million. In which case he’s not the norm, he’s an outlier. And yet you seem to be saying we should all be like him. That is never going to happen because in crucial ways we are not like him and never will be. It’s a completely unrealistic goal. Just as awakening in the traditional sense is a completely unrealistic goal for 99% Buddhists in a million. Even amongst that small percentage for whom it is a realistic goal seldom achieve it.
I think you also play down the vital role played by plodders in science who steadily build up observations for the paradigm changers to work with (keeping in mind that Feynman was a theoretical not an experimental physicist). For every Feynman there are thousands of people toiling away on minor observations that look unimportant both before and after the paradigm shift, but without whom the paradigm would not have shifted.
What you seem to be saying is fuck the orchestra, all music has to be free jazz - in which everyone is a composer and everyone is a soloist. Which is OK if you are one of those people to whom that sort of noise appeals. But if you want to hear a symphony then you need 20 violin players who mainly scrape out the same old thing time and again, and can do so without getting so bored that they fall asleep. This does not take genius or the skill to compose a symphony. It is what it is. The middle ranks of musicians make symphonies possible. For most of them the goal is association with a brilliant conductor, soloist, or composer, not to be that person.
We have one model of what it looks like when there is a push for everyone to be a genius. 20th art. Which was just awful a lot of the time. If everyone has to found their own movement and create an entirely new style, then change becomes entirely arbitrary and art becomes meaningless. It’s all just sound and fury signifying nothing.
It seems to me that you’ve inadvertently Romanticised Feynman, and presented an unrealistic picture of how science works in practice. It relies a lot less on genius and a lot more on persistence than you make out.
What’s worse, it’s becoming clear that in Buddhism, meditators are extremely prone to mixing up epistemology and ontology. Over emphasising the intellectual contribution of those who meditate a lot has produced a kind of cargo-cult Buddhism, in which people regularly confuse experience and reality.
Niches
OK, and as a more substantial comment…
I liked the last sentence of that ‘scenius’ link: “Let it remain inefficient, wasteful, edgy, marginal, in the basement, downtown, in the ‘burbs, in the hotel ballroom, on the fringes, out back, in Camp 4.”
That ‘inefficient, wasteful, edgy, marginal’ thing is key, I think. Much of academia is obsessed with hill-climbing, competing to reach some local maximum defined by one dubious metric or another. Anybody who plans to wander about to see if there are any more interesting mountains elsewhere is doomed in the short term, because they will fail at this competition. Wandering around always looks inefficient and wasteful in the short term, and often is, but has far more potential in the long run.
On a slightly more practical level, what I’d like to see is more niches, in the ecological sense: more acceptable ways to be ‘successful’ in academia so that everyone isn’t stuck trying to shove each other down the same boring hill in their quest for the summit. Off the top of my head I would like all these things to be valued as highly as ‘high-impact’ research: teaching, reproduction of experiments, new ways of visualising or conceptualising existing results, communication of existing concepts to people outside of your speciality, programming new tools to make research easier, digging around in the historical archives of the field for interesting lost insights…
Of course any one of these could probably still be reduced to gameable metrics, but at least we’d have more niches to hide out in!
Normal Science
Hi David
A very gracious response to criticism as always. Yes, I was trying to get across the value and importance of routine science and also, now that I think of it, the value of routine Buddhism. A lot of Buddhists cannot rationally expect to become awakened (even when insight is a common as it now is) and yet value the social and ethical aspects of being a member of a religious group. They don’t deserve to be criticised for that.
Distinguishing routine science from cargo-cult science is relatively easy. Distinguishing routine Buddhism from cargo-cult Buddhism is more difficult I think.
I’m not sure that the crisis in psychology is that easy to diagnose. It looks like science in action to me. One can only ever disprove a hypothesis. A number have been disproved. The delay was regrettable, but there were delays in the early days of other sciences also.
Which historians?
