Ontological realism: Methodology or misdirection?

Abstract

In a series of papers over a period of several years Barry Smith and Werner Ceusters have offered a number of cogent criticisms of historical approaches to creating, maintaining, and applying biomedical terminologies and ontologies. And they have urged the adoption of what they refer to as a "realism-based" approach. Indeed, at times they insist that the realism-based approach not only offers clear advantages and a well-founded methodological basis for ontology development and evaluation, but that such a realist perspective is in fact necessary for understanding and using terminologies and ontologies in science.This paper explores a number of questions surrounding such claims, provides a careful characterization of the type of realism recommended by Smith and Ceusters, and evaluates the role that realism plays in the critiques and recommendations that they offer. The conclusion reached is that while Smith's and Ceusters' criticisms of prior practice in the treatment of ontologies and terminologies in medical informatics are often both perceptive and well founded, and while at least some of their own proposals demonstrate obvious merit and promise, none of this either follows from or requires the brand of realism that they propose.Editor's note: A response to this paper from Barry Smith and Werner Ceusters is scheduled to appear in a future issue of Applied Ontology.

Ontology, Ontologies, and Science

“…9 For some insight into issues pertaining to the characterizations of scientific ontologies, their comparability and commensurability, and how an appeal to philosophical positions may be relevant to such concerns and their consequences, see Euzenat and Shvaiko 2007. Merrill 2008, 2010a, 2010b, Smith and Ceusters 2010, Dumontier and Hohendorf 2010, Lord and Stevens 2010, and Kutz et al 2010 computer scientist. In the case of each of these disciplines there is a certain core knowledge (pertaining to useful concepts, theories, techniques, and methods) that is applicable to empirical science and without which empirical science cannot function.…”

Ontological realism and classification: Structures and concepts in the Gene Ontology

“…Others argue that concepts in an ontology have their referents in the thinking minds of scientists rather than in reality; even some biologists resist the extreme notion that mental concepts have no place in ontologies for biology. For various perspectives on this issue, see Merrill (), Dumontier and Hoehndorf (), Lord and Stevens (), and Smith and Ceusters ().…”

Anatomy of a cladistic analysis

“…Good, rigorous practice means annotating the underlying considerations for a particular solution extensively, laying out plausible alternatives and their effects of cladistic outcomes, signalling uncertainty and vagueness in one's assessments, and making explicit use of the contingencies that reflect the specific scope of an analysis. One should be bold in the practice of seeking the best‐fitting scope to codify homology; however, a good working criterion for stopping is when one senses that certain codings are no longer defensible in discussions with one's most highly regarded peers. Systematists should understand that sophisticated, parsimony‐contingent formulations of homology pose a challenge for ontology‐based representation and reasoning, particularly if the latter adheres to the tenet of homology neutrality. However, alternative schools of ontology design hold that ontologies should follow the representation and inference needs of particular scientific disciplines, and not vice versa (Merrill, ,b). Until such ontologies are developed, it is prudent not to water down the special linguistic contributions that systematists bring to biology by virtue of their parsimony‐contingent homology propositions and scope refinements.…”