Noah Smith & I have made Henry Farrell unhappy. But we do believe that non-economist social scientists are greatly confusing themselves via excessive, promiscuous, and insufficiently thoughtful use...
I think Farrell makes the stronger argument. As a not to recent example, recall how Microsoft c. 1990s exreted power over the OS and applications market. Gates could just state that MSoft was going to do [X] and this then dissuaded any number of companies from even trying to build anaything. Announcing vaporware was effectively coercise power, but framed as a technology development decision. Of course in teh end this put them in the crosshairs of the antitrust law (that ultimately failed) but did trigger the slide of MSoft from its dominance and power.
Your examples 6 & 8 seem to be very much what the oil majors are doing, and I would say clearly coercise. i.e. don't restrict us extracting oil or the consequences will be BAD (at least in the short to medium term). 8 is very much in this vein, as they have systematically eliminated alternatives, even greenwashing their pitiful developments to delay actual widespread deveopment of alternatives. They use arguments to slow up alternatives and continue to buy the legislators to ensure they continue. Quite a few products and services that are net BAD are allowed to continue. Without available alternatives, it is coercive power.
Their argument, to the inadequate extent I know it, would seem to me conditioned by the environment in which the _default_ Valley position is that any check by the State were illegitimate. This gives them incentive to highlight, and Our Gracious Host would hold overstate, the existence of coërcive non-State power that has been mystified, for example as Just How Things Work, or, even more obnoxiously, as the exercise of the natural rights of lions which must not, on pain of Not Progressing, be limited to those of oxen.
> in modern society “power over”—the P-word—is seen as ipso facto illegitimate .... Only consent—and only true consent, not coërced consent.
It seems to me that this principle of non-coercion is somewhat limited by considerations of granularity. While I may consent to the operations and demands of government as an aggregate or bundle, I may not consent to any number of the specific actions of government. Government is sort of like a cable or streaming package: If you subscribe to the bundle, then in order to get the things you like, you'll need to tolerate a number of things that you don't like or don't care about.
But if you don't wish to subscribe, what are the alternatives? Even a democracy requires consent to the wishes of the majority. But what happens when teh minority refuses to give that consent to the majority. Even teh majority doesn't have carte blanche, as the US Constitution includes articles to prevent a "tyranny of the majority" under certain circumstances. Any hierarchical social structure results in some coercision of the lower levels and hence power over those lower levels.
I would nominate two things for more-current bad tech: 1) audio-visual recordings and (much more purely) 2) automobiles. These are both VERY favored by market forces, and both VERY close to un-debatable in our democracy.
Why are AV recordings bad tech? That wounf include photos, music, tv and movie recordings, personal film and video, etc. That destroys their historical value if nothing else.
I'm not sure they're bad on balance. But there's a Postmanian case against them as dangerous replacements for print culture. Of course, putting Bull Connor on TV has its place, nonetheless.
First thought is that debates about the conceptual line between freedom and coercion have a very long and complex history and are very much ongoing. One could compare Isaiah Berlin’s attempt to distinguish positive and negative freedom: influential, but highly contested as well.
I don’t see that using the word “power” to describe 2,4,6,8,9 is unusually distorting or necessarily leads people to think of them pejoratively. And I’m not sure what other word one could use. It seems some well-chosen adjectives can keep the distinction between types of power. For example, international relations people distinguish soft power and hard power, without this leading them to conflate them or think of them as equally bad. It all comes down to context in concrete cases. One could classify China’s belt and road initiatives as soft power and still think of it as quite pernicious/exploitative and a kind of coercion, even if the Chinese themselves think it is just fair persuasion and fair exchange.
