As a practical matter, you may want a different word than 'accountability'. The reason I like the expression 'high trust' and dislike 'high accountability' is that once people start talking about accountability, what we end up with is a system whereby the usual suspects are scapegoated and
take the blame. This I know is not what you want. But the thing that erodes trust the most in my experience is discovering that people have lied to you, most especially if they did so as part of a cover-up. But when you investigate why cover-ups happen, you often find it is to hide the most minor of transgressions. People do not feel free to admit the most inconsequential of sins because they think the accountability police will unleash all sorts of hell upon them. The accountability police is disproportionately formed of people who love humiliating others and making them suffer -- it is why
they sign up for such things.
So, if you want real accountability the first thing you may need to do excuse such people from your process. This has turned out to be very difficult to implement in practice. Teaching people how to admit minor wrong doing and have it not matter very much is a hard lesson, especially if you do not wish to instead teach 'the well connected people can get away with anything' which is a different serious risk.
I suggest that all people elected or selected to a high authority position must sign and make public a Negative Confession, both at the end of their mandate and every time they are reelected or reappointed. Read https://www.worldhistory.org/The_Negative_Confession/
I have small children. I can’t really trust them ti so much. They require constant supervision which is expensive on my time and resources.
I try my best to teach them and to provide a good “incentive structure”, but they are children. They lack the mental abstraction capabilities, ability to predict the consequences of their actions, or emotional control necessary to act unsupervised or self direct without massive guardrails.
Low IQ adults are basically children. You wouldn’t trust them unsupervised. You wouldn’t believe they could accomplish complex tasks left to their own devises.
Children are a net resource sick. Either my time or the time of someone I pay is necessary to keep them in line. They produce no economic output.
Now in the case of my children this will change on day. They will grow up and contribute. But this isn’t the case for low iq adults. They will always require outsized supervision and subsidy that exceeds whatever meager tasks they can perform.
It’s pointless to talk about all of the ways in which high iq people act in an untrustworthy manner. The bottom line is that in spite of these inefficiencies, we just keep getting richer. It’s an acceptable level of rot in the nation.
Only totalitarianism and war was able hold down high iq societies in the 20th century, and nearly all of them have left that behind by now.
But low iq societies simply can’t escape their trap. No matter how many resources get poured in, now matter how much they can draw on the examples and guidance of more functional societies, they just can’t escape. They can’t even escape within successful nation states (look at Detroit).
High iq is not *sufficient* to succeed, but it’s *necessary*. And it tends to find a way a beat the most extreme circumstances.
Most people would endorse the following two observations: Culture matters. Incentives matter.
Arnold makes the case that incentives are more fundamental.
It's hard to design coherent, effective incentives, which can't be gamed. This is the principal-agent problem. Moreover, it's hard to find Archimedean points to uplift major institutions. Endogeneity everywhere.
What counts as cheating is endogenous. For example, alas, people no longer count endless deficit spending as cheating. Culture matters.
Re: "At one time, a religious community could hold its members accountable for living up to moral standards." There is a countervailing, cognitive mechanism, an instance of social epistemology, which can help high-commitment religious communities in recruitment. Dogma is inscrutable insofar as no one can check, empirically, here and now, if specific assertions about the afterlife or providence are true or false. If a person is anxious about the afterlife or about religious identity, she might take observable evidence of high commitment by others, who are members of religion X, to be indirect evidence that X might well be the one true religion. She might come to trust religion X because the community's high-commitment bespeaks trust in its dogma.
Good essay. I'd add immigration policy as a major area where this dynamic is present. Support for legal immigration is jeopardized because there is no trust that people who abuse the system will be punished.
A high trust society would not have decriminalized most shoplifing. You can't put the accountability horse in front of the trust cart. A culture where shopkeepers put their wares on the shelves for the customer to brouse and handle is the one that ensures you don't steal the goods and get away with it.
I suspect that you cannot use 'fear of being punished' as the glue that holds together a high trust society. The sort of people who do not value being virtuous for its own sake are the sort of people whom you 'cannot trust any further than you can throw them'. Certain religious communities have managed to create high-trust social groups by withdrawing from the larger world and, to the extent it is possible, only transacting with members of the community. These communities can handle significant disagreement over 'what does it mean to be virtuous' and significant sinfulness -- but they fall apart when people stop trying to be virtuous. These communities often have a very significant number of people who have joined not out of faith, or shared religious beliefs, but simply because they recognise that 'these are good people', and they want to be part of this community.
