The second half of my conversation with Mark Mandeles has been released on all major podcast apps. In this episode Mark and I talk about whether science is slowing down, how path dependence analysis works, the benefits of positive feedback loops, what the central task of analysts ought to be, how we create high reliability organizations, what’s the role of history in weapon system projects, and much more.
I also ask Mark about how some of the great thinkers of the 20th century should impact our thinking about acquisition. From Charles Perrow on high reliability organizations and Martin Landau on the benefits of social redundancy, to Friedrich Hayek on the uses of knowledge and Herbert Simon on, well, almost everything. But Mark’s favorite philosopher of the 20th century is Karl Popper, who has much to teach us on falsification, the growth of knowledge, and the open society.
I again can’t recommend enough reading through some of Mark’s great papers, mostly online for free, including the B-52 Development book, Systems Design and Project Management Principles, and Needs and Opportunities in US Naval History which we discuss on the program. His books are also available to buy on Amazon here.
Commentary on the conversation.
Eric: Do you think that the rate of progress in weapon systems is slowing down? After all, they want to extend the life of the B-52 until 2050, and I believe the last B-52H was produced in 1962. What do you make of that?
Mark: That’s a really hard question. I don’t have the data, but I do have some feelings. I think technological progress is accelerating. This gives us a situation where there is a tremendous amount of accumulated knowledge… It leads to new research areas. We have to figure out how to index and access and gather and employ this vast amount of information…
We can expect that there are many revolutions in military affairs or transformations proceeding in parallel without our notice. Enabling technologies are being developed and that’s being moved fairly rapidly into commercial practice or other venues.
I tend to agree with Mark that the base of information has grown a great deal. We are expanding on many different fronts of technology that cannot be captured by single metrics, or even a diversity of metrics, because the relevant number of metrics are ever-expanding and changing. So I think Mark is right, a lot is happening in parallel without our notice.
By the way, there was an excellent EconTalk episode with Patrick Collison on whether science is slowing down. I think Collison was making the point that science does appear to be slowing down, but it doesn’t have to be. He thinks that there’s a lot of social issues at play. I agree broadly, but I think that science isn’t slowing down as much as Collison claims due to the measurement problem, or perhaps I think the bias in our interpretation of the data makes us think it is slowing down faster than it is in reality.
Mark Mandeles also picked up on these issues:
On the other hand, we have an alternative view, which is that it is conceivable that technological progress can be slowed or stymied by poor interaction, inadequate support for thinkers and entrepreneurs, inappropriate budgeting procedures, inadequate enabling technologies, opportunities costs imposed by Gresham’s Law of Planning or other sunk costs.
Lots more in the back-and-forth between Mark and myself. Here’s a teaser:
Eric: You’ve really done a great job of taking great thinkers over time and integrating their ideas into defense acquisition. When I read some of that it really inspired me. I’m going to throw out a few names, and can you tell us why they might be important for us to think about in terms of defense?
It feels like a crime that these thinkers that Mark has identified are not integral parts of every acquisition debate. The thinkers include Karl Popper, Herbert Simon, Friedrich Hayek, Charles Perrow, and Martin Landau. Check out the podcast for his take.
Here is Mark on the principles for high reliability organizations:
Another professor of mine from UC Berkeley, Todd La Porte, in his research program “high reliability organizations” noted that there are organizations that contribute to the extraordinary use of complex technologies in very difficult task environments. They have several attributes that are important.
First, there is the high technical and interpersonal selection criteria for positions — merit criteria. Second, continual training and continual improvement efforts of those people in the organization. Third, the attitude of mindfulness, of the importance and necessity of identifying potential errors before they occur. Fourth, development of latent networks of expertise that are activated at the realization of an unanticipated event. And finally, alignment of organization structure of expertise and authority.
I think these 5 points are really quite insightful. Do systems program offices in the DOD exhibit these principles?
One of the obstacles to acquisition reform is that the organizational rules and processes [of the DOD] conflict in many ways with the goals of designing and developing effective technologies. The organizational budget process or promotion process, these processes provide incentives for individuals to behave in a certain way, that may make it harder, more difficult, or stymie our development efforts.
Acquisition reform will continue, I think, to cycle around without creating any real improvement unless these problems of organizational design are fixed. I expect that pivoting to a high reliability organization would, in fact, require a complete change in the patterns of thought in the DOD.
Here’s a nice bit, perhaps related:
That just makes the point that experts are sometimes wrong about the way to derive implications from what they know. I recall that Admiral William D. Leahy said before the trinity test that the nuclear bomb wouldn’t work. And he said that as an expert in ordnance of all types…
I advocate tolerance of these kinds of disagreements. And we have a political system, at least in America, over time, the experts view can inform and guide public policy.
Here is Mark on Popper and the growth of knowledge:
The growth of knowledge is central to so many things that the defense department does. The unwillingness or inability of defense officials to understand the implications of Popper’s thinking on the growth of knowledge, falsifications, and research programs, impoverishes us. It makes it more difficult for us to accomplish our goals.
I couldn’t agree more. I’d like to thank Mark Mandeles again for joining me on Acquisition Talk. Please check out the podcast, as there are tons of great insights from Mark that were not excerpted here.
Leave a Reply