The debate here seems rather analogous to long-standing debates about advertising – is it just distributing information about products to rational well-informed consumers, or is it actually creating wants and desires in consumers? While there is probably a lot of fairly innocuous advertising, It is not hard to think of highly pernicious cases like cigarette advertising. It would be hard to argue that as people became aware of the negative consequences and the addictive characteristics of nicotine, the people making money from them were not highly resistant to accepting that view. Acemoglu and Johnson are just saying something similar might well happen with AI. We can expect those making money from it are likely to be overly-optimistic in arguing for the upside and downplaying the downside. Though by the same token any kind of regulator, no matter how conscientious, is probably going to err on the pessimistic side. History would seem to say that it is very difficult to predict the upsides and downsides of new technologies. Keynes would tell us that in the face of true uncertainty (knightian uncertainty) about the future, we fall back on animal spirits and fluctuate between being overly-optimistic and overly-pessimistic.
Yes: "hard power" and "soft power" is an essential distinction to make if you are going to sweep everything under the rug and call it "power". But wouldn't it be better to distinguish "power" from something else—call it influence, call it leadership, call it preëminence, call it positive-sum coördination, call it management, call it alliance...
Fair point -- and in particular empirical cases all those terms may be appropriate descriptors. In which case, calling it an exercise of power "over" another party is maybe a little too Hobbesian. On the other hand, I still think what is a stake in such cases can be fairly described as a use of power, or perhaps, more neutrally, a capacity or ability. What one is going to call the use of power in a particular case depends on one's normative evaluation of how the power is being used -- is it being used in a way consistent with the interests and consent of the other party. Then the problem is less the use of the term power, but whether one thinks it is a cooperative exercise of power "with" another party, or an exercise of power "over." Though I think your right to say that we tend to slip easily form talking about a use of power, to infer that it it must be a use of power over someone.
I can't help thinking that this is an instance of the Hayek v Polanyi framing of STU. Your concern here being that most social scientists are too Polanyi, and not enough Hayek. Or in Schumpeterian terms, too focussed on the destructive and not the creative side.
I'm also curious that you framed your answer by identifying yourself as an economist and not an economic historian -- does that imply that you think most of your fellow economic historians and even just plain historians, would not agree with economists on this use of the term power?
Noah's outrage at A&J's book, before he had even read it, seemed over the top at the time. His reading of the book added no nuance. It seemed like a strawman was built and burned on Hexapodia. I wish Acemoglu had been on the podcast to defend his work.
Then, of course, there are the at least nominally regretfully post-libertarian, who reject the notion of power-over as illegitimate. I recall there being a nice exchange in "A Fish Called 'Wanda'"* about such persons….
(Not strictly accurate, though, we shouldn't white-wash Nietzsche: though he would have loathed the Nazis' vulgarity and popularism, his regret over the brutal nature of 'that which survives'—and he did _not_ mean Lee Meriweather—at other times seems crocodilachrymose. He wouldn't be the first weedy and bullied boy who ended worshipping strength.)
I'm sure you are familiar with Robert Moses, but you might want to spend some time browsing the Instagram archive of Segregation By Design. They have some excellent before-and-after illustrations, both fixed and animated, of how the California Department of Highways (now California Dept of Transportation [with a 90% focus on automobile transportation] was used over and over again to destroy any buildup of social stability or economic wealth by African-Americans in the Golden State. Every single time African-Americans developed a stable prosperous community in the LA Basin the Dept of Highways would announce that the only feasible route - the only possible route - for a new or expanded highway just happened to run through that heart of that neighborhood. Local politicians would shake their heads in sorrow and proclaim "there is no choice", and another minority neighborhood would be smashed. I would call that power to destroy (more precise version: power to destroy opportunity), which is not the same as (1) or (3) and which despite more light being shone on it continues to this day.
Not only economists: For the take of an historian (who paid attention to Lukes) on ways to study power from more than forty years ago, see David C. Hammack, “Problems in the Historical Study of Power in the Cities and Towns of the United States, 1800-1960,” American Historical Review, April, 1978, pp. 323-349, https://www.jstor.org/stable/1862320 . The book that employed this approach is Hammack, POWER AND SOCIETY: GREATER NEW YORK AT THE TURN OF THE CENTURY (Russell Sage Foundation, 1982): https://www.jstor.org/stable/10.7758/9781610442657
I think Farrell makes the stronger argument. As a not to recent example, recall how Microsoft c. 1990s exreted power over the OS and applications market. Gates could just state that MSoft was going to do [X] and this then dissuaded any number of companies from even trying to build anaything. Announcing vaporware was effectively coercise power, but framed as a technology development decision. Of course in teh end this put them in the crosshairs of the antitrust law (that ultimately failed) but did trigger the slide of MSoft from its dominance and power.
