Mechanical Thinking Part Deux
In a previous post I tried to justify mechanical thinking by saying
[Mechanical Thinking is] an attempt to create "laws of thought" that are akin to Newton's "laws of motion". In the worldview of Newton's laws, if you could know the exact state of everything in the universe, then you can determine the path that an apple will take when it falls from a tree. Well that's what axiomatic thinking gives you in the realm of thinking and reasoning. The axioms are the exact state of everything that you initially know about. Then the rules of logic tell you how to determine everything after that. It's neat. It's clean. There's no way to mess it up. Like clockwork, as the common analogy says. Newton and friends thought of God as a Watchmaker, setting up the initial state of the universe and letting it run by universal predetermined for the rest of time.
However, that's all rather abstract, and I think there's an even better practical reason for investigating mechanical thinking. Just look at that word, mechanical, and what it relates to: machines.
In many ways, the history of human civilization is the history of its machines. The wheel is a famous example from early civilization. It led to great advances in both art (potter's wheel) and war (the chariot). It also incredibly increased the speed of travel and hence the flow of ideas across ancient Eurasia. It is extremely obvious that modern society could not exist without the wheel.
Machines are tools that augment a human's (or animal's) ability to do labor. In ordinary usage, machine refers to something more complicated than a tool, but the line can get fuzzy and I don't want to get into that distinction right now.
magus341 would discuss this as gains in productivity. A machine is more productive than some other process if it allows us to take the same amount of input resources and turn them into more output products. As machines have gotten more and more sophisticated, they have further increased more and more in the productivity gains they have given us.
An important property of a large class of machines is that they can perform a large amount of the work without human intervention. A water wheel or windmill turns the gears without a human directing it to do so. An old-fashioned watch similarly operates with minimal intervention from people. A human's physical activity doesn't actually propel an automobile forward. And so on. A world without machines would be a Flintstones world, where humans and animals have to provide all of the labor on the planet. But unlike the fictional Flintstones, in a real world like that one nothing could ever get done!
This property, of machines working with little/no human intervention, is what we call automation. We can think of it as "outsourcing" our human labor to machines to do it for us. Except without the negative connotations of outsourcing (unless you're a luddite). Instead, it's liberating! If we have machines do these tedious things for us automatically, then that frees up time for us to do art and music and philosophy and those other things that make us uniquely human. And nothing's to stop you from getting down there and still dig in the dirt or whatever, if that's what you want to do too.
This aspect, of machines augmenting human physical labor and increasing our productivity, I think is rather well known to most people. Now let's apply it to "mechanical thinking".
First, since we're in the realm of "thought", which isn't physical (in the conventional sense), we have to dispose of the requirement that the "machine" be physical as well. Think of the process you learned in elementary school for multiplying multi-digit numbers. First, you memorized the times tables, using only the human power of your brain. But if you had to continue memorizing multiplication values, that would obviously end up being impossible! So you learned a mechanical process for multiplying a two digit number by another two digit number. And now you can (in principle) multiply any two numbers together! This multiplication "machine" has augmented your puny human ability to multiply numbers. By the way, any "mechanical" process like this is what computer programmers refer to as an "algorithm", but that term isn't important for the time being.
It wouldn't be until about sometime in the nineteenth century that the ability to automate this type of thinking really became possible, with Charles Babbage and his Difference Engine (an extremely fascinating historical story that even predates Alan Turing). Any sort of automation invariably requires some sort of physical embodiment in the real world. And hence, after many permutations, a modern digital computer.