It used to be that scientific and academic professions and professional institutions worked the same way. Now, it seems, that the deconstructionists have completely undermined the virtue of being truthful. And they are out there holding people accountable -- but not to the truth, but to academic consensus, even if false. There are a lot of virtuous and would-be virtuous people out there who would really like to produce new virtuous institutions, and then compete against the ones that they left, and drive them out of business. But how to bell this particular cat?
My answer to your challenge is that prosecutors seem to have very little accountability but are highly trusted. As you said, sometimes "soft on crime" prosecutors lose elections, but there are a vanishingly small number of cases where a prosecutor was severely punished for the atrocity of depriving an innocent person of their liberty.
In some sense I think accountability is overrated. Developing some technological system that is highly resistant to many forms of cheating is far superior to developing a system that is completely open to cheating and then relying on a separate social system (likely with its own vulnerabilities) to provide accountability.
A high-accountability society is indeed what we need. But a society is only as accountable as the least rigorous of its institutions permit; and there are groups, including some racial, religious, ethnic, and/or political groups, which seek in an organized manner to take over a society's institutions in order to protect the group's loyal members from accountability when they cheat or bully non-members of that group. The "woke" are a major current example.
It seems to me that when a society begins to be taken over by these groups who undermine accountability, it becomes necessary for the remaining honest people to defend themselves against the subset of society's institutions that have been subverted -- either by segregation, to form a smaller but safe society, or to escalate the problem by undermining their own accountability to the now-corrupt institutions. This is where the Right find ourselves in today's western countries, and it is why I feel it worthwhile to push to re-legalize voluntary segregation. Because it's the best alternative left to us, the second best being secession movements.
“a high-trust society is one that has an effective process in place for detecting and punishing cheaters.”
Then would a country with pervasive surveillance, and swift punishment be by your definition high trust? Let’s imagine a current, or better, a future country like a China or a Middle Eastern theocracy only with technologically much enhanced powers to detect and punish norm violators. Perhaps those imaginary societies would be high trust by your definition, but they don’t sound like the desirable and pleasant places to live that one commonly associates with the description high trust society.
Let me suggest a somewhat opposite idea. Perhaps societies that expend great effort on rule creation, effective violation detection and punishment, expend that effort because they are in fact low trust societies. This idea is closer to my understanding of what people are getting at when they use the words ‘high trust’. I have lived in societies where it wasn’t uncommon for individuals to consider it personally foolish to abide by a rule if one could violate that rule to personal advantage and also be sure that no adverse consequences would result. Someone with a different cultural bent might decide not to violate a rule for personal gain because his morals were against it, for example: ‘that’s not the way I choose to do business’. I’ve also seen it commonly suggested that high trust societies, are characterized, not just by, the above mentioned, moral restraint, but also the punitive expectation of the same in others. Reusing the example above: ‘I wont deal with people that do business that way, even if it might cost me a bit more’.
You seem to focus on economic idea that incentives matter. So detection and punishment provide incentives leading to greater trust within society. I have always thought that the idea, high trust, focuses on the economic idea of transaction costs. So endogenous levels of trust enable interactions with lower overheads of detection and punishment. Both points are true and I would suggest that in practice lower transaction costs via trust have created better conditions for human flourishing than stronger incentives via punishment.
Regarding the crypto currency quote, Izra’s point about ‘trusted services’ is noted but I am not satisfied that he has offered a value free discussion of trust and crypto currencies. In the interest of completeness, I would observe that crypto currency services have selected trade-offs that place trust into the service as follows: (1) trust in sovereign management of a currency has been replaced by trust in management by an auditable computer algorithm. The idea is that individuals can have higher trust in the policies of the algorithm than policies driven by the passions and pressures of government functionaries. (2) Trust in institutional custodianship has been replaced by trust in personal custodianship. The idea is that personally holding ones digital property (hopefully in a hardware wallet) means one doesn’t have to trust a holding institution to remain solvent (FTX) or trust a government to abstain from managing rules in a self-interested way that appropriates private assets. (I do not associate any value judgements to trade-offs.)