Replication crisis
I agree that the replication crisis is more than just a normal scientific self-correction. For one thing, it is about widespread methods problems, rather than refining or replacing a particular theory. But I don’t agree that “they should have known” based on statistics education. First, I came away from three semesters of statistics with a vague impression that multiple comparisons correction and conditional stopping were “picky” theoretical considerations. There was just no discussion of how wrong you can be if you break “the rules”.
The more subtle and fascinating part of the replication crisis is how fields evolved standards of practice that seem “designed” to satisfy the institutional incentives for maximum publishable results per unit effort. This is evolution in action.
Some people like Meehl in psychology noticed the problem decades ago, but everyone was too busy writing papers to pay attention to these party poopers.
More broadly, it isn’t obvious what the right level of rigor and the strictness standards for evidence should be. The skeptical position is always strong. Formal statistical procedures for confident knowledge are only about 100 years old. Somehow we got along before then, and succeeded in socially constructing functional systems. Social psychology is an example of just such a functional system; it just wasn’t optimizing what we thought it was.
Rob
Replication crisis
To put it another way, the current replication crisis narrative is level 5, comparing systems. To say it is just about bad methods is level 4. The level 5 understanding is that the system structure and incentives must change.
Historians
Citing when you can’t remember the author or title is bad method. Yes? If you’re saying that its a fact you ought to be able to back it up. I’m not familiar with the claim you make, though it is not central to my own work, so I might have missed it. Still…
Getting things completely wrong used to be a lot more common than it is today. Which is a testament to progress.
I agree that the problem goes deeper. But my story about what has gone wrong is different - I’ve tried to make this comment stand-alone, but it may rely on terminology I’ve been developing this year on my blog. It’s hard to get across my point without the jargon.
The field of psychology tried to move from non-science to science by ostensibly adopting scientific methods. The trouble is that at this level of the science stack, the methods of reductive materialism are very difficult to apply or simply don’t apply. That difficulty was never really acknowledged or accounted for. Methods for studying systems, or even the acknowledgement that human beings are social animals, seemed to get lost in the rush to find results that would make the physicists pay attention.
This is really a failure of metaphysical reductionism, because it did not allow those working at the upper levels of the stack the proper freedom to chose methods appropriate to their level. Reductionists who dominate science only recognise the lowest levels as real, thus psychology had try too hard to make their results fit that skewed paradigm. Psychology was never going to be very amenable to analysis because systems dominate at that level of the stack. The havoc already wrecked by the Romantics following Freud meant that psychology had been so decontextualised from its social milieu that once reductionists got involved it was always going to be a disaster.
Many biologists are antireductive in outlook, because once you take your organism apart it ceases do anything interesting. Even so the gold standard for knowledge and method is reductionism, even when it is entirely inappropriate to the scale and level.
Since you are right about hindsight here is some foresight:
Biology is currently dominated by Neoliberal political ideology and Utilitarian philosophy that is going to have to change at some point. Compare Richard Dawkins to the late Lynn Margulis or Frans de Waal and you realise that Dawkins perpetuates some completely ludicrous myths about people and evolution.
The role of symbiosis, hybridisation, cooperation, communities, and other systemic features of living systems are almost completely written out of history. Witness the extent of hybridisation amongst human species that makes a mockery of the “tree” metaphor for evolution. Similarly our nature as social animals, which requires that we operate on the basis of in-group empathy and reciprocity as all social animals do, is still being nudged out of the mainstream by the (literally) insane assumptions of Nash and game theory.
A time will come when we have a paradigm shift in biology that will expose wrong-headed methods that obscured knowledge and obstructed progress after the discovery of DNA. This will be on the same scale as the psychology debacle, but it might not get the same level of press as psych because people respect biology and psych has always been suspect.
The unfolding debacle in psychology is nothing compared to the fantasy island that is economics. In economics many of the major tenets still taught up to graduate level have already been disproved, but that has not changed anything. For example in the 1970s (!) Sonnenschein, Mantel, and Debreu proved that a demand curve need not be a simple curve with a downward slope - in practice a demand curve can be any polynomial, i.e. any wavy line (Steve Keen Debunking Economics).
