MODELING and METAMODELING
| Collection | International Encyclopedia of Systems and Cybernetics |
|---|---|
| Year | 2004 |
| Vol. (num.) | 2(2) |
| ID | ◀ 2167 ▶ |
| Object type | Epistemology, ontology or semantics, Methodology or model |
J van GIGCH emphasizes the important distinction between modeling and metamodeling: “Modeling… implies that the modeler ”abstracts“ properties from things in order to obtain a representation of the physical world. It is easy to conceptualize that the model stands at a level of abstraction higher than the things from which the properties are obtained. The process of abstraction can be applied to modeling itself, to obtain a model of the modeling process which we call a metamodel.
- “The metamodel embodies the properties which are abstracted from all models (There is more than one model). By metamodeling we imply a process of design which is carried out at the metalevel, and by which we define how the process of modeling (at the object level of abstraction), is to be carried out. METAMODELING defines the epistemology or design foundations of modeling. Design foundations consist of the reasoning processes, guarantees of truth, proofs, axioms of validity, or any other logic which underlies a methodology.
- “Any discussion of modeling must, of necessity, involve metamodeling from which the former evolved” (1992, p.2-3).
If it does not involve it explicitly, it does it in an implicite — if not more or less unconscious — way. This can be risky, because:
- hidden epistemological assumptions may be unsound
- they may even be perfectly sound in the modeler's view, but not admitted by all stakeholders. This may remain unknown if the metamodel is not openly discussed.
This is still more important at the meta-metamodeling level, where values are paramount.
This “ladder” for validation is related to GÖDEL 's Theorem, as shown by Ch. FRANÇOIS (1987), as well as with KORZYBSKI'S structural differential.