Disagreeing with Paul Graham
rothbrothers.net
rothbrothers.net
It can be done, but you have to make it a deliberate project and put some time and thought into it, rather than just stumbling on a bunch of programmers in a cafe.
The story about the programmers in the cafe was not evidence for any claim. It was just an explanation of what set me thinking about the problem. It's not a pillar holding up anything, so attacking it proves nothing.
(Incidentally, a moment's thought would have made it clear that I have in fact "put some time and thought into it." I've spent 24 years as a professional programmer, during which I've observed thousands of programmers working for all sizes of companies. Plus I've seen first hand the transformation undergone by roughly 200 YC-funded founders so far.)
The second half is fallacy:
But rather than pointing out fallacies, a better way to refute Graham's evolutionary argument is by reductio ad absurdum. ... Our ancestors lived in a world that was shrouded in darkness half the day, therefore we would be happier without electric lights.
This supposed refutation can itself be disproven by reductio ad absurdum, because it could be applied to any explanation of our inclinations based on how we evolved.
(Bad example, incidentally. There do seem to be conflicts between electric lights and human biology: http://www.washingtonpost.com/wp-dyn/content/article/2008/02...)
Neither of these arguments even attempts to refute the central point of the essay. In case anyone wants to try it, the central point is that in an organization organized as a tree structure, structural forces tend to give each person freedom in inverse proportion to the size of the whole tree. That we work better in groups of 10 than 100 I feel is obvious enough not to need justification. The argument from evolution is just an attempt to explain why groups that size work.
Though on the whole my reaction is "I want my 20 minutes back," there was one encouraging thing about this experience. Even these dishonest DH5s are more civil than Atwood DH2ing about my choice of metaphors.
He clearly states he thought your central point was: "It will always suck to work for large organizations, and the larger the organization, the more it will suck," because "humans weren't meant to work in such large groups."
How is this so different than your stated thesis that: "the central point is that in an organization organized as a tree structure, structural forces tend to give each person freedom in inverse proportion to the size of the whole tree. That we work better in groups of 10 than 100 I feel is obvious enough not to need justification?"
His refutation relies on examples of humans partaking in behavior that they were not originally "evolved" to do. It is a direct response to your reliance on evolution. If you thought the whole point of your essay was "obvious enough not to need justification," then why did you write a whole essay on the subject?
To address your central point, I would cite the space program. Humans were obviously not evolved to be space-faring creatures, and yet hundreds (thousands?) of human intellects in concert (in a tree-structured organization) managed to land a man on the moon. The idea of synergy, where many humans in an organizational structure can accomplish more than any single human's evolved capabilities, has long been recognized (http://www.complexsystems.org/magic.html). Many people that work in such organizations draw personal satisfaction from the fact that they are furthering what they consider a greater good, rather than their own personal desires. Where a start-up glorifies individuals, organizations tend to glorify the whole. The fact that this is not satisfying to some does not mean it is not satisfying to others, and certainly does not imply that “You were not meant to have a boss.”
And you're right, he does state something close to the central point of my essay. But if you keep reading, that's not the part he tries to refute.
Your last paragraph is actually the most thoughtful response I've read about this whole controversy. But I think you're wrong to say that startups glorify the individual. The startup founders I know are on average more idealistic than people working for big companies. They tend to be the ones who want to make the world a better place; though there are of course exceptions, the people working at big companies are the ones who tend to think of their job as something they do mainly for the money.
"Making the world a better place (and getting really damn rich)."
The justification for this attitude is that one can supposedly participate in a positive-sum game for both parties -- one party is you, the other is the world. This is the altruistic lure of market structures. Whether or not you agree with that non-zero-sum stance (do markets actually tend to produce such situations, all externalities considered?), I think the "making a lot of money indicates I'm filling a truly needed demand" mentality produces entrepreneurs who unnecessarily tie their personal increase in wealth to the feeling of an increase in some greater good.
But we do get pretty cool products out of this, whatever the psychology involved.
Well it is his site, deal with it. In the meantime please stop polluting my damn RSS feeds. Please consider the comments feature a side effect of this site's PRIMARY FUNCTION and not worthy of news itself. Thanks!! (GRRRRR)
Of course, persons can sacrifice for the greater good or a greater civilizational vision. This produces societies, not tribes. Your space program example serves to illustrate what is possible in the context of such a vision.
Your argument doesn't do what you think. Just because every evolutionary psychology argument can be attacked this way doesn't make it a bad attack. The logic in your essay is far from being able to pass scientific scrutiny. Maybe you don't care.... This IS a fundamental problem in evolutionary psychology. Why do we engage in some weird behavior? Because it was adaptive. How do we know it was adaptive? Well, we wouldn't engage in it otherwise.
