This 'we are better at criticism of the statements of others than creating our own' is, I think, an important piece of the puzzle regarding 'best decision making'. Another piece is how our own convictions work and why they are normally stronger than our own observations and reasonings (for evolutionary reasons, I estimate, both for the speed of the individual as the effectiveness of the tribe, it might thus be evolutionary necessary for us to automatically believe our own bullshit and to believe what close 'relatives' tell us — see https://ea.rna.nl/2022/10/24/on-the-psychology-of-architecture-and-the-architecture-of-psychology/)
This 'collaborative criticism' has been part of my setup for Enterprise/IT decision making/governance since I first set it up myself. This means that criticism is important, but it needs to happen in a collaborative setting (we manage the criticism consent-based in a group, adversarial criticism doesn't work). So, we have all forms of peer review at all sorts of levels. This is embedded in the 'political organisation' that an enterprise is, but if it works well enough (and doesn't become adversarial) it can coexist with and maybe even stabilise the political 'hacking'. But I have also seen 'political hacking' destroy the 'collaborative criticism' as it was seen as 'adversarial'.
Food for thought. Thank you for repeating it here.
Very thoughtful comment. Organizational decision-making, e.g. within a company or a school governing body, is beset by bias traps, but doesn't fall neatly into Weber's political versus scientific dichotomy that Farrell refers to. From "Politics as a vocation":
"What do we understand by politics? The concept is extremely broad and comprises any kind of independent leadership in action. One speaks of the currency policy of the banks, of the discounting policy of the Reichsbank, of the strike policy of a trade union; one may speak of the educational policy of a municipality or a township, of the policy of the president of a voluntary association, and, finally, even of the policy of a prudent wife who seeks to guide her husband. Tonight, our reflections are, of course, not based upon such a broad concept. We wish to understand by politics only the leadership, or the influencing of the leadership, of a political association, hence today, of a state."
As you point out, within modern organizations, a small community of people engages in knowledge-acquisition through scientific inquiry alongside collective decision-making on the practical application of such, and this often leads to tension. This doesn't just apply to technology start-ups, but to any organization wanting to improve under conditions of Knightian uncertainty, including e.g. educational establishments such as school and Universities (effective teaching methods are not god-given, but have to be discovered).
In my own experience I have never been involved in an organization, either in the private or public sectors, that satisfactorily resolved the tension between political and scientific modes. For example, school governors are encouraged by government policy to challenge school leaders, but leaders are political animals who typically have big egos, do not respond well to criticism, and therefore marginalize critics. Anticipating this, politically-savy governors attempt to gain influence by agreeing with the powerful, leading to many decisions that are dominated by group-think.
The modern "diversity and inclusion" movement was meant to overcome some of this, for precisely some of the reasons outlined by Farrell, but sadly D&I has become a bureaucratic tick-box nightmare leveraged by culture warriors on both sides of an argument that long ago missed the point.
You mention peer-review as a possible solution. In academia peer-review is typically anonymous precisely to try and isolate the influence of politics One problem with anonymous peer-review, however, is that there is often such a small community of scholars in highly specialised fields, that it can be tempting to start guessing who the reviewers are. Does the same problem beset, e.g. technology companies? In contrast to academic peer-review, how do you make a code-review non-adversarial and free from political factors?
No one - so far - is making negative comments about this piece - which would be in spirit of the piece, right?
One thing - Gelman doesn’t exactly quote Cowen out of context, but the full quote makes clear that Cowen thinks the risk of going negative is that it’s the easy way out - it can be an obstacle to thinking harder.
“2. Avoid criticizing other public intellectuals. In fact, avoid the negative as much as possible. However pressing a social or economic issue may be, there is almost always a positive and constructive way to reframe your potential contribution. This also will force you to keep on thinking harder, because it is easier to take apparently justified negative slaps at the wrongdoers.”
"...we are more likely to be closer to the truth when we are trying to figure out why others may be wrong, than when we are trying to figure out why we ourselves are right."
In prediction markets, when others are wrong they put their money up at a bad price. When I can figure out why they're wrong, I profit by moving the price to what I think is a better price. I don't even have to bother thinking about what I think is right.
How can you claim to be doing this with a straight face when half the universe of political opinion -- conservatives -- is barely represented in academia?
Moreover, doesn’t the fact of that underrepresentation indicate that academics have been failing to choose good interlocutors, -- graduate students -- for several generations now?
In fairness, if you define conservative as to the right of the median (whatever that would be since there are many possible continua), then it's a tautology that half are conservatives. I agree that the second needs a lot more scrutiny and depends on how you are defining the first.
The Enigma of Reason, published in 2019, is currently ranked #54,033 in Books on Amazon. It has 305 ratings. Amazon suggests "Thinking in Systems" is frequently bought with it.
Thinking Fast and Slow, published in 2013, is currently ranked #30 in Books on Amazon. It has 39,316 ratings. Amazon suggests that "The Psychology of Money" and "Read People Like a Book" are frequently bought with it.
My negative comments on this situation are, first, that Amazon really ought to recommend these books together. Second, while both books have their merits and defects, I suspect that Thinking Fast and Slow tells us the story we prefer to hear. Perhaps that is because it focuses on individual rather than collective cognition. However that may be, it does suggest limits to the power of criticism: the 3rd party listening to the debate has its own biases and interests - the marketplace of ideas is rigged.
Very good indeed.