Otherwise regarding "if someone tries to screw me over or steal my stuff..." then the technology to deal with that contingency is a system called law enforcement. That system protects property rights, and, as far as I know, its competency is not a function of the decentralization or centralization of the thing that issues the property. After all, there are decentralized systems that issue all kinds of property. Some issue crypto currency; some issue the variety of real-world currencies; some issue cars, some houses and whatever. And when someone tries to screw you over and take, say your car, we rationally appeal a competent legal system that we trust to enforce those property rights, rather than say General Motors.
Cancel culture represents a form of accountability. If you associate with, host, publish or have any other dalliance with someone deemed ideologically inappropriate or express views much outside of the 'correct' ones, you will be publicly vilified, potentially fired, removed from social media and other tech platforms etc.
Does this lead to high trust levels by those that share the appropriate ideology? Maybe. They could trust that people they associate with are unlikely to say anything too alarming to them, that they are unlikely to see views expressed in their chosen media outlets that go against the prevailing view, or hear such views at work. Perhaps what we are seeing with Twitter currently is a breakdown of trust and the mechanisms around it that underpinned a key part of this construct.
On: "I challenge you to point to an institution or segment of society that is accorded high trust with little accountability, or vice-versa."
Over the last half a century the teachers unions are still trusted by a large number of people while they have been very effective and preventing any measuring (aka testing) or allowing accountability of their results.
Great piece by Arnold. However, he seems to make no distinction between crypto and the exchanges through which many trade in it. For example, Bitcoin is not the same as an exchange which trades it. So far, the problems seem to be with the exchanges, rather than the crypto itself.
It may be that there is a set amount of trust in a society, and it can either be concentrated or dispersed. In some societies there is a high degree of solidarity within a family, clan, and or tribe; and almost none outside. By contrast, in the modern West, trust and solidarity is dispersed; which is why the brother of Ted Kaczynski (the Unabomber) could inform the authorities about his suspicions, something almost unthinkable in a more traditional society.
You seem to be confusing trust in people with trust in institutions. If you have person A and person B and an accountability system C, you can judge their trust in each other based on what they would do without C. Would A borrow money to B without a receipt and without witnesses? That's high trust. Accountability systems is what you put in place because people don't trust each other.
Also, what or whom is it you trust? Do you trust the outcome, D?
When people ask: do you trust the media, or the courts, or scientists, or some other institution with checks and balances, it is not the people in those institutions they think of. If they trust any of these institutions, it's because they know that there are checks and balance systems in place, which let them trust the outcome D.
I don't trust journalists, but I might trust the content in an article I read because there (presumably) is a system of fact checkers involved. And other newspapers that would love to point out faults in their competitors articles. If those systems don't exist, there is no reason to trust D.
Do you trust lawyers? Prosecutors? Judges? Maybe not, but if you put them against each other, you might still trust D (the verdict in a case, for instance).
Unfortunatly, the universities have become monocultures, which means the system is broken. So should you trust scientists? I only trust them in areas that are politically neutral, which is mostly the natural sciences (biology is in danger, though).
Like I said, if A trust B, he/she doesn't need C. Why do they trust each other? Mostly, people trust people they know, and if they don't know them, they are more likely to trust people who are similar to themselves. A wouldn't cheat B, so B probably won't cheat A. Similarity has a lot of dimensions, and one of them is ethnicity. Homogeneous societies have more trust, but it is possible to build trust among different ethnicities as well. Culture is important.
Finally, one thing that needs to be said is that there is a flip side to high trust societies. When we trust people, we can get fooled. Germany in the 30s and 40s was a high trust society. People trust that their own soldiers, people they know, won't commit crimes. But they do.
Sweden, where I live, used to be a high trust society, and in many ways it still is. If you knew things about Sweden, you would know that in reality there is little accountability. We don't think we need institutions that constrain our law enforcement agencies etc. We believe in the goodness of power. This has many perverse effects, but I won't bother you with that here. Trusting people is efficient, but it is not just efficient when we do good things but also when we murder.
