ABSTRACT
In this paper I examine the question of how a diagrammatic demonstration (a “proof without words”) could be understood by a computational model. The computational model (a) has a means of representing geometric diagrams composed exclusively of points, line segments, triangles, and quadrilaterals, including the special cases of parallelograms, rhombuses, rectangles, and squares; (b) accepts step-by-step descriptions of specific diagrams, and constructs in computer memory a representation of the diagram as it is described; (c) includes the ability to make modifications to the diagram by construction steps that specify movement of previously constructed components; (d) after each construction step notices any new objects (line segments, triangles, etc.) that are created by the step; (e) accepts a goal statement that the construction sequence is allegedly demonstrating; and (f) attempts to find a justification that confirms the goal statement.