The equivalent is teaching the planetary model of an atom to PhD level. And then noting that an atom is nothing like a solar system, but continuing to model it as a solar system because anything else is too hard! The failure of the vast majority of professional economists to predict either the Great Crash of 1929 and the subsequent Depression or the Great Recession of 2008 and the subsequent stagnation is an indictment that goes far deeper than what is happening in psychology. Neoclassical Economics is no better than a hoax. No amount of failure to replicate seems to have any effect on the economics hegemony.
Finally, here’s a weird thing about the replication crisis in psychology. The very same bloggers and tweeters who promote the “replication crisis” narrative, still routinely blog and tweet about one-off, non-replicated studies recently published in psych journals as if there is no crisis.
Virtue and method may be insufficient; unflattering anecdata
This helped clarify some thoughts I’ve had floating around in my mind for a while, thanks.
Around 2012, in the context of algorithmic trading, I tasked myself (with the help of a team) with finding a way to predict price movements of a certain set of securities at a certain frequency (or lower). The details of the problem aren’t that important but one critical side effect was a limited data set.
Methodological issues
I myself had no more than undergrad statistics knowledge although some members of the team had grad-level training and experience working on applied problems (most relevant being in physics).
Nevertheless, we fell into the trap of bad statistics – fooling ourselves to make methodological exceptions due to small expected effect sizes (sounds like psych, eh?), data snooping, etc
For one thing, it is about widespread methods problems, rather than refining or replacing a particular theory. But I don't agree that "they should have known" based on statistics education. First, I came away from three semesters of statistics with a vague impression that multiple comparisons correction and conditional stopping were "picky" theoretical considerations. There was just no discussion of how wrong you can be if you break "the rules".
My guess is more time spent learning to not break the rules would have not helped…
Epistemic virtue issues
If we fooled ourselves into seeing something that wasn’t there, we would waste money doing further research on it, and then directly lose money by trying to make real-world, real money predictions with it. This wasn’t enough! In this case, I think it was the mindset that failing to find something wasn’t an option because of the plan I had created (for various reasons) to find a way along this path.
Skin in the game isn’t sufficient if you are hamstrung to play the game anyway. You need a way to surrender a path and look for alternatives
In economics many of the major tenets still taught up to graduate level have already been disproved, but that has not changed anything. For example in the 1970s (!) Sonnenschein, Mantel, and Debreu proved that a demand curve need not be a simple curve with a downward slope - in practice a demand curve can be any polynomial, i.e. any wavy line (Steve Keen Debunking Economics).
They are forced to play the game – the path cannot be abandoned until an alternative exists. But it will be tough to find until the path is abandoned…
Mentorship/community issues
I think this could have saved my endeavor. Unfortunately I was in a position where there was no one around me to learn how to do the science in this particular domain, and there was team-wide skepticism of how much knowledge on how to do the science was transferable cross-domain. In hindsight, the answer is much more than we thought.
This is both discouraging and encouraging for the future of social sciences. The discouraging aspect is there may be very few, or no one, left in them that is capable of doing the science properly. From whom will they be mentored and how will the community guide itself, then? On top of this, you have the problem of cultural momentum. The encouraging aspect is the surprising amount of transferability of science-skill between fields. Maybe some physicists will be able to move into psych and revitalize it? This leaves out the issues of applying reductive positivism to “higher domain” fields raised by Jayarava, however.
This is really a failure of metaphysical reductionism, because it did not allow those working at the upper levels of the stack the proper freedom to chose methods appropriate to their level. Reductionists who dominate science only recognise the lowest levels as real, thus psychology had try too hard to make their results fit that skewed paradigm. Psychology was never going to be very amenable to analysis because systems dominate at that level of the stack.
Do you have a sense of what these more appropriate methods are? More exploratory science being labeled as such?
