I think the world needs more engineering, not just in the realm of natural sciences, but also in the social sciences. In this context, I’m using the word "engineering" to describe the process of understanding a preexisting system or natural occurrence, and then building additional systems on top of it to help humans influence or control that thing. I know that definition is overly broad and generic, but don’t worry, it’ll be useful.
I think, generally, there are two types of knowledge.
The first is knowledge for the sake of knowledge, just knowing that something exists or doesn’t exist, or understanding how something works. This kind of knowledge helps us answer questions and satisfy our curiosity. Like knowing why the sky is blue, or why the sun always rises in the east. It helps us understand our environment, how things came to be the way they are, and so on. This kind of knowledge is tremendously useful because it helps us map the domains of our existence. This is what I think science does.
The second type is knowledge for tools, knowledge that helps us create things that take us from merely understanding our existence to actually controlling it. This type of knowledge is built on top of the first kind. Obviously, you need some understanding of a thing before you can begin developing tools to manipulate it. This is what I think engineering is.
So, in brief: we use science to understand systems, and we use engineering to control them.
Here’s an example: we use knowledge from astronomy to understand why seasons exist and why winter happens. That’s the understanding part. But that knowledge alone isn’t very helpful to someone who’s dying of hypothermia in the winter. So, we combine our understanding of astronomy, thermodynamics, and electronics to create a tool, an air conditioner or heater, that lets us regulate indoor temperatures regardless of the whims of the planet. We can go even further and build gas pipelines to heat every home in an entire country.
These two kinds of knowledge don’t oppose each other, they build on top of each other. Engineering uses what we’ve learned from science to create new and better tools. And those tools, in turn, help us gain new scientific insights. I’m not saying one is better than the other; both are important. The key difference is that one diagnoses, and the other prescribes.
Science itself is also split into two broad domains. One seeks to understand the natural world, what exists independent of human influence. The other tries to understand the systems we’ve built for ourselves: our societies, institutions, and the behaviors that emerge from human interaction. Of course, not all social systems are intentionally designed, many are emergent properties of large numbers of people interacting. The first domain is what we call natural science, and the second is social science.
The main point I want to make is that natural science has embraced tools and engineering from the very beginning. It grew alongside them. But social science, in contrast, often seems willfully ignorant of, or even fearful towards engineering. It seems nearly incapable of prescribing solutions.
To me, it feels like as we make progress in natural science and its associated engineering, we’re hitting bottlenecks caused by the lack of engineering tools coming out of social science. That’s why I’m saying: we need more engineering. We need more social engineering, not in the pejorative or manipulative sense, but in the practical, systems building sense.
There are some good reasons why we don’t see much engineering coming out of social sciences.
One is the potential for fallout from a failed or misused tool. The consequences can be severe. Economic policy is a good example of a tool born from social science engineering. Implementing a particular economic policy in a given region makes the policymakers directly responsible for any poverty or suffering it might cause. We often see this with demographic policies, too. Because of this, many experts in social sciences are wary of making prescriptive, engineering style moves, there are just too many examples of them going wrong.
Another challenge is that social engineering often requires holistic application to be effective. That means everyone in a given area has to adopt it for it to work. These changes can’t usually be applied incrementally, so they require a lot of power and effort to execute. And often, the people with the power to make such changes aren’t the same as those who would have designed the new tools or systems.
I don't really have a solution for the lack of enginering tools coming out of social science. I just wish more social science experts focused on controlling, not just understanding. I wish more of them had an engineering mindset.