This 'we are better at criticism of the statements of others than creating our own' is, I think, an important piece of the puzzle regarding 'best decision making'. Another piece is how our own convictions work and why they are normally stronger than our own observations and reasonings (for evolutionary reasons, I estimate, both for the speed of the individual as the effectiveness of the tribe, it might thus be evolutionary necessary for us to automatically believe our own bullshit and to believe what close 'relatives' tell us — see https://ea.rna.nl/2022/10/24/on-the-psychology-of-architecture-and-the-architecture-of-psychology/)
This 'collaborative criticism' has been part of my setup for Enterprise/IT decision making/governance since I first set it up myself. This means that criticism is important, but it needs to happen in a collaborative setting (we manage the criticism consent-based in a group, adversarial criticism doesn't work). So, we have all forms of peer review at all sorts of levels. This is embedded in the 'political organisation' that an enterprise is, but if it works well enough (and doesn't become adversarial) it can coexist with and maybe even stabilise the political 'hacking'. But I have also seen 'political hacking' destroy the 'collaborative criticism' as it was seen as 'adversarial'.
Food for thought. Thank you for repeating it here.
Very thoughtful comment. Organizational decision-making, e.g. within a company or a school governing body, is beset by bias traps, but doesn't fall neatly into Weber's political versus scientific dichotomy that Farrell refers to. From "Politics as a vocation":
"What do we understand by politics? The concept is extremely broad and comprises any kind of independent leadership in action. One speaks of the currency policy of the banks, of the discounting policy of the Reichsbank, of the strike policy of a trade union; one may speak of the educational policy of a municipality or a township, of the policy of the president of a voluntary association, and, finally, even of the policy of a prudent wife who seeks to guide her husband. Tonight, our reflections are, of course, not based upon such a broad concept. We wish to understand by politics only the leadership, or the influencing of the leadership, of a political association, hence today, of a state."
As you point out, within modern organizations, a small community of people engages in knowledge-acquisition through scientific inquiry alongside collective decision-making on the practical application of such, and this often leads to tension. This doesn't just apply to technology start-ups, but to any organization wanting to improve under conditions of Knightian uncertainty, including e.g. educational establishments such as school and Universities (effective teaching methods are not god-given, but have to be discovered).
In my own experience I have never been involved in an organization, either in the private or public sectors, that satisfactorily resolved the tension between political and scientific modes. For example, school governors are encouraged by government policy to challenge school leaders, but leaders are political animals who typically have big egos, do not respond well to criticism, and therefore marginalize critics. Anticipating this, politically-savy governors attempt to gain influence by agreeing with the powerful, leading to many decisions that are dominated by group-think.
The modern "diversity and inclusion" movement was meant to overcome some of this, for precisely some of the reasons outlined by Farrell, but sadly D&I has become a bureaucratic tick-box nightmare leveraged by culture warriors on both sides of an argument that long ago missed the point.
You mention peer-review as a possible solution. In academia peer-review is typically anonymous precisely to try and isolate the influence of politics One problem with anonymous peer-review, however, is that there is often such a small community of scholars in highly specialised fields, that it can be tempting to start guessing who the reviewers are. Does the same problem beset, e.g. technology companies? In contrast to academic peer-review, how do you make a code-review non-adversarial and free from political factors?
No one - so far - is making negative comments about this piece - which would be in spirit of the piece, right?
One thing - Gelman doesn’t exactly quote Cowen out of context, but the full quote makes clear that Cowen thinks the risk of going negative is that it’s the easy way out - it can be an obstacle to thinking harder.
“2. Avoid criticizing other public intellectuals. In fact, avoid the negative as much as possible. However pressing a social or economic issue may be, there is almost always a positive and constructive way to reframe your potential contribution. This also will force you to keep on thinking harder, because it is easier to take apparently justified negative slaps at the wrongdoers.”
https://marginalrevolution.com/marginalrevolution/2020/02/how-public-intellectuals-can-extend-their-shelf-lives.html
"...we are more likely to be closer to the truth when we are trying to figure out why others may be wrong, than when we are trying to figure out why we ourselves are right."
In prediction markets, when others are wrong they put their money up at a bad price. When I can figure out why they're wrong, I profit by moving the price to what I think is a better price. I don't even have to bother thinking about what I think is right.
How can you claim to be doing this with a straight face when half the universe of political opinion -- conservatives -- is barely represented in academia?
Moreover, doesn’t the fact of that underrepresentation indicate that academics have been failing to choose good interlocutors, -- graduate students -- for several generations now?
How can you claim with a straight face that half the universe of political opinion are conservatives?
How can you claim with a straight face that conservatives are barely represented in academia?
Neither of these claims can withstand an iota of scrutiny.
In fairness, if you define conservative as to the right of the median (whatever that would be since there are many possible continua), then it's a tautology that half are conservatives. I agree that the second needs a lot more scrutiny and depends on how you are defining the first.
The Enigma of Reason, published in 2019, is currently ranked #54,033 in Books on Amazon. It has 305 ratings. Amazon suggests "Thinking in Systems" is frequently bought with it.
Thinking Fast and Slow, published in 2013, is currently ranked #30 in Books on Amazon. It has 39,316 ratings. Amazon suggests that "The Psychology of Money" and "Read People Like a Book" are frequently bought with it.
My negative comments on this situation are, first, that Amazon really ought to recommend these books together. Second, while both books have their merits and defects, I suspect that Thinking Fast and Slow tells us the story we prefer to hear. Perhaps that is because it focuses on individual rather than collective cognition. However that may be, it does suggest limits to the power of criticism: the 3rd party listening to the debate has its own biases and interests - the marketplace of ideas is rigged.