User:Phlsph7/Ontology - Methods

Methods
Methods of ontology are ways of conducting ontological inquiry and deciding between competing theories. There is no single standard method; the diverse approaches are studied by metaontology.

Conceptual analysis is a method to understand ontological concepts and clarify their meaning. It proceeds by analyzing their component parts and the necessary and sufficient conditions under which a concept applies to an entity. This information can help ontologists decide whether a certain type of entity, such as numbers, exists. Eidetic variation is a related method in phenomenological ontology that aims to identify the essential features of different types of objects. Phenomenologists start by imagining an example of the investigated type. They proceed by varying the imagined features to determine which ones cannot be changed, meaning they are essential. The transcendental method begins with a simple observation that a certain entity exists. In the following step, it studies the ontological repercussions of this observation by examining how it is possible or which conditions are required for this entity to exist.

Another approach is based on intuitions in the form of non-inferential impressions about the correctness of general principles. These principles can be used as the foundation on which an ontological system is built and expanded using deductive reasoning. A further intuition-based method relies on thought experiments to evoke new intuitions. This happens by imagining a situation relevant to an ontological issue and then employing counterfactual thinking to assess the consequences of this situation. For example, some ontologists examine the relation between mind and matter by imagining creatures identical to humans but without consciousness.

Naturalistic methods rely on the insights of the natural sciences to determine what exists. According to an influential approach by Willard Van Orman Quine, ontology can be conducted by analyzing the ontological commitments of scientific theories. This method is based on the idea that scientific theories provide the most reliable description of reality and that their power can be harnessed by investigating the ontological assumptions underlying them.

Principles of theory choice aim to offer guidelines for assessing the advantages and disadvantages of ontological theories rather than guiding their construction. The principle of Ockham's Razor says that simple theories are preferable. A theory can be simple in different respects, for example, by using very few basic types or by describing the world with a small number of fundamental entities. Ontologists are also interested in the explanatory power of theories and give preference to theories that can explain many observations. A further factor is how close a theory is to common sense. Some ontologists use this principle as an argument against theories that are very different from how ordinary people think about the issue.

In applied ontology, ontological engineering is the process of creating and refining conceptual models of specific domains. Developing a new ontology from scratch involves various preparatory steps, such as delineating the scope of the domain one intends to model and specifying the purpose and use cases of the ontology. Once the foundational concepts within the area have been identified, ontology engineers proceed by defining them and characterizing the relations between them. This is usually done in a formal language to ensure precision and, in some cases, automatic computability. In the following review phase, the validity of the ontology is assessed using test data. Various more specific instructions for how to carry out the different steps have been suggested. They include the Cyc method, Grüninger and Fox's methodology, and METHONTOLOGY. In some cases, it is feasible to adapt a pre-existing ontology to fit a specific domain and purpose rather than creating a new one from scratch.