Your examples 6 & 8 seem to be very much what the oil majors are doing, and I would say clearly coercise. i.e. don't restrict us extracting oil or the consequences will be BAD (at least in the short to medium term). 8 is very much in this vein, as they have systematically eliminated alternatives, even greenwashing their pitiful developments to delay actual widespread deveopment of alternatives. They use arguments to slow up alternatives and continue to buy the legislators to ensure they continue. Quite a few products and services that are net BAD are allowed to continue. Without available alternatives, it is coercive power.
Their argument, to the inadequate extent I know it, would seem to me conditioned by the environment in which the _default_ Valley position is that any check by the State were illegitimate. This gives them incentive to highlight, and Our Gracious Host would hold overstate, the existence of coërcive non-State power that has been mystified, for example as Just How Things Work, or, even more obnoxiously, as the exercise of the natural rights of lions which must not, on pain of Not Progressing, be limited to those of oxen.
Touché...
> in modern society “power over”—the P-word—is seen as ipso facto illegitimate .... Only consent—and only true consent, not coërced consent.
It seems to me that this principle of non-coercion is somewhat limited by considerations of granularity. While I may consent to the operations and demands of government as an aggregate or bundle, I may not consent to any number of the specific actions of government. Government is sort of like a cable or streaming package: If you subscribe to the bundle, then in order to get the things you like, you'll need to tolerate a number of things that you don't like or don't care about.
Yes...
" If you subscribe to the bundle"
But if you don't wish to subscribe, what are the alternatives? Even a democracy requires consent to the wishes of the majority. But what happens when teh minority refuses to give that consent to the majority. Even teh majority doesn't have carte blanche, as the US Constitution includes articles to prevent a "tyranny of the majority" under certain circumstances. Any hierarchical social structure results in some coercision of the lower levels and hence power over those lower levels.
I would nominate two things for more-current bad tech: 1) audio-visual recordings and (much more purely) 2) automobiles. These are both VERY favored by market forces, and both VERY close to un-debatable in our democracy.
Why are AV recordings bad tech? That wounf include photos, music, tv and movie recordings, personal film and video, etc. That destroys their historical value if nothing else.
I'm not sure they're bad on balance. But there's a Postmanian case against them as dangerous replacements for print culture. Of course, putting Bull Connor on TV has its place, nonetheless.
Does Brad mean that "market power" isn't power?
First thought is that debates about the conceptual line between freedom and coercion have a very long and complex history and are very much ongoing. One could compare Isaiah Berlin’s attempt to distinguish positive and negative freedom: influential, but highly contested as well.
I don’t see that using the word “power” to describe 2,4,6,8,9 is unusually distorting or necessarily leads people to think of them pejoratively. And I’m not sure what other word one could use. It seems some well-chosen adjectives can keep the distinction between types of power. For example, international relations people distinguish soft power and hard power, without this leading them to conflate them or think of them as equally bad. It all comes down to context in concrete cases. One could classify China’s belt and road initiatives as soft power and still think of it as quite pernicious/exploitative and a kind of coercion, even if the Chinese themselves think it is just fair persuasion and fair exchange.