Looking for your keys in the light
You’ve added much analysis & detail, but this strangely reminds of something I was taught as a child: If you lose your keys, you won’t find them by staying under the street lamp. You have to go back out there in the dark, where it’s cold and scary.
(The person who taught me this was a Sunday school teacher. I’ve noticed that mentioning anything to do with Christianity gets up the backs of most scientists, but as was so rightly pointed out, there is cargo-cult science just as there is cargo-cult religion, but there are still a few people in both fields who are trying to get it right. )
Bounded internet communities, Ericsson, podcast questions
Both Kevin Kelly, and you, make the key point that productive scenes need boundaries. My thesis will be that in the "atomized era," the internet and other forces have made boundary-maintenance much more difficult—so scenes are less prevalent, to the culture's detriment.
My (Millenial) friend and (Millenial) I were speculating about this just the other day – specifically how much more exclusive and comfy online communities used to feel. Funny that the growth of the Internet has caused this boundary-removing effect even on itself. Looking forward to this thesis!
The situated learning stuff is fascinating. Are you familiar with Ericsson’s work on expertise? I wonder what the hybrid theory between situated learning and deliberate practice is. Are there contradictions?
I have some questions pertaining to your recent Imperfect Buddha podcast. Sorry if this is the wrong place.
You sounded nonchalant when saying that you don’t know how to account for people’s frequent comments that they only developed some of the aspects (cognitive, ethical, social, etc) of a particular stage, or even developed stages out of order. Why so nonchalant? Why not more suspicion for Kegan’s experimental methodology?
I have a specific (apparent) contradiction in mind about Stage 4. You’ve written that STEM people are more likely to develop Stage 4 cognitive skills. But also there’s a (reasonable) stereotype of STEM people being disdainful for (what appear to them as) arbitrary rules/systems/bureaucracy/etc. How do you square this?
Screw the rules I have better rules
@Duckland “Ugh, this is so byzantine, why can’t this bureaucracy act logically and efficiently? All we have is a slow and illegible mess of incentives with no rhyme or reason to it. They really need to optimize this. If only they had the right system.”
Sounds like stage 4 to me.
From a young researcher
When I enrolled the PhD programme It was clear to me that it’s high time someone should be able to predict or at least understand which are the key success factors for LGBT healthy relationships in adult life.
While reading your essay I could but think of all my fears as a young and unexperienced researcher. It hurt when reading “the guy who figured out the problem was ignored and never cited” and I figured sincerely one of my deepest stupid motivations for doing this research – Ego satisfaction?
Do I really care about the phenomena I am supposed to study? What is the link than with my engagement, with the time and effort I dedicate to it?
I believe I do have a lust for understanding, I do hope to figure out what is actually going on. But that seems nothing compared to what I should know. The how. The methodology, the ability to study what’s relevant, to extract from literature the important issues and to be able to create a research from a methodological point of view that is helpful and that will eventually pinpoint my questions.
Thank you for stating I could be wrong and that’s ok, it gives some sort of empowerment to a young researcher like me. I will reflect more on what you call methods of technical rationality, especially publishing negative results and abandoning the famously flawed p<0.05 significance that used to scare the curiosity in me.
And yes, the mythical “scientific method”! I thought I was here to predict and measure. And that there are clear ways for me to be able to control and predict my thesis.
So what I’m taking with me from your essay is firstly becoming a cargo cultist and go towards a meta-systematic approach while looking to master at least some of the methods.
Curiosity and progress
Tackling “cargo cult” phenomenon is something I began to ask myself since I started to teach. How can I help the students create something original? How long is it acceptable to imitate and after what stage are they expected to produce something original? What context could benefit them? This are legitimate questions in any field. I see the phenomenon of imitation as an early stage of learning process or skill aquision. I enjoyed reading your article, first because it validated some of my intuitions :) and secondly because it underlined the prerequisite for overcoming the imitation phase - to overpass the form without substance phase.
