You are viewing Version 2 of this post: a major revision written for the LessWrong 2018 Review. The original version published on 9th November 2018 can be viewed here.
See my change notes for major updates between V1 and V2.
Combat Culture
I went to an orthodox Jewish high school in Australia. For most of my early teenage years, I spent one to three hours each morning debating the true meaning of abstruse phrases of Talmudic Aramaic. The majority of class time was spent sitting opposite your chavrusa (study partner, but linguistically the term has the same root as the word “friend”) arguing vehemently for your interpretation of the arcane words. I didn’t think in terms of probabilities back then, but if I had, I think at any point I should have given roughly even odds to my view vs my chavrusa’s view on most occasions. Yet that didn’t really matter. Whatever your credence, you argued as hard as you could for the view that made sense in your mind, explaining why your adversary/partner/friend’s view was utterly inconsistent with reality. That was the process. Eventually, you’d reach agreement or agree to disagree (which was perfectly legitimate), and then move onto the next passage to decipher.
Later, I studied mainstream analytic philosophy at university. There wasn’t the chavrusa, pair-study format, but the culture of debate felt the same to me. Different philosophers would write long papers explaining why philosophers holding opposite views were utterly confused and mistaken for reasons one through fifty. They’d go back and forth, each arguing for their own correctness and the others’ mistakeness with great rigor. I’m still impressed with the rigor and thoroughness of especially good analytic philosophers.
I’ll describe this style as combative, or Combat Culture. You have your view, they have their view, and you each work to prove your rightness by defending your view and attacking theirs. Occasionally one side will update, but more commonly you develop or modify your view to meet the criticisms. Overall, the pool of arguments and views develops and as a group you feel like you’ve made progress.
While it’s true that you’ll often shake your head at the folly of those who disagree with you, the fact that you’re bothering to discuss with them at all implies a certain minimum of respect and recognition. You don’t write lengthy papers or books to respond to people whose intellect you have no recognition of, people you don’t regard as peers at all.
There’s an undertone of countersignalling to healthy Combat Culture. It is because recognition and respect are so strongly assumed between parties that they can be so blunt and direct with each other. If there were any ambiguity about the common knowledge of respect, you couldn’t be blunt without the risk of offending someone. That you are blunt is evidence you do respect someone. This is portrayed clearly in a passage from Daniel’s Ellsberg recent book, The Doomsday Machine: Confessions of a Nuclear War Planner (pp. 35-36):
From my academic life, I was used to being in the company of very smart people, but it was apparent from the beginning that this was as smart a bunch of men as I have ever encountered. That first impression never changed (though I was to learn, in the years ahead, the severe limitations of sheer intellect). And it was even better than that. In the middle of the first session, I ventured—though I was the youngest, assigned to taking notes, and obviously a total novice on the issues—to express an opinion. (I don’t remember what it was.) Rather than showing irritation or ignoring my comment, Herman Kahn, brilliant and enormously fat, sitting directly across the table from me, looked at me soberly and said, “You’re absolutely wrong.”
A warm glow spread throughout my body. This was the way my undergraduate fellows on the editorial board of the Harvard Crimson (mostly Jewish like Herman and me) had routinely spoken to each other: I hadn’t experienced anything like it for six years. At King’s College, Cambridge, or in the Society of Fellows, arguments didn’t take this gloves-off, take-no-prisoners form. I thought, “I’ve found a home.” [emphasis and paragraph break added]
That a senior member of the RAND group he had recently joined was willing to be completely direct in shooting down his idea didn’t cause the author to shut down in anguish and rejection, on the contrary, it made it author feel respected and included. I’ve found a home.
Nurture Culture
As I’ve experienced more of the world, I discovered that many people, perhaps even most people, strongly dislike combative discussions where they are being told that they are wrong for ten different reasons. I’m sure some readers are hitting their foreheads and thinking “duh, obvious,” yet as above, it’s not obvious if you’re used to a different culture.
While Combat Culture prioritizes directness and the free expression of ideas, in contrast, Nurture Culture prioritizes the comfort, wellbeing, and relationships of participants in a conversation. It wants everyone to feel safe, welcome, and well-regarded within a general spirit of “we’re on the same side here”.
Since disagreement, criticism, and pushback can all lead to feelings of distance between people, Nurture Culture tries to counter those potential effects with signals of goodwill and respect. Nurture Culture wants it to be clear that notwithstanding that I think your idea is wrong/stupid/confused/bad/harmful, that doesn’t mean that I think you’re stupid/bad/harmful/unwelcome/enemy, or that I don’t wish to continue to hear your ideas.