As a practical matter, you may want a different word than 'accountability'. The reason I like the expression 'high trust' and dislike 'high accountability' is that once people start talking about accountability, what we end up with is a system whereby the usual suspects are scapegoated and
take the blame. This I know is not what you want. But the thing that erodes trust the most in my experience is discovering that people have lied to you, most especially if they did so as part of a cover-up. But when you investigate why cover-ups happen, you often find it is to hide the most minor of transgressions. People do not feel free to admit the most inconsequential of sins because they think the accountability police will unleash all sorts of hell upon them. The accountability police is disproportionately formed of people who love humiliating others and making them suffer -- it is why
they sign up for such things.
So, if you want real accountability the first thing you may need to do excuse such people from your process. This has turned out to be very difficult to implement in practice. Teaching people how to admit minor wrong doing and have it not matter very much is a hard lesson, especially if you do not wish to instead teach 'the well connected people can get away with anything' which is a different serious risk.
I suggest that all people elected or selected to a high authority position must sign and make public a Negative Confession, both at the end of their mandate and every time they are reelected or reappointed. Read https://www.worldhistory.org/The_Negative_Confession/
Arnold, if you like the idea, please develop it.
I have small children. I can’t really trust them ti so much. They require constant supervision which is expensive on my time and resources.
I try my best to teach them and to provide a good “incentive structure”, but they are children. They lack the mental abstraction capabilities, ability to predict the consequences of their actions, or emotional control necessary to act unsupervised or self direct without massive guardrails.
Low IQ adults are basically children. You wouldn’t trust them unsupervised. You wouldn’t believe they could accomplish complex tasks left to their own devises.
Children are a net resource sick. Either my time or the time of someone I pay is necessary to keep them in line. They produce no economic output.
Now in the case of my children this will change on day. They will grow up and contribute. But this isn’t the case for low iq adults. They will always require outsized supervision and subsidy that exceeds whatever meager tasks they can perform.
It’s pointless to talk about all of the ways in which high iq people act in an untrustworthy manner. The bottom line is that in spite of these inefficiencies, we just keep getting richer. It’s an acceptable level of rot in the nation.
Only totalitarianism and war was able hold down high iq societies in the 20th century, and nearly all of them have left that behind by now.
But low iq societies simply can’t escape their trap. No matter how many resources get poured in, now matter how much they can draw on the examples and guidance of more functional societies, they just can’t escape. They can’t even escape within successful nation states (look at Detroit).
High iq is not *sufficient* to succeed, but it’s *necessary*. And it tends to find a way a beat the most extreme circumstances.
Most people would endorse the following two observations: Culture matters. Incentives matter.
Arnold makes the case that incentives are more fundamental.
It's hard to design coherent, effective incentives, which can't be gamed. This is the principal-agent problem. Moreover, it's hard to find Archimedean points to uplift major institutions. Endogeneity everywhere.
What counts as cheating is endogenous. For example, alas, people no longer count endless deficit spending as cheating. Culture matters.
Re: "At one time, a religious community could hold its members accountable for living up to moral standards." There is a countervailing, cognitive mechanism, an instance of social epistemology, which can help high-commitment religious communities in recruitment. Dogma is inscrutable insofar as no one can check, empirically, here and now, if specific assertions about the afterlife or providence are true or false. If a person is anxious about the afterlife or about religious identity, she might take observable evidence of high commitment by others, who are members of religion X, to be indirect evidence that X might well be the one true religion. She might come to trust religion X because the community's high-commitment bespeaks trust in its dogma.
Good essay. I'd add immigration policy as a major area where this dynamic is present. Support for legal immigration is jeopardized because there is no trust that people who abuse the system will be punished.
A high trust society would not have decriminalized most shoplifing. You can't put the accountability horse in front of the trust cart. A culture where shopkeepers put their wares on the shelves for the customer to brouse and handle is the one that ensures you don't steal the goods and get away with it.
I suspect that you cannot use 'fear of being punished' as the glue that holds together a high trust society. The sort of people who do not value being virtuous for its own sake are the sort of people whom you 'cannot trust any further than you can throw them'. Certain religious communities have managed to create high-trust social groups by withdrawing from the larger world and, to the extent it is possible, only transacting with members of the community. These communities can handle significant disagreement over 'what does it mean to be virtuous' and significant sinfulness -- but they fall apart when people stop trying to be virtuous. These communities often have a very significant number of people who have joined not out of faith, or shared religious beliefs, but simply because they recognise that 'these are good people', and they want to be part of this community.
