the CltIes represent not only the elementary symbols, such as those for
"grass", "house", and "car", but also symbols which get created as a result of
the chunking ability of a brain-symbols for such sophisticated concepts as
"crab canon", "palindrome", or "ASU".
Now if it is granted that the notion of taking a trip is a fair counterpart
to the notion of having a thought, then the following difficult issue comes
up: virtually any route leading from one city to a second, then to a third,
and so on, can be imagined, as long as one remembers that some interven-
ing cities are also passed through. This would correspond to the activation
of an arbitrary sequence of symbols, one after another, making allowance for
some extra symbols-those which lie en route. Now if virtually any se-
quence of symbols can be activated in any desired order, it may seem that a
brain is an indiscriminate system, which can absorb or produce any thought
whatsoever. But we all know that that is not so. In fact, there are certain
kinds of thoughts which we call knowledge, or beliefs, which play quite a
different role from random fancies, or humorously entertained absur-
dities. How can we characterize the difference between dreams, passing
thoughts, beliefs, and pieces of knowledge?
Possible, Potential, and Preposterous PathwaysThere are some pathways-you can think of them as pathways either in an
ASU or in a brain-which are taken routinely in going from one place to
another. There are other pathways which can only be followed if one is led
through them by the hand. These pathways are "potential pathways",
which would be followed only if special external circumstances arose. The
pathways which one relies on over and over again are pathways which
incorporate knowledge-and here I mean not only knowledge of facts
(declarative knowledge), but also knowledge of how-to's (procedural knowl-
edge). These stable, reliable pathways are what constitute knowledge.
Pieces of knowledge merge gradually with beliefs, which are also rep-
resented by reliable pathways, but perhaps ones which are more susceptible
to replacement if, so to speak, a bridge goes out, or there is heavy fog. This
leaves us with fancies, lies, falsities, absurdities, and other variants. These
would correspond to peculiar routes such as: New York City to Newark via
Bangor, Maine and Lubbock, Texas. They are indeed possible pathways,
but ones which are not likely to be stock routes, used in everyday voyages.
A curious, and amusing, implication of this model is that all of the
"aberrant" kinds of thoughts listed above are composed, at rock bottom,
completely out of beliefs or pieces of knowledge. That is, any weird and
snaky indirect route breaks up into a number of non-weird, non-snaky
direct stretches, and these short, straightforward symbol-connecting routes
represent simple thoughts that one can rely on-beliefs and pieces of
knowledge. On reflection, this is hardly surprising, however, since it is quite
reasonable that we should only be able to imagine fictitious things that are
somehow grounded in the realities we have experienced, no matter how
(^378) Minds and Thoughts