Nurture Culture makes a lot of sense in a world where criticism and disagreement are often an attack or threat– people talk at length about how their enemies and outgroups are mistaken, never about how they’re correct. Even if I have no intention to attack someone by arguing that they are wrong, I’m still providing evidence of their poor reasoning whenever I critique.
There is a simple entailment: holding a mistaken belief or making a poor argument is evidence of poor reasoning such that when I say you are wrong, I’m implying, however slightly, that your reasoning or information is poor. And each time you’re seen to be wrong, you (rightly) lose a few points in people’s estimation of you. It might be a tiny fractional loss of a point of respect and regard, but it can’t be zero.
So some fraction of the time people express disagreement because they generally believe it, but also in the wider world, a large fraction of the time people express disagreement it is as an attack [1]. A healthy epistemic Nurture Culture works to make it possible to safely have productive disagreement by showing that disagreement is safe and not intended as an attack.
To offer a concrete example of Nurture Culture and why you might want it, I wrote the following fictional account of Alex Percy:
I was three months into my new role as an analyst at Euler Associates. It was my first presentation making any recommendations of consequence. It was attended by heads of three other departments. Senior managers who managed teams of hundreds and budgets of millions. I was nervous. Sure, I had a doctorate in operations management, but that was merely classroom and book knowledge.
Ten minutes into the meeting I’d finished my core pitch. Ms. Fritz seemed mildly irritated, Mr. Wyman radiated skepticism. My heart sank. I mean, I shouldn’t have been surprised. If (if!) I keep my job long enough, I’m sure I’ll get there...
“You lost me on slide 4, but let’s talk it through. It seems you’re assuming that regional sales growth is going to plateau, which I hadn’t been assuming myself, but I could see it being true. Let’s assume it for now and can chat for a few minutes to see if your right about the flow-through effects”, these were the first words from Dr. Li.
I felt elation. Engagement! A chance to explain my reasoning! Being taken seriously! Maybe contributing to Euler Associates wouldn’t be such a painful grind.
IN A LESS NURTURING UNIVERSE
My heart sank. I mean, not really surprising. If (if!) I keep my job long enough, I’m sure I’ll get there...
“To get that result, you assume regional sales growth is going to plateau. That would be very surprising. So that must be wrong.”
Such was the terse response to my months of work. I’d triple-checked that premise– it wasn’t certain but it was reasonable. Was I supposed to argue back with someone who probably doubted I was worth their time? Did I meekly accept that I hadn’t justified this enough in my pitch and try to do better next time? My boss gave no indication when I look at him. Ach, I’m such a fool. I should have known you need to really justify the controversial points. My first big presentation and I blew it.
It might be a virtue to be tough and resilient, but it also costly to to bring that degree of emotional fortitude to bare. Someone might be pushing against a strong prior that they are unwelcome or that others are trying to make them push uphill contribute.
The first scenario describes Nurturing behavior. A busy senior manager signaling to the new person: “I’m going to give you a chance, I know you want to help.” If nothing else, I expect the consulting firm where Nurturing behavior is commonplace to be a far more pleasant place to work, and probably better for people’s health and wellbeing.
The Key Cultural Distinction: Significance of a Speech-Act
It is possible to describe to Combat and Nurture cultures by where they fall on a number of dimensions such as adversarial-collaborative, “emotional effort and concern”, filtering, etc. I discuss these dimensions in Appendix 1.
However, I believe that the key distinction between Combat and Nurture cultures is found in one primary question:
Within a culture, what isthemeaning/significance of combative speech acts?
By combative speech acts, I mean things like blatant disagreement, directly telling people they are wrong, and arguing one-sidedly against someone’s ideas with no attempt to find merit. The question is, within a culture, do these speech acts imply negative attitudes or have negative effects for the receiver? Or, conversely, are they treated like any other ordinary speech?
The defining feature of a Combat Culture is that these traditionally combative speech acts do not convey threat, attack, or disdain. Quite the opposite– when Herman Kahn says to Daniel Ellsberg, “you’re absolutely wrong”, Ellsberg interprets this as a sign of respect and inclusion. The same words said to Alex Percy by the senior managers of Euler Associates are likely to be interpreted as a painful “I’m not interested in you or what you have to say.”
And like in language in general, the significance of speech acts (combative or otherwise) needs to be coordinated between speakers and listeners. Obviously, issues arise when people from different cultures assign different meanings to the same thing in an interaction. Someone with Nurture Culture meanings associated to speech-acts can feel attacked in a Combative space. A Combatively-cultured person [2] in a Nurture culture space can make others feel attacked.