It used to be that scientific and academic professions and professional institutions worked the same way. Now, it seems, that the deconstructionists have completely undermined the virtue of being truthful. And they are out there holding people accountable -- but not to the truth, but to academic consensus, even if false. There are a lot of virtuous and would-be virtuous people out there who would really like to produce new virtuous institutions, and then compete against the ones that they left, and drive them out of business. But how to bell this particular cat?
My answer to your challenge is that prosecutors seem to have very little accountability but are highly trusted. As you said, sometimes "soft on crime" prosecutors lose elections, but there are a vanishingly small number of cases where a prosecutor was severely punished for the atrocity of depriving an innocent person of their liberty.
In some sense I think accountability is overrated. Developing some technological system that is highly resistant to many forms of cheating is far superior to developing a system that is completely open to cheating and then relying on a separate social system (likely with its own vulnerabilities) to provide accountability.
A high-accountability society is indeed what we need. But a society is only as accountable as the least rigorous of its institutions permit; and there are groups, including some racial, religious, ethnic, and/or political groups, which seek in an organized manner to take over a society's institutions in order to protect the group's loyal members from accountability when they cheat or bully non-members of that group. The "woke" are a major current example.
It seems to me that when a society begins to be taken over by these groups who undermine accountability, it becomes necessary for the remaining honest people to defend themselves against the subset of society's institutions that have been subverted -- either by segregation, to form a smaller but safe society, or to escalate the problem by undermining their own accountability to the now-corrupt institutions. This is where the Right find ourselves in today's western countries, and it is why I feel it worthwhile to push to re-legalize voluntary segregation. Because it's the best alternative left to us, the second best being secession movements.
“a high-trust society is one that has an effective process in place for detecting and punishing cheaters.”
Then would a country with pervasive surveillance, and swift punishment be by your definition high trust? Let’s imagine a current, or better, a future country like a China or a Middle Eastern theocracy only with technologically much enhanced powers to detect and punish norm violators. Perhaps those imaginary societies would be high trust by your definition, but they don’t sound like the desirable and pleasant places to live that one commonly associates with the description high trust society.
Let me suggest a somewhat opposite idea. Perhaps societies that expend great effort on rule creation, effective violation detection and punishment, expend that effort because they are in fact low trust societies. This idea is closer to my understanding of what people are getting at when they use the words ‘high trust’. I have lived in societies where it wasn’t uncommon for individuals to consider it personally foolish to abide by a rule if one could violate that rule to personal advantage and also be sure that no adverse consequences would result. Someone with a different cultural bent might decide not to violate a rule for personal gain because his morals were against it, for example: ‘that’s not the way I choose to do business’. I’ve also seen it commonly suggested that high trust societies, are characterized, not just by, the above mentioned, moral restraint, but also the punitive expectation of the same in others. Reusing the example above: ‘I wont deal with people that do business that way, even if it might cost me a bit more’.
You seem to focus on economic idea that incentives matter. So detection and punishment provide incentives leading to greater trust within society. I have always thought that the idea, high trust, focuses on the economic idea of transaction costs. So endogenous levels of trust enable interactions with lower overheads of detection and punishment. Both points are true and I would suggest that in practice lower transaction costs via trust have created better conditions for human flourishing than stronger incentives via punishment.
Regarding the crypto currency quote, Izra’s point about ‘trusted services’ is noted but I am not satisfied that he has offered a value free discussion of trust and crypto currencies. In the interest of completeness, I would observe that crypto currency services have selected trade-offs that place trust into the service as follows: (1) trust in sovereign management of a currency has been replaced by trust in management by an auditable computer algorithm. The idea is that individuals can have higher trust in the policies of the algorithm than policies driven by the passions and pressures of government functionaries. (2) Trust in institutional custodianship has been replaced by trust in personal custodianship. The idea is that personally holding ones digital property (hopefully in a hardware wallet) means one doesn’t have to trust a holding institution to remain solvent (FTX) or trust a government to abstain from managing rules in a self-interested way that appropriates private assets. (I do not associate any value judgements to trade-offs.)
