Given the many
theories
of explanation, given their inability
to deal successfully with many everyday
scenarios,
and given the lack of a standard means to
evaluate them,
is there a way to think about this space in general?
One way to begin is to codify the constellation of interrelated
concepts into a list of ontology terms and their definitions. The
following provides the vocabulary used throughout the
Explanatarium. These terms are also included in the glossary.
-
Observation: an assertion made about the
world or some part of it
-
Phenomenon: an observation worthy of
remark and explanation
-
Explanandum: a description of a phenomenon for
which an explanation is desired
-
System: a demarcated part of the world in
which the phenomenon occurs (synonym: context)
-
Element: a tangible, cohesive and discernable
aspect of the system
-
Property: a measurable attribute of an element
-
Status: the values of properties of an element
or set of elements, including the system itself (synonym: state)
-
Event: an instantaneous unit of activity in the
context
-
Stimulus: an event to which the system reacts
-
Response: a change in system status that
results from a stimulus
-
Explanation: a description accounting for a phenomenon
(synonym: explanans)
-
Evidence: a property measurement that support
an explanation
-
Understanding: the perceived increase in
knowledge engendered by an explanation
-
Prediction: an assertion about a future
event or about the future status of system elements that derives
from an explanation
-
(Explanatory) Framework: a cohesive set
of definitions and rules used to organize explanations and
test their validity
-
Representation: an extended and
structured metaphor upon which a framework is based
Several things should be noted about the above definitions. Although
the vocabulary is precise about the part of the world in which
the phenomenon occurs, it delegates to a framework the responsibility
for structuring and validating the explanation. In this sense, it is
similar to the
Unificationist Theory.
Second, it forces on the modeler the responsibility for delineating the
system boundary, much like the
Structural Theory. Third,
it notably delegates concepts like cause, laws,
probabilities, and mechanisms to the specific
framework chosen to express an explanation. The implication is that
the adequacy of an explanation is relative to the context, the
accuracy of the modeled state and the framework used to express
the explanation.