The prevailing culture often isn’t clear, and often spaces are mixed. You can imagine how that goes.
When does each culture make sense?
In the original version of this post, I mostly refrained from commenting on which culture was better. At this point, I think I can say a little more.
Combat Culture has a number of benefits that make it attractive, particularly for truth-seeking discussion: it has greater freedom of speech, no extra effort is required to phrase things politely, one doesn’t need to filter as much or dance around, and less attention is devoted to worrying about offending people.
And as Said Achmiz mentioned in a comment on the original post, “generally speaking, as people get to know each other, they tend to relax around each other, and drop down into a more “combative” orientation in discussions and debates with each other. They say what they think.”Or they outright countersignal. That’s some reason to have a generic preference for a culture where fewer speech acts parse as attacks or threats.
But Combat Culture only works when you’ve established that the standardly hostile speech acts aren’t unfriendly. It works when you’ve got a prior of friendliness, or have generally built up trust. And that means it tends to work well where people know each other well, have common interests, or some other strong factor that makes them feel aligned and on the same team. Something that creates a sense of trust and safety that you’re not being attacked even if your beliefs, works, or choices are.
Interestingly, while in any given particular exchange Combat Culture might seem to be more efficient and “pay less communicative overhead”, that’s only possible because the Combat Culture previously invested “overhead” into creating a high-safety context. It is because you’ve spent so much time with a person and signaled so much caring and connection that you’re now able to bluntly tell them that they’re absolutely wrong.
However, there’s more than one path to being able to tell someone point-blank that they’re dead wrong. Being old friends is one method, but you also get the relevant kind of trust and safety in contexts where not everyone knows each other well, like in philosophy and mathematics departments, as well as my Talmud class and Ellsberg’s group at RAND.
In those contexts, I believe the mechanics work out to produce priors such as everyone makes mistakes; mistakes aren’t a big deal; you point out mistakes and move on. Perhaps it’s because in mathematics mistakes are very legible once identified and so everyone is seen to make them, or that for the whole endeavor to work, people have to people to point them out and people get completely used to it as normal. In philosophy, everyone’s been saying that everyone else is wrong for millennia. Someone saying you’re wrong just means you’re part of the game.
Somehow, these domains have done the work required to give blunt disagreement different significance than is normal in the wider world. They’ve set up conducive priors [A2] for healthy combat.
Combat without Safety, Nurture without Caring
One of my regrets with the original Combat vs Nurture post is that some people starting describing generically aggressive and combative conversation cultures as “Combat Culture.” No! I meant more specific and better! As clarified above, I meant specifically a culture where combative speech-acts aren’t perceived as threatening by people. If people do feel threatened or attacked, you don’t have a real Combat Culture!
Others pointed out that there are non-Combative cultures that aren’t actually nurturing at all. This is entirely correct. You can avoid blatantly being aggressive while still being neutral to hostile underneath. That isn’t what I meant to call Nurture Culture.
In a way, both Combat Culture and Nurture are my attempted steelmen for two legitimate conversational cultures which are situated within a larger space of many other conversational cultures. Abram Demski attempts to do a better job of carving up the entire space.
Know thyself and others
Compared to 5-10 years ago, I’ve updated that people operating in different cultures from my own are probably not evil. To my own surprise, I’ve seen myself both be an advocate for Combat and Nurture at different points.
I continue to think the most important advice in this space is being aware of both your own cultural tendencies and those of the people around you. Even if others are wrong, you’ll probably do better by understanding them and yourself relative to them.
Conversational Cultures: Combat vs Nurture (V2)
You are viewing Version 2 of this post: a major revision written for the LessWrong 2018 Review. The original version published on 9th November 2018 can be viewed here.
See my change notes for major updates between V1 and V2.
Combat Culture
I went to an orthodox Jewish high school in Australia. For most of my early teenage years, I spent one to three hours each morning debating the true meaning of abstruse phrases of Talmudic Aramaic. The majority of class time was spent sitting opposite your chavrusa (study partner, but linguistically the term has the same root as the word “friend”) arguing vehemently for your interpretation of the arcane words. I didn’t think in terms of probabilities back then, but if I had, I think at any point I should have given roughly even odds to my view vs my chavrusa’s view on most occasions. Yet that didn’t really matter. Whatever your credence, you argued as hard as you could for the view that made sense in your mind, explaining why your adversary/partner/friend’s view was utterly inconsistent with reality. That was the process. Eventually, you’d reach agreement or agree to disagree (which was perfectly legitimate), and then move onto the next passage to decipher.