One of them was curiosity. You’ve identified curiosity as a source of honesty: “Honesty comes out of curiosity, mostly, I think. If you really do want to know, there’s much less motivation to promote a wrong answer—arrived at either through deliberate fraud or sloppy, inadequately-controlled experimentation.”
As an european working in the Midle East as an aid worker I had many friends (nationals) who were looking for ways to migrate to Europe, Australia or north America for a better life. I was wondering why aren’t they interested to improve things in their own country? What coudl change their minds? Reflecting more upon these issues, I understood that among different options to understand the phenomenon, going back to school is the best step towards finding some good leads. I ended up looking at the entire phenomenon of migration. Curiosity was the nudge that led me towards pursuing a phd in the field of migration.
Your analysis about legitimate peripheral participation helepd me better understand why some people trive and some are failing. I also had in mind something related to imprinting. Who are the ones that have a major impact on you in an early stage as a junior researcher? How much is the context influencing one’s pursue of inovation and how much are other factors influencing its success?
Staying quiet in the boat is hard
My Da was a friend of Feynman and I grew up w/ the term “Cargo culting it”. I am sorry we missed Feyman’s damn subdural. Did you also grow up w/ the term? I wonder how life would have been less not knowing my motives –always understanding my own brain put Cargo culting in everything you do personally, as close as humor, as friendly, as inescapable, as oneself. I was far more fey –and will never be the scientistt –that my wise physicist Da was; everything’s magic for me–sealed off , secret. I grew up thinking he built roads through mountains but he “tunneled through solids”. He taught me quantum and SqUIDS but hey became something else in my head. My life is clapping hands and things appear–that is what I understood & agreed. Even though my Da was the poet too. I am just missing the right brain. When younger, I even saw myself waving the planes down as a metaphor though I could not fold the maps in the acr as we drove. Do you fold maps anymore I wonder. I don’t but I
miss handing them over hopelessly to my da and laughing. he taught me calclus, solving the area under a graph –all while we drove across Illinois once. He just passed. Feynman was a scamp but I liked that term. We have dangerous brains, so trecherous is what they meant. I work in cargo cult medicine. You don’t even know and suddenly you realize people have wandered in a path of false efficent fakery–either way too. Both directions are yawning caverns that yourself and your patients want you to bullshit in. hence the title. You cannot lie. Sure take this fishoil. So then I spend 23 hours deeply researching it and I have just started and no one really wants that 230 page thesis. Like handing over a basket of garbage. “Want to read this? ” Well, and I have 200 more data bits to go and really it’s trivial compared with–hey where did you go?
And, meanhwile I have gone backwards asking–oh! the real question might be–no, back that up–
Still growing up that way–grasping–what is the question before the question and simple, but not too simple. Then running after someone–I FORGOT TO ASK!?_I just need to know WHY YOU WANT TO KNOW!! This is fun, isn’t IT!!? Hello!?
No skill but also no fear of asking–what is the question, what is the result we seek? When you ask–what are you really asking–treachery often lies–and how can we ask–how should we move from what we ask? I went into medicine after philosophy but everywhere I found the world of cargo cult medicine. Slide one way and let’s play a game of “nutritional” magic. OK, I raised my kids on an organic dairy farm. I don’t mind tossing chance one way for their safety–I mind very much pretending I know any larger “truth” in that direction. In 40 years of medicine, I have learned this: many become addicted to one version of a story–the more they want to tell it, the more they get hooked. I fear, there is a very deeply cognitive bias–fair enough–but no one wants to pull their’s out even as I pull out my motives–it lurks, it taints, it makes me want to chase preventative care for people & make pretty OCD charts, I adore SNP’s, DNA but –that’s fun and engaging not intelligent. If I wanted to do the most good maybe I should argue w/ those who misapprehend the 2nd amendment, go paper Earl Warren’s opinion upon tall graffiti. That would be more honest in a way. But I don’t like arguing.