The argument itself I need to think more about.
a. I tend not to be convinced by arguments that invoke nature. It's a heuristic at best. Like you said it didn't bother me since it never claimed to be more than circumstantial evidence.
b. Your 'central' point was certainly the most interesting, but it wasn't clear it was central. In fact, I can't see how your style of essay can admit a central anything. Which is a strength, IMO.
In general I don't care as much about accuracy as I do about interestingness. I found myself recently looking back on notes from your essays of 3 years ago. I was quoting 1 in 3 sentences, sometimes 1 in 2. And these were long essays! It's safe to say that's not true anymore.
This supposed refutation can itself be disproven by reductio ad absurdum, because it could be applied to any explanation of our inclinations based on how we evolved.
No, that's overbroad. It could be applied to any argument that takes the form, "Our ancestors evolved to do x, therefore we will be happier if we do x." And, indeed, all such arguments are false, for the reason stated in my post.
Neither of these arguments even attempts to refute the central point of the essay.
I believed I had identified the central point of your argument. If you want to designate a different aspect the main point, that's your prerogative, but nothing in the essay prioritizes one over the other, and in fact the title refers to the evolutionary argument rather than the "central point." I don't have any thoughts about the "tree-structure/inverse freedom" point.
I'm not going to respond to your accusation of dishonesty, except to ask what DH number you'd assign it.
Instead of parsing the perceived inconsistencies in the words, how about looking at the spirit of what he's trying to do.
Look at the default use of people's minds who have been trained to solve difficult problems are being put to today: In bureauacracies on menial things that won't mean much of a difference in the world.
What is the problem with someone saying: "This is wrong, and I'm going to say it. I'm going to push the meme out there that this is bullshit and the default should be to maximize your return on your talent and work, and have a positive impact on society as a result."
If a metaphor or phrasing or analogy is misconstrued, who cares? That's too subjective anyway. If the default spirit was imbued in more technical people than not, I'm guessing we'd be in a better place.
[descend from soapbox]
Already done: http://news.ycombinator.com/item?id=142597
If I hadn't already done it, I wouldn't bother with it now, though, because:
Even these dishonest DH5s are ...
Roth's contribution is CLEARLY NOT DISHONEST (couldn't resist the uppercase here). I get the impression that the chance you would ever take advantage of any refutations to plug holes and thus further improve the standard of your essay collection is precisely zero. It's all about "picking a winner" (this term actually appears in paulgraham.com/disagree.html) instead of advancing towards the truth. Readers of your essays will do best by just enjoying the incredible density of ideas, then deciding for themselves which ones to accept, rather than following the "comment" link at the bottom.
If anyone has a link to an instance where Paul Graham helps the search for the truth by conceding a nontrivial point in an argument, I would be grateful. That would help repair my impression, and probably some other people's as well.
Paul is not saying that there's some kind of moral or ethical imperative to working in small groups. He just observes that it works well. So there is no fallacy, though I can see how Roth could have gotten this mixed up.
Second, it is just silly to try to refute the notion that we might be good at and enjoy anything we might have evolved to do. For one thing, that's clearly the case -- evolution works well. For another thing, PG's not trying to prove anything, he's relating his personal experience and forwarding a theory that might explain it.
The fact that people were so bothered by this is the most telling fact. As Abbie Hoffman used to say, if someone ever says something that gets your goat, they struck gold!
Example: the honest, well-intentioned pastor who delivers a sermon on creationism. I witnessed one of these last year -- I was stunned by the intellectual dishonesty as he made a false statements about what "science" and "scientists" say, but understand that he's a loving man who cares deeply and authentically for his congregation.
The fact that he believes what he's saying doesn't change the dishonesty of his arguments.
The pastor in my example misled people with wrong arguments, but it was unintentional. While this type of action can lead to horrible results, it's hard to call the pastor "dishonest" if he didn't intend to deceive. Perhaps a better descriptor would be "fool" or "ignoramus".
For instance, if you titled the essay 'the inverse relationship between personal freedom and organization size', I don't think it would elicit such an outcry (or any attention whatsoever). Instead, you state that the loss of freedom makes going into large organizations a bad choice for certain groups of people (namely, programmers, because programming is about freedom).
The programming community is made up of many people with different backgrounds, mindsets, and interests. While I myself may program for the freedom to create things, I know other people who program for the fun of it, and others who program because its what they know how to do. All of these types of programmers exist in the startup community, because a startup can mean much more than just programming freedom to them (it can mean money, prestige, adventure, a challenge, etc). If I am disagreeing at all, I would only do so concerning the scope of people that 'programming freedom' applies to. I think only a small subset of the startup community really thinks this point is meaningful - the rest do not give it much weight.