The debate here seems rather analogous to long-standing debates about advertising – is it just distributing information about products to rational well-informed consumers, or is it actually creating wants and desires in consumers? While there is probably a lot of fairly innocuous advertising, It is not hard to think of highly pernicious cases like cigarette advertising. It would be hard to argue that as people became aware of the negative consequences and the addictive characteristics of nicotine, the people making money from them were not highly resistant to accepting that view. Acemoglu and Johnson are just saying something similar might well happen with AI. We can expect those making money from it are likely to be overly-optimistic in arguing for the upside and downplaying the downside. Though by the same token any kind of regulator, no matter how conscientious, is probably going to err on the pessimistic side. History would seem to say that it is very difficult to predict the upsides and downsides of new technologies. Keynes would tell us that in the face of true uncertainty (knightian uncertainty) about the future, we fall back on animal spirits and fluctuate between being overly-optimistic and overly-pessimistic.
Yes: "hard power" and "soft power" is an essential distinction to make if you are going to sweep everything under the rug and call it "power". But wouldn't it be better to distinguish "power" from something else—call it influence, call it leadership, call it preëminence, call it positive-sum coördination, call it management, call it alliance...
Fair point -- and in particular empirical cases all those terms may be appropriate descriptors. In which case, calling it an exercise of power "over" another party is maybe a little too Hobbesian. On the other hand, I still think what is a stake in such cases can be fairly described as a use of power, or perhaps, more neutrally, a capacity or ability. What one is going to call the use of power in a particular case depends on one's normative evaluation of how the power is being used -- is it being used in a way consistent with the interests and consent of the other party. Then the problem is less the use of the term power, but whether one thinks it is a cooperative exercise of power "with" another party, or an exercise of power "over." Though I think your right to say that we tend to slip easily form talking about a use of power, to infer that it it must be a use of power over someone.
I can't help thinking that this is an instance of the Hayek v Polanyi framing of STU. Your concern here being that most social scientists are too Polanyi, and not enough Hayek. Or in Schumpeterian terms, too focussed on the destructive and not the creative side.
I'm also curious that you framed your answer by identifying yourself as an economist and not an economic historian -- does that imply that you think most of your fellow economic historians and even just plain historians, would not agree with economists on this use of the term power?
Noah's outrage at A&J's book, before he had even read it, seemed over the top at the time. His reading of the book added no nuance. It seemed like a strawman was built and burned on Hexapodia. I wish Acemoglu had been on the podcast to defend his work.
Then, of course, there are the at least nominally regretfully post-libertarian, who reject the notion of power-over as illegitimate. I recall there being a nice exchange in "A Fish Called 'Wanda'"* about such persons….
(Not strictly accurate, though, we shouldn't white-wash Nietzsche: though he would have loathed the Nazis' vulgarity and popularism, his regret over the brutal nature of 'that which survives'—and he did _not_ mean Lee Meriweather—at other times seems crocodilachrymose. He wouldn't be the first weedy and bullied boy who ended worshipping strength.)
*(how it _should_ be spelt)
I'm sure you are familiar with Robert Moses, but you might want to spend some time browsing the Instagram archive of Segregation By Design. They have some excellent before-and-after illustrations, both fixed and animated, of how the California Department of Highways (now California Dept of Transportation [with a 90% focus on automobile transportation] was used over and over again to destroy any buildup of social stability or economic wealth by African-Americans in the Golden State. Every single time African-Americans developed a stable prosperous community in the LA Basin the Dept of Highways would announce that the only feasible route - the only possible route - for a new or expanded highway just happened to run through that heart of that neighborhood. Local politicians would shake their heads in sorrow and proclaim "there is no choice", and another minority neighborhood would be smashed. I would call that power to destroy (more precise version: power to destroy opportunity), which is not the same as (1) or (3) and which despite more light being shone on it continues to this day.
Not only economists: For the take of an historian (who paid attention to Lukes) on ways to study power from more than forty years ago, see David C. Hammack, “Problems in the Historical Study of Power in the Cities and Towns of the United States, 1800-1960,” American Historical Review, April, 1978, pp. 323-349, https://www.jstor.org/stable/1862320 . The book that employed this approach is Hammack, POWER AND SOCIETY: GREATER NEW YORK AT THE TURN OF THE CENTURY (Russell Sage Foundation, 1982): https://www.jstor.org/stable/10.7758/9781610442657