You make good points here. Any ideas why those other shifts happened and how can we help reverse them or prevent them from happening elsewhere?
Mostly it looks to me like a series of unrelated changes built up over time, and the unintended consequences were mostly adverse.
An example is the War on Cancer and the changes that came with it to funding. It had long been the case that funding was mostly handed out on a project-by-project basis, but in order to get the funding dedicated to cancer research it was necessary to explain how cancer research would benefit. The obvious first-order impact is an increase in administrative overhead for getting the money.
Alongside this science sort of professionalized. I expect that when the sense of how important something is permeates, professionalization is viewed as a natural consequence, but it seems to have misfired here. Professionalization, like other forms of labor organization, isn’t about maximizing anything but about ensuring a minimum. This means things like more metrics, which is why our civilization formally prefers a lot of crappy scientific papers to a few good ones, and doesn’t want any kind of non-paper presentation of scientific progress at all. Science jobs become subject to Goodharting, because people start thinking that the right way to get more science is just to increase the number of scientists, on account of them all being interchangeable professionals with a reliable minimum output.
The university environment also got leaned on as a lever for progress; the student loan programs all grew over this same period, which seems to have driven a long period of competition for headcount. This shifted universities’ priorities from executing their nominal mission towards signalling desirability among students/parents/etc. I am certain at least part of that came at the expense of faculty, even if only by increasing the administrative burden still further by yet more metrics.
On the fixing side, I am actually pretty optimistic. A few simple things would probably help a lot, two examples being funding and organization. Example: Bell Labs and Xerox PARC have been discussed here a lot. Both cases deviated significantly from the standard university/government system of funding individual projects case by case. Under the project/grant system being a scientist reduces to being able to successfully get funding for a series of projects over time. At Bell and at PARC, they rather made long-term investments on a person-by-person basis. I think this has wide-ranging effects, but not least among them is that there wasn’t a lot of administrative overhead to a given investigation; rather they could all be picked up, put down, or adapted as needed. Another effect, maybe intentional but seemingly happenstance, is that they built a community of researchers in the colloquial sense. This is pretty different from the formal employee relationships that dominate now. Around 7 years ago I listened to a recruiting pitch from Sandia National Laboratories for engineering students, and asked how communication was between different groups in the lab. The representative said that she knew of a case where two labs right across the hall from each other were investigating the same thing for over a year before they realized it, because nobody talks.
This suggests to me that a university that was struggling financially, or maybe just needed to take a gamble on moving up in the world, could cheaply implement what appears to be a superior research-producing apparatus, just by shifting their methods of funding and tracking results.
Mostly it looks to me like a series of unrelated changes built up over time, and the unintended consequences were mostly adverse.
An example is the War on Cancer and the changes that came with it to funding. It had long been the case that funding was mostly handed out on a project-by-project basis, but in order to get the funding dedicated to cancer research it was necessary to explain how cancer research would benefit. The obvious first-order impact is an increase in administrative overhead for getting the money.
Alongside this science sort of professionalized. I expect that when the sense of how important something is permeates, professionalization is viewed as a natural consequence, but it seems to have misfired here. Professionalization, like other forms of labor organization, isn’t about maximizing anything but about ensuring a minimum. This means things like more metrics, which is why our civilization formally prefers a lot of crappy scientific papers to a few good ones, and doesn’t want any kind of non-paper presentation of scientific progress at all. Science jobs become subject to Goodharting, because people start thinking that the right way to get more science is just to increase the number of scientists, on account of them all being interchangeable professionals with a reliable minimum output.
The university environment also got leaned on as a lever for progress; the student loan programs all grew over this same period, which seems to have driven a long period of competition for headcount. This shifted universities’ priorities from executing their nominal mission towards signalling desirability among students/parents/etc. I am certain at least part of that came at the expense of faculty, even if only by increasing the administrative burden still further by yet more metrics.
On the fixing side, I am actually pretty optimistic. A few simple things would probably help a lot, two examples being funding and organization. Example: Bell Labs and Xerox PARC have been discussed here a lot. Both cases deviated significantly from the standard university/government system of funding individual projects case by case. Under the project/grant system being a scientist reduces to being able to successfully get funding for a series of projects over time. At Bell and at PARC, they rather made long-term investments on a person-by-person basis. I think this has wide-ranging effects, but not least among them is that there wasn’t a lot of administrative overhead to a given investigation; rather they could all be picked up, put down, or adapted as needed. Another effect, maybe intentional but seemingly happenstance, is that they built a community of researchers in the colloquial sense. This is pretty different from the formal employee relationships that dominate now. Around 7 years ago I listened to a recruiting pitch from Sandia National Laboratories for engineering students, and asked how communication was between different groups in the lab. The representative said that she knew of a case where two labs right across the hall from each other were investigating the same thing for over a year before they realized it, because nobody talks.
This suggests to me that a university that was struggling financially, or maybe just needed to take a gamble on moving up in the world, could cheaply implement what appears to be a superior research-producing apparatus, just by shifting their methods of funding and tracking results.