Thinking and Computation - Thinking as Computation - page 19

enormous collection of sentences, an entire web of belief. Such a collection is called a
knowledge base (KB). The collection shown in figure 1.5 is just a very small sample.
So yes, there is a connection between thinking and logic, but it is misleading to
think of it as Logic, the subject studied in philosophy. The way Logic is normally
taught, one starts with a small set of premises and concentrates on ensuring that the
conclusions from them are always correct:
Socrates is a man.
All men are mortal.
Therefore, Socrates is mortal.
Thinking is very different. It starts with an enormous collection of premises (maybe
millions of them) about a very wide array of subjects. There will be facts in the knowl-
edge base about George and marriage, but also about barber shops, ferris wheels,
Academy Awards, hate literature, and so on. The question then is, what are the
logical entailments of all those sentences?
This leads to two hypotheses:
Much of the subjective feeling of richness experienced in thinking might be
explained as simple mechanical and logical operations, but applied to a very
rich collection of sentences.
To build computer systems with a number of desirable properties (versatile,
flexible, extendible, easily maintained, and so on), one must
- represent much of what the system needs to know as symbolic sentences of
some sort, called its knowledge base;
- perform processing over the knowledge base using the rules of logic to
obtain new conclusions;
- have the system act based on the conclusions it can derive.
Systems built this way are called knowledge-based systems .
The big picture
In summary, thinking means bringing what one knows to bear on what one is doing.
But how does this work? How do concrete, physical entities like people engage with
something formless and abstract like knowledge? What is proposed in this chapter
(via Leibniz) is that people engage with symbolic representations of that knowledge. In
other words, knowledge is represented symbolically as a collection of sentences in a
knowledge base, and then entailments of those sentences are computed as needed.