I think the main point of the essay is to convince people to startup early in their careers, even out of college. (If 'you weren't meant to have a boss', what are you meant to do?) You back it up by saying that corporate culture, from your experience, does more harm than good from a startup's perspective (especially in the matter of programming), and you explore that line until you come up with a good model that explains why corporate culture is harmful in its essence.
In contrast, many people don't think that programming freedom is a powerful incentive to starting up early, and are disagreeing because that is the only evidence that you provide to back up your claim that 'you weren't meant to have a boss'.
Perhaps Pg should go mano-a-mano with Zed Shaw.
You're taking inputs and synthesizing them in the context of intellectual, emotional, biological contexts all in parallel. You're artfully communicating these intuitions to other people using all of those contexts. The ideas and messages are very complex and subtle because they involve life and human nature. It is not posible to reduce these intuitions down to a purely rational argument.
This "output" then is taken as input by other people (each hugely rich and complex). They process it. Some approximation of your original intuitions form in their minds. Some of these people will "get it". Others will not.
Trying to respond to the people who didn't understand what you meant using purely intellectual tools will not necessarily yield results. Because it's highly likely that nonintellectual aspects of what you're saying and how you're saying it are the causes of interference that are preventing the concepts from getting across.
In that sense what you wrote is not perfect. But I don't think it can be perfect. You need to choose one set of tradeoffs over another. You're choosing to be bold and decisive rather than measured and cautious.
There are plenty of deeply important true statements that can't be proven true. How much time and effort you want to put into trying to convince every last person of your ideas is your own business. There are a lot of people who do understand and appreciate what you're thinking.
I agree, but I think you're missing a bigger point than your essay's thesis.
In your definition of DH2, you say:
"So if the worst thing you can say about something is to criticize its tone, you're not saying much."
"It matters much more whether the author is wrong or right than what his tone is."
You keep trying to redefine all criticism as refutations of the central point, even though many of those criticisms didn't object and even agreed with your main assertion.
Which incidentally means that all such arguments are fundamentally flawed, the classic is/ought dilemma.
Pg's argument moves from 'is' to 'ought' without explanation or justification.
So perhaps you are right about humans, but wrong about freedom.
One problem with direct observation is that it's hard to get a representative sample. It can be done, but you have to make it a deliberate project and put some time and thought into it, rather than just stumbling on a bunch of programmers in a café.
False dichotomy. PG's sample is neither a deliberate project, nor the people in the cafe. It's the programmers he has known and observed over his life. There's no reason to doubt representativeness of his sample.
a case where an ad hominem argument is relevant and valid.
There's never a case where ad hominem is valid. If it's valid, it's not ad hominem, by definition. A logical fallacy can't be valid. But I'm nitpicking here. His point is:
The fact that this observation confirms his preexisting belief in the worthiness of startups makes it less credible than if the same observation were made by someone who had no particular interest in startups.
Occam's Razor suggests that PG works with startups because he thinks they're better, not that he thinks they're better because he works with them. He is rich enough to work on whatever he wants.
A creature that's perpetually dissatisfied, always striving for advantage, wins out over a creature that's happy.
Then why does happiness exist? Why hasn't it been eradicated millions of years ago? Answer: because it serves a purpose. It's how evolution tells the organism, "keep doing what you're doing". Not everything that mentions nature is an appeal to nature.
So the sample of programmers that he has known and observed over his life is unquestionably representative?
>There's never a case where ad hominem is valid.
Are you saying that biases and potential biases of people have absolutely no relevance in any "valid" discussion, even in response to subjective observations about why people act the way they do?
>Occam's Razor suggests that PG works with startups because he thinks they're better, not that he thinks they're better because he works with them. He is rich enough to work on whatever he wants.
Are you suggesting that this makes PG completely free from bias?
>It's how evolution tells the organism, "keep doing what you're doing".
So happiness is just a reward we get from pursuing those things that will result in the propagation of our genes?
>Not everything that mentions nature is an appeal to nature.
Sure, but that's irrelevant. Gabe says PG's essay is an appeal to nature and backs it up. If you disagree, then please do the same.
So the sample of programmers that he has known and observed over his life is unquestionably representative?"
there's no reason to doubt != it is unquestionable that
As for the last one, I did back it up. I refuted a central premise of the author, that happiness is evolutionarily useless.
They have absolutely NO relevance to the argument. Ad hominem is an error in logic. Bias has nothing to do with the argument. It is the argument we refute, not the person. Unfortunately, you
It is utterly amazing to me people still argue about these errors in logic. One can only assume it is defensiveness.