Sorry, for someone a scientist background, I speak in strange Irish blather. Everyone moves too quickly to the wrong question, it seems–like it will bite. -Cargo Cult –we run to it, the very ones who are left to teach about it. It wasn’t those w/ the husks on the ears, that was my Da’s lasting lesson. Caro culting–yes, develop the ability, the habit to catch yourself–the laughter to do it too. But, the real part my da and Richrad used to say was not to let it be a bully a whip–an institution. We all cargo Cult, we do it in our own heads and it’s in the way I tell myself adorable stupidity but also how I forget I am slipping past known data into–oh just skipped a step there–I can’t do that and be true. One head is dangerous, a second head can be help or danger–it s in how brave both are in virtue or in investing in falsehood. What kind of truth the tribe builds–and how can one leave that tribe if it falls down on itself. There are all kinds of mountains to climb in such endeavors–and all sorts of hidden avalanches too. I have seen people I never imagined become trapped on the ropes both ways if you will forgive stupid metaphors. I have begun to write about how to move carefully in these matters. I never understood before that people would attach such meaning to having “their” truth being enlisted as a permanent step after all examined it together–what did it matter I thought–data is data–fun to look at? But the attachment to who picked the rock up–and then if it was put down–I was naive –these 2 physicists long ago spoke of ego–damn they were right. I think there is a way around it if people move slowly enough and speak in a way we do not usually. I work in this way, a gentle logic with patients–who are much smarter than me–and pick up logic as if they born to teach it themselves. Which may be part of the key. Debate –is the worst mindset–our brains believe they have been enlisted to save our lives. “Here we are, swords drawn” that silly brain calls out– rigid with some opinion)–any one will do in a pinch–)from playground, sibling fights on. Truth, data sorting–none of it mattered–survival was that brain task. Turning that off, telling that brain do NOT be ready to fake, to cargo cult, see what you would want to see if–well, you just wanted to see what your questioner, was… the brain drops the sword, (don’t worry, some other part picks it up) and here comes Samuel Beckett, that’s the Irish –that’s me again–because this takes practice: you have to be weird, laugh, be vulnerable. Love the question–love the question more than being a goose: “Ever tried. Ever failed. No matter. Try Again. Fail again. Fail better.” You see ? The scientist–it is the only part I have of my Da, I am an empathy goose but I love curiosity and I love the question more than anything. It makes you love truth no matter what. If you do–you will never be addicted to looking good, you won’t be seduced . And, if you understand what I am saying–than you really, really know what cargo culting is all about. You laugh when you do it yourself–though I doubt you thank the television for turning on like me… but you will never sell out. Cheers. –MB Lambe
Typo in URL
There is a space in the amazon book link:
” I’m reading about the Balinese Rangda-Barong ritual so I can use existential phenomenology to figure out how to make breakfast.”
Science
RMCM: 5 … Mathematics is the manipulation of Function and Form to elucidate Cause and Effect.
RMCM: 6 … Science is the manipulation of Cause and Effect to elucidate Function and Form.
RMCM: 7 … In the cases of mathematics and science, technology is the final proof.
I'm grateful you put enough sense into your thoughts...
To remind me of awe-inspiring sensibility which occurs though overweight at by the darker side in structural discipline.. unfortunately, my conclusion is evolutionary discipline as mimeses are profoundly weighted toward on one hand inadequacy by ressentiments and a dual symmetry needed in both the casablanca shocked by gambling sense in inadequate risks of horror and backups serving the skeptic-cynic survivors and on the other hand chance by skinner vicarity.
Big statements. Lack of evidence
“Companies run on cargo cult business management; states run on cargo cult policies; schools run on cargo cult education theories (Feynman mentioned this one); mainstream modern medicine is mostly witch doctoring.”
Could you please provide some evidence of this? And something to show it’s a significant percentage?
B00HUWQUQO?
I’m sure I’ll have more to say when I finish this, but just a quick nitpick - there’s a broken link in footnote 7. (I googled this genuinely expecting B00HUWQUQO to be a real acronym for something and was kind of disappointed that it just seems to be part of the broken Amazon link. Maybe it should be a real acronym.)