Later, I studied mainstream analytic philosophy at university. There wasn’t the chavrusa, pair-study format, but the culture of debate felt the same to me. Different philosophers would write long papers explaining why philosophers holding opposite views were utterly confused and mistaken for reasons one through fifty. They’d go back and forth, each arguing for their own correctness and the others’ mistakeness with great rigor. I’m still impressed with the rigor and thoroughness of especially good analytic philosophers.
I’ll describe this style as combative, or Combat Culture. You have your view, they have their view, and you each work to prove your rightness by defending your view and attacking theirs. Occasionally one side will update, but more commonly you develop or modify your view to meet the criticisms. Overall, the pool of arguments and views develops and as a group you feel like you’ve made progress.
While it’s true that you’ll often shake your head at the folly of those who disagree with you, the fact that you’re bothering to discuss with them at all implies a certain minimum of respect and recognition. You don’t write lengthy papers or books to respond to people whose intellect you have no recognition of, people you don’t regard as peers at all.
There’s an undertone of countersignalling to healthy Combat Culture. It is because recognition and respect are so strongly assumed between parties that they can be so blunt and direct with each other. If there were any ambiguity about the common knowledge of respect, you couldn’t be blunt without the risk of offending someone. That you are blunt is evidence you do respect someone. This is portrayed clearly in a passage from Daniel’s Ellsberg recent book, The Doomsday Machine: Confessions of a Nuclear War Planner (pp. 35-36):
That a senior member of the RAND group he had recently joined was willing to be completely direct in shooting down his idea didn’t cause the author to shut down in anguish and rejection, on the contrary, it made it author feel respected and included. I’ve found a home.
Nurture Culture
As I’ve experienced more of the world, I discovered that many people, perhaps even most people, strongly dislike combative discussions where they are being told that they are wrong for ten different reasons. I’m sure some readers are hitting their foreheads and thinking “duh, obvious,” yet as above, it’s not obvious if you’re used to a different culture.
While Combat Culture prioritizes directness and the free expression of ideas, in contrast, Nurture Culture prioritizes the comfort, wellbeing, and relationships of participants in a conversation. It wants everyone to feel safe, welcome, and well-regarded within a general spirit of “we’re on the same side here”.
Since disagreement, criticism, and pushback can all lead to feelings of distance between people, Nurture Culture tries to counter those potential effects with signals of goodwill and respect. Nurture Culture wants it to be clear that notwithstanding that I think your idea is wrong/stupid/confused/bad/harmful, that doesn’t mean that I think you’re stupid/bad/harmful/unwelcome/enemy, or that I don’t wish to continue to hear your ideas.
Nurture Culture makes a lot of sense in a world where criticism and disagreement are often an attack or threat– people talk at length about how their enemies and outgroups are mistaken, never about how they’re correct. Even if I have no intention to attack someone by arguing that they are wrong, I’m still providing evidence of their poor reasoning whenever I critique.
There is a simple entailment: holding a mistaken belief or making a poor argument is evidence of poor reasoning such that when I say you are wrong, I’m implying, however slightly, that your reasoning or information is poor. And each time you’re seen to be wrong, you (rightly) lose a few points in people’s estimation of you. It might be a tiny fractional loss of a point of respect and regard, but it can’t be zero.
So some fraction of the time people express disagreement because they generally believe it, but also in the wider world, a large fraction of the time people express disagreement it is as an attack [1]. A healthy epistemic Nurture Culture works to make it possible to safely have productive disagreement by showing that disagreement is safe and not intended as an attack.
To offer a concrete example of Nurture Culture and why you might want it, I wrote the following fictional account of Alex Percy:
IN A LESS NURTURING UNIVERSE
It might be a virtue to be tough and resilient, but it also costly to to bring that degree of emotional fortitude to bare. Someone might be pushing against a strong prior that they are unwelcome or that others are trying to make them push uphill contribute.
The first scenario describes Nurturing behavior. A busy senior manager signaling to the new person: “I’m going to give you a chance, I know you want to help.”
If nothing else, I expect the consulting firm where Nurturing behavior is commonplace to be a far more pleasant place to work, and probably better for people’s health and wellbeing.
The Key Cultural Distinction: Significance of a Speech-Act
It is possible to describe to Combat and Nurture cultures by where they fall on a number of dimensions such as adversarial-collaborative, “emotional effort and concern”, filtering, etc. I discuss these dimensions in Appendix 1.
However, I believe that the key distinction between Combat and Nurture cultures is found in one primary question:
By combative speech acts, I mean things like blatant disagreement, directly telling people they are wrong, and arguing one-sidedly against someone’s ideas with no attempt to find merit. The question is, within a culture, do these speech acts imply negative attitudes or have negative effects for the receiver? Or, conversely, are they treated like any other ordinary speech?