You might want to look up "straw man" as well. It's another logical fallacy, and you're using it.
Evolution, at its core, strives for only one thing: propagation of the species. Taken to this level, it's a wonder that everyone isn't spending their time either at drunken orgies, or fighting each other to get into the drunken orgies. Somehow, "civilization", "culture", and "society" have guided human development away from that model - and I don't presume to know how it happened.
Taken to the extreme, why have we even developed arms, legs, eyes, ears, and minds? Simple bacterial reproduction does just fine. In fact, there is so much bacteria in our gut and sinuses that 90% of the DNA molecules in our bodies are foreign. One begins to wonder, then, whose purpose our human bodies really serve.
What do you mean by evolution being "easy"?
Evolution, at its core, strives for only one thing: propagation of the species.
Even if we agree to attribute intentions to evolution (rhetorically or otherwise), and to narrow the concept of evolution to adaptation through gene transmission/recombination/mutation/selection, you can just as well argue that evolution "strives for" stable ecosystems or, as you argue below, for a better habitat for our microscopic guests.
Taken to this level, it's a wonder that everyone isn't spending their time either at drunken orgies, or fighting each other to get into the drunken orgies.
Why so?
Somehow, "civilization", "culture", and "society" have guided human development away from that model - and I don't presume to know how it happened.
It doesn't surprise me that constructive competitiveness in a social setting is evolutionarily stable before shortsighted, greedy selfishness. Other less intelligent animals have developed social behaviours. The surprising thing would be that a species with more brainpower to spare wouldn't have figured that out eventually.
I don't mean to downplay culture as a factor. But I don't see a dichotomy between culture and "evolution"; I see fuzzy borders and much interplay between genetic and cultural evolution, in most of our history as a species.
The fact that culture evolves much faster than genes and is the overriding adaptation factor doesn't mean genetic influence is going away just as fast. Much on the contrary, the relative slowness of genetic evolution serves to explain why we carry such a legacy of behavioral tendencies so long after its original usefulness has faded.
Although most of us don't spend our lives in endless orgies and fights, it's easy to observe some ancestral substrate that makes many of us basically prone to promiscuity and domination (although I don't think those are the only evolutionarily stable heuristics, even in a selfish scope). The cultural rules that keep these instincts at line can't be wholly attributed to reason either; genetically driven social tendencies play an important part too.
Taken to the extreme, why have we even developed arms, legs, eyes, ears, and minds? Simple bacterial reproduction does just fine.
I don't get what you are trying to prove here. What is your explanation for the fact that we developed those things?
In fact, there is so much bacteria in our gut and sinuses that 90% of the DNA molecules in our bodies are foreign. One begins to wonder, then, whose purpose our human bodies really serve.
Again, that's a matter of interpretation. Applying the concepts of "purpose" and "serving" to these topics is like carrying our everyday physical intuitions when we are discussing relativity theory. I think it's often more a trap than a help.
With that caveat, I'd say our bodies "serve", at different levels, our own cells and mythocondria, alien microorganisms living within them, ourselves as individuals, other humans and animals for whom our net influence is positive, society as a whole, and the Earth's ecosystem.
Yes, even if we are screwing up Earth as we know it, we are "helping" it become Earth as our children will know it. I have no reason to believe Earth "prefers" either version. Whatever the outcome, it will have turned out the most evolutionarily stable one.
>> Taken to the extreme, why have we even developed arms, legs, eyes, ears, and minds? Simple bacterial reproduction does just fine.
> I don't get what you are trying to prove here. What is your explanation for the fact that we developed those things?
Back to my point, everyone assumes that evolution simply means "survival of the fittest" or the "propagation of the gene". If that's all there is to it, there's no obvious explanation for the emergence of higher-order organisms, since bacterial reproduction provides by far the most efficient mechanism for copying genes. All this is to say that the devil is in the details and there is actually quite a bit more to evolution than what is taught in 7th grade biology.
No, actually. The one thing it strives for is propagation of the genes, not the species. Read The Selfish Gene, by Richard Dawkins. It will make things clearer.
Nonsense. If that's your "statement", it's wrong. The entire field of sociobiology (read Wilson) speaks to phenomenon like "civilization", "culture", and "society" being driven solely by nucleotide replication.
Like I said, I only know enough to know that I can't comfortably refer to evolution to explain human nature. My point is that I doubt that most people who do reference evolution are familiar with the likes of sociobiology either.
Sure, until one bacterium in your neighborhood starts working out---then you better hit the gym, too. Onwards and upwards.
