14.3.2 Automated reasoning
At about the same time that some computer scientists were studying ways of creating computer systems that could play games (late 1950’s and 1960’s), other computer scientists were thinking about the problem of how to build a system that could “reason”. Much of the early work on automated reasoning focused on constructing computer programs to carry out the rules of symbolic logic. A symbolic logic, or formal logic, combines a way of representing information using symbols together with a collection of rules for manipulating those symbols, in order to reach logically valid conclusions.
The reason that symbolic logic was of such interest to computer scientists studying artificial intelligence is that these logics are capable of “reasoning” about concepts and ideas based purely on the form of the statements used to represent them. In other words, a symbolic logic can draw logically valid conclusions about a collection of statements based not on the meaning of those statements, but on the form in which they are written.[10]
There are many kinds of symbolic logics. In this chapter we will look at one logic, propositional logic, in some detail. 10
Footnotes
[10] Believe me, I know this sounds odd. But it will actually make sense once we look at some examples.