The defining feature of a Combat Culture is that these traditionally combative speech acts do not convey threat, attack, or disdain. Quite the opposite– when Herman Kahn says to Daniel Ellsberg, “you’re absolutely wrong”, Ellsberg interprets this as a sign of respect and inclusion. The same words said to Alex Percy by the senior managers of Euler Associates are likely to be interpreted as a painful “I’m not interested in you or what you have to say.”
And like in language in general, the significance of speech acts (combative or otherwise) needs to be coordinated between speakers and listeners. Obviously, issues arise when people from different cultures assign different meanings to the same thing in an interaction. Someone with Nurture Culture meanings associated to speech-acts can feel attacked in a Combative space. A Combatively-cultured person [2] in a Nurture culture space can make others feel attacked.
The prevailing culture often isn’t clear, and often spaces are mixed. You can imagine how that goes.
When does each culture make sense?
In the original version of this post, I mostly refrained from commenting on which culture was better. At this point, I think I can say a little more.
Combat Culture has a number of benefits that make it attractive, particularly for truth-seeking discussion: it has greater freedom of speech, no extra effort is required to phrase things politely, one doesn’t need to filter as much or dance around, and less attention is devoted to worrying about offending people.
And as Said Achmiz mentioned in a comment on the original post, “generally speaking, as people get to know each other, they tend to relax around each other, and drop down into a more “combative” orientation in discussions and debates with each other. They say what they think.” Or they outright countersignal. That’s some reason to have a generic preference for a culture where fewer speech acts parse as attacks or threats.
But Combat Culture only works when you’ve established that the standardly hostile speech acts aren’t unfriendly. It works when you’ve got a prior of friendliness, or have generally built up trust. And that means it tends to work well where people know each other well, have common interests, or some other strong factor that makes them feel aligned and on the same team. Something that creates a sense of trust and safety that you’re not being attacked even if your beliefs, works, or choices are.
Interestingly, while in any given particular exchange Combat Culture might seem to be more efficient and “pay less communicative overhead”, that’s only possible because the Combat Culture previously invested “overhead” into creating a high-safety context. It is because you’ve spent so much time with a person and signaled so much caring and connection that you’re now able to bluntly tell them that they’re absolutely wrong.
However, there’s more than one path to being able to tell someone point-blank that they’re dead wrong. Being old friends is one method, but you also get the relevant kind of trust and safety in contexts where not everyone knows each other well, like in philosophy and mathematics departments, as well as my Talmud class and Ellsberg’s group at RAND.
In those contexts, I believe the mechanics work out to produce priors such as everyone makes mistakes; mistakes aren’t a big deal; you point out mistakes and move on. Perhaps it’s because in mathematics mistakes are very legible once identified and so everyone is seen to make them, or that for the whole endeavor to work, people have to people to point them out and people get completely used to it as normal. In philosophy, everyone’s been saying that everyone else is wrong for millennia. Someone saying you’re wrong just means you’re part of the game.
Somehow, these domains have done the work required to give blunt disagreement different significance than is normal in the wider world. They’ve set up conducive priors [A2] for healthy combat.
Combat without Safety, Nurture without Caring
One of my regrets with the original Combat vs Nurture post is that some people starting describing generically aggressive and combative conversation cultures as “Combat Culture.” No! I meant more specific and better! As clarified above, I meant specifically a culture where combative speech-acts aren’t perceived as threatening by people. If people do feel threatened or attacked, you don’t have a real Combat Culture!
Others pointed out that there are non-Combative cultures that aren’t actually nurturing at all. This is entirely correct. You can avoid blatantly being aggressive while still being neutral to hostile underneath. That isn’t what I meant to call Nurture Culture.
In a way, both Combat Culture and Nurture are my attempted steelmen for two legitimate conversational cultures which are situated within a larger space of many other conversational cultures. Abram Demski attempts to do a better job of carving up the entire space.
Know thyself and others
Compared to 5-10 years ago, I’ve updated that people operating in different cultures from my own are probably not evil. To my own surprise, I’ve seen myself both be an advocate for Combat and Nurture at different points.
I continue to think the most important advice in this space is being aware of both your own cultural tendencies and those of the people around you. Even if others are wrong, you’ll probably do better by understanding them and yourself relative to them.
Further Content
Changes from V1 to V2
Footnotes
Appendices
Conversational Dimensions
Priors of Trust
How to Nurture
Advice & Further Descriptions (from V1)
Author’s Favorite Comments from V1