Huh? I think you'd have to know pg pretty darn well before being able to apply Occam's Razor to the logic of his preferences. It seems perfectly reasonable to me to think that pg works with startups because he thinks they'd be better because he works with them.
It also seems to me that might be a distinction between work and play: you work on the things you work on to make them better, you play with the things you play with because they are better.
Lastly, if you think startups are better, and that you can make them better by working with them, then it seems like the perfect combination of work and play (read: passion), and Occam's Razor isn't needed.
You know what else would be better if PG worked with it? Potholes in the street. So why doesn't PG work with potholes, or any of the infinity of other things he could be working with, and making better? The obvious answer: because he thinks startups are neat, as opposed to potholes, or big companies.
like the purpose of leadership/management/social groups it can be taken to a detrimental level. example: addiction. "keep doing what you're doing at all costs, care about nothing else, even if it kills you or otherwise destroys your life."
>A creature that's perpetually dissatisfied, always striving for advantage, wins out over a creature that's happy.
I, too, thought this statement was conspicuously unsupported by any evidence (which is part of what i thought the author disagreed with). moreover, it is in direct contradiction to the parenthetical that it preceeds (are we to take it that monks "override... unhappiness" in order to lose out?)
The difficulty with this argument is a fallacy about implication. Roth states that the "evolutionary argument" supports pg's view, but that he can then use that same "evolutionary argument" to prove an absurd point of view. Then he holds that he has disagreed with pg's central viewpoint.
This method of argumentation is fundamentally flawed because pg's view might be supported by a multitude of arguments, the truth or usefulness of the "evolutionary argument" is not a necessary condition for the truth of pg's view of organizations, bosses, and human nature. Hence, Roth has merely attacked pg's method of proof while leaving the central claim untouched. I have other criticisms of Roth's argument, but even if I am wrong in such criticism the argument would fail to disagree with pg.
"I don't understand why the audience was asked if it was 'ready to rock', as they clearly did not have instruments. Also, it was illogical to ask them to 'fight the power', since power is an abstract physical quantity that cannot meaningfully be 'fought'."
And did PG's monumental catalogue of rhetorical errors mention the rule about "not arguing with straw men"? Because I don't think the real PG actually believes that literally everyone "wasn't meant to have a boss", nor would the real PG ever endorse the concept that "you weren't meant to have a chair, because our ancestors didn't evolve to sit in chairs."
You weren't meant to have a bed. The better a bed "feels" to most people, the worse it is for the health of your back, and vice versa. Most non-serious back pain can be corrected by sleeping on a dirt floor, or a hay bale.
You also weren't meant to have shoes. Orthopedics today are mostly needed because of the shoes we wear when we're young; your feet are meant to rest flat against the ground, to be able to curve around things like rocks and tree branches, and to mush up into wet dirt or sand. I personally get back-aches whenever I wear anything other than flat-soled ("skate") shoes, approximating the natural terrain below them.
From these two things, I can guess that chairs really aren't a good idea either, but I've never looked into them specifically.
He also makes some guesses about the effects of those limitations on the happiness of employees in companies of different sizes, which I think he marks off pretty clearly as being drawn from observation. What's so incendiary about those, anyway? Regardless of whether it's true or not, isn't the stereotype of the creatively stifled employee of a large company a cultural cliche?
I'm ready to stop hearing about this now.
One difficulty with disagreeing with people is that you have to present their argument and your argument, and so your essay ends up being longer than theirs.
No, you only need to include the aspect you disagree with so your rebuttal can be shorter.
>In a general sense, this is the logical fallacy known as the "appeal to nature" -- the idea that what's natural is ipso facto good or right. There is no reason to believe this: plenty of natural things are neither good nor right.
Paul is not saying that working in groups of about 8 is not "good or right." He's saying that it's satisfying and functional. It's perfectly reasonable to say that something we were evolved to do is something we do well, and something we were not evolved to do, we have trouble with.
I could start cheating on my girlfriend pretty easily, but that would not be "good and right" to me, though it would arguably be satisfying and functional for me. Sometimes it seems satisfying and functional for me to punch people in the face. The problem with "satisfying and functional" = "good and right" is that both sides of that equation are highly complex, contextual, and subjective.
It seems like both of these possible explanations are endemic to large organizations, and would be impossible if you were a startup founder. If you start your own company and your boss is an asshole and your project is boring, you only have yourself to blame. You also have the authority to fix it immediately.
Only if you're bootstrapping. A lot of start-ups are not so lucky as to get funding and help from a company like Y-Combinator. End up with a bad VC, and you can easily find yourself feeling 'caged' again.