Putting aside his rabid, CIA-sponsored anti-communism, I am a big fan of James C. Scott’s Seeing Like A State. Above all, it’s a useful and well-researched critique of engineering.
What Scott critiques are grand schemes to change systems in which people and nature are deeply implicated. A characteristic such scheme is what was called Scientific Forestry, in which forests were demolished and re-planted as tree monocrops in nice clean lines. The artificial forests withered and wood production crashed. It turns out, Scott argues, that complexity cannot be engineered out of complex systems.
The trouble is that Scott doesn’t call this process "engineering." He calls it science.
Science, properly, is an episteme. The word “science” is also often used to refer to the body of scientific knowledge, which is the set of claims that are believed to be supported by that episteme. This is generally an acceptable conflation. For example, the act of comparing models for gene replication by comparing them with data can be called “science.” Through this process, some of those models gain status as truth claims about genes. They are inaugurated into the set of claims about biology that we think are true, because we did science!
What is essential to science is its epistemological nature. It is a process of research, of learning.
Let’s say that we know how to make humans glow in the dark using CRISPR gene editing. This is a technique, which is a kind of knowledge, which we acquired through a scientific process.
There is nothing scientific about designing and implementing a national program to make every American glow in the dark. It is not about learning (even if Scott would eventually make it into a teachable experience). It is an act of engineering.
Engineering is an act of changing the world materially using claims from the body of scientific knowledge.
It is not often that a failure of engineering indicates a fault of science. The power of science to support knowledge claims is proportional to the simplicity of the system it attempts to describe. The process of science often involves acts of engineering in order to produce experimental contexts that are adequate for the production of knowledge claims. But engineering itself necessarily encounters the world in all its complexity. There are a million things that can go wrong with a machine in the world.
In other words, there was nothing “scientific” about Scientific Forestry. It was a grandiose kind of engineering. The only kinds of “scientific” claims that its failure challenged were claims of total knowledge, which are not scientific in the first place. A scientific claim is one like “within certain parameters, a tree in soil with a certain level of phosphorus will grow faster than in soil with higher or lower levels, all else being equal.” Claims of total knowledge attempt to enumerate or eliminate this “all else.” Scientific Forestry was an engineering project premised on a total knowledge of the factors of wood production.
Scott’s attack on science as a process of knowing, then, is entirely attributable to this conflation of science and engineering.
This ill-founded critique of science is Scott’s contribution to the postmodern discourse supporting ontological relativism. It must be understood that for every anti-science right-wing writer, there is a corresponding "left-wing" academic who spends their career undermining science in the name of the subaltern. This is extremely ironic, given the meaning that science, progress, and communism has had for the oppressed in the 20th century. But given the choice between rejecting the idea that science can justify oppression and rejecting science altogether, they choose the quixotic and self-defeating latter.