Otherwise regarding "if someone tries to screw me over or steal my stuff..." then the technology to deal with that contingency is a system called law enforcement. That system protects property rights, and, as far as I know, its competency is not a function of the decentralization or centralization of the thing that issues the property. After all, there are decentralized systems that issue all kinds of property. Some issue crypto currency; some issue the variety of real-world currencies; some issue cars, some houses and whatever. And when someone tries to screw you over and take, say your car, we rationally appeal a competent legal system that we trust to enforce those property rights, rather than say General Motors.
Cancel culture represents a form of accountability. If you associate with, host, publish or have any other dalliance with someone deemed ideologically inappropriate or express views much outside of the 'correct' ones, you will be publicly vilified, potentially fired, removed from social media and other tech platforms etc.
Does this lead to high trust levels by those that share the appropriate ideology? Maybe. They could trust that people they associate with are unlikely to say anything too alarming to them, that they are unlikely to see views expressed in their chosen media outlets that go against the prevailing view, or hear such views at work. Perhaps what we are seeing with Twitter currently is a breakdown of trust and the mechanisms around it that underpinned a key part of this construct.
On: "I challenge you to point to an institution or segment of society that is accorded high trust with little accountability, or vice-versa."
Over the last half a century the teachers unions are still trusted by a large number of people while they have been very effective and preventing any measuring (aka testing) or allowing accountability of their results.
You might not like the phrase, but you seem to understand that we are losing what it represents.
Great piece by Arnold. However, he seems to make no distinction between crypto and the exchanges through which many trade in it. For example, Bitcoin is not the same as an exchange which trades it. So far, the problems seem to be with the exchanges, rather than the crypto itself.
It may be that there is a set amount of trust in a society, and it can either be concentrated or dispersed. In some societies there is a high degree of solidarity within a family, clan, and or tribe; and almost none outside. By contrast, in the modern West, trust and solidarity is dispersed; which is why the brother of Ted Kaczynski (the Unabomber) could inform the authorities about his suspicions, something almost unthinkable in a more traditional society.
You seem to be confusing trust in people with trust in institutions. If you have person A and person B and an accountability system C, you can judge their trust in each other based on what they would do without C. Would A borrow money to B without a receipt and without witnesses? That's high trust. Accountability systems is what you put in place because people don't trust each other.
Also, what or whom is it you trust? Do you trust the outcome, D?
When people ask: do you trust the media, or the courts, or scientists, or some other institution with checks and balances, it is not the people in those institutions they think of. If they trust any of these institutions, it's because they know that there are checks and balance systems in place, which let them trust the outcome D.
I don't trust journalists, but I might trust the content in an article I read because there (presumably) is a system of fact checkers involved. And other newspapers that would love to point out faults in their competitors articles. If those systems don't exist, there is no reason to trust D.
Do you trust lawyers? Prosecutors? Judges? Maybe not, but if you put them against each other, you might still trust D (the verdict in a case, for instance).
Unfortunatly, the universities have become monocultures, which means the system is broken. So should you trust scientists? I only trust them in areas that are politically neutral, which is mostly the natural sciences (biology is in danger, though).
Like I said, if A trust B, he/she doesn't need C. Why do they trust each other? Mostly, people trust people they know, and if they don't know them, they are more likely to trust people who are similar to themselves. A wouldn't cheat B, so B probably won't cheat A. Similarity has a lot of dimensions, and one of them is ethnicity. Homogeneous societies have more trust, but it is possible to build trust among different ethnicities as well. Culture is important.
Finally, one thing that needs to be said is that there is a flip side to high trust societies. When we trust people, we can get fooled. Germany in the 30s and 40s was a high trust society. People trust that their own soldiers, people they know, won't commit crimes. But they do.
Sweden, where I live, used to be a high trust society, and in many ways it still is. If you knew things about Sweden, you would know that in reality there is little accountability. We don't think we need institutions that constrain our law enforcement agencies etc. We believe in the goodness of power. This has many perverse effects, but I won't bother you with that here. Trusting people is efficient, but it is not just efficient when we do good things but also when we murder.