beliefs. Here I propose to define beliefs as multiply-encoded cognitive/affective
configurations, usually including (but not limited to) prepositional encoding, to
which the holder attributes some kind of truth value. The latter term is not taken in
the technical sense of symbolic logic, but as a term that may variously refer to
logical truth, empirical truth, validity, applicability to some degree of
approximation, metaphysical truth, religious truth, practical truth, or conventional
truth. Within mathematics, the “truth value” might be logical truth in the sense of
deducibility from specific formal assumptions, or it might sometimes be
conventional truth in the sense of satisfying some agreed-upon, arbitrary rules of
definition or notation.
A belief structure is a set of mutually consistent, mutually reinforcing, or
mutually supportive beliefs and warrants (see below) in the individual, mainly
cognitive but often incorporating supportive affect. A belief system is an elaborate or
extensive belief structure that is socially or culturally shared. Since I may be
employing these terms rather differently from their casual uses, let me stress that my
intent is to distinguish individuals' belief structures from socially or culturally
shared belief systems.
A belief is, to begin with, individually held: for example, I believe that guiding
children to discover logical patterns for themselves generally fosters their enjoyment
and learning of mathematics. Such a belief may or may not be shared - some may
agree with me, others may not. Belief structures, like cognitive structures, refer here
to the individual's complex, personal, internal representational configurations: my
belief about guiding children to discover patterns for themselves does not stand in
isolation in my head; it is part of a structure of mutually reinforcing beliefs that I
hold, together with a variety of reasons - or warrants - I have for holding them.
Beliefs and belief structures are important in understanding individuals'
mathematical problem solving heuristics and strategies (Schoenfeld, 1985; Lester,
Garofalo, & Lambdin Kroll, 1989).
Belief systems, on the other hand, refer to socially or culturally shared belief
structures, that are sufficiently broad to warrant the term. Shared beliefs, or belief
systems are, in turn, not exactly the same as normative beliefs. The latter are
idealized, approximate descriptions, at the societal or cultural level, of beliefs that
one “should” hold (but may or may not actually be held by very many people).
Relations among social norms and shared beliefs, the beliefs of individuals, and
emotions are discussed further by Cobb, Yackel, & Wood (1989) and Cobb &
Yackel(1996).
Note that it is the attribution of truth (of some kind) that turns mere propositions,
conjectures, stories, or hypotheses, into beliefs. This attribution is by the holder, and
not necessarily by others. It is not to be assumed, even when a belief is shared and
normative, that the believer or believers are correct in their attributions of truth or
validity to it. That is, some beliefs may - in fact - be false ones.
There is an unfortunate tendency among cultural relativists to use the word
“knowledge” as if it were synonymous with “belief”, “shared belief”, “normative
belief”, “warranted belief”, or some combination of these. For example, according to
Confrey (2000), constructivism - which, in its radical formulation, has significantly