The study of mechanical thinking, then, is the study of the mechanization of thinking (i.e., the formal processes of reasoning and thought, like the multiplication example above) as well as the automation, or physically carrying out those mechanical processes with little-to-no human intervention (as in the PC you are reading this on). I may be attracted to the austere beauty implied by looking at mechanical thinking as Laws of Thought akin to the Laws of Motion of Newton's Watchmaker. But hopefully this view of mechanization and automation shows that its also interesting for other reasons, as well :-)
[Mechanical Thinking is] an attempt to create "laws of thought" that are akin to Newton's "laws of motion". In the worldview of Newton's laws, if you could know the exact state of everything in the universe, then you can determine the path that an apple will take when it falls from a tree. Well that's what axiomatic thinking gives you in the realm of thinking and reasoning. The axioms are the exact state of everything that you initially know about. Then the rules of logic tell you how to determine everything after that. It's neat. It's clean. There's no way to mess it up. Like clockwork, as the common analogy says. Newton and friends thought of God as a Watchmaker, setting up the initial state of the universe and letting it run by universal predetermined for the rest of time.
However, that's all rather abstract, and I think there's an even better practical reason for investigating mechanical thinking. Just look at that word, mechanical, and what it relates to: machines.
In many ways, the history of human civilization is the history of its machines. The wheel is a famous example from early civilization. It led to great advances in both art (potter's wheel) and war (the chariot). It also incredibly increased the speed of travel and hence the flow of ideas across ancient Eurasia. It is extremely obvious that modern society could not exist without the wheel.
Machines are tools that augment a human's (or animal's) ability to do labor. In ordinary usage, machine refers to something more complicated than a tool, but the line can get fuzzy and I don't want to get into that distinction right now.
magus341 would discuss this as gains in productivity. A machine is more productive than some other process if it allows us to take the same amount of input resources and turn them into more output products. As machines have gotten more and more sophisticated, they have further increased more and more in the productivity gains they have given us.
An important property of a large class of machines is that they can perform a large amount of the work without human intervention. A water wheel or windmill turns the gears without a human directing it to do so. An old-fashioned watch similarly operates with minimal intervention from people. A human's physical activity doesn't actually propel an automobile forward. And so on. A world without machines would be a Flintstones world, where humans and animals have to provide all of the labor on the planet. But unlike the fictional Flintstones, in a real world like that one nothing could ever get done!
This property, of machines working with little/no human intervention, is what we call automation. We can think of it as "outsourcing" our human labor to machines to do it for us. Except without the negative connotations of outsourcing (unless you're a luddite). Instead, it's liberating! If we have machines do these tedious things for us automatically, then that frees up time for us to do art and music and philosophy and those other things that make us uniquely human. And nothing's to stop you from getting down there and still dig in the dirt or whatever, if that's what you want to do too.
This aspect, of machines augmenting human physical labor and increasing our productivity, I think is rather well known to most people. Now let's apply it to "mechanical thinking".
First, since we're in the realm of "thought", which isn't physical (in the conventional sense), we have to dispose of the requirement that the "machine" be physical as well. Think of the process you learned in elementary school for multiplying multi-digit numbers. First, you memorized the times tables, using only the human power of your brain. But if you had to continue memorizing multiplication values, that would obviously end up being impossible! So you learned a mechanical process for multiplying a two digit number by another two digit number. And now you can (in principle) multiply any two numbers together! This multiplication "machine" has augmented your puny human ability to multiply numbers. By the way, any "mechanical" process like this is what computer programmers refer to as an "algorithm", but that term isn't important for the time being.
It wouldn't be until about sometime in the nineteenth century that the ability to automate this type of thinking really became possible, with Charles Babbage and his Difference Engine (an extremely fascinating historical story that even predates Alan Turing). Any sort of automation invariably requires some sort of physical embodiment in the real world. And hence, after many permutations, a modern digital computer.
The study of mechanical thinking, then, is the study of the mechanization of thinking (i.e., the formal processes of reasoning and thought, like the multiplication example above) as well as the automation, or physically carrying out those mechanical processes with little-to-no human intervention (as in the PC you are reading this on). I may be attracted to the austere beauty implied by looking at mechanical thinking as Laws of Thought akin to the Laws of Motion of Newton's Watchmaker. But hopefully this view of mechanization and automation shows that its also interesting for other reasons, as well :-)