Evidence to Hypothesis

This scheme asserts that an unsupported CLAIM in an argument map is in fact a hypothesis supported by evidence. The QUALIFIER is presumably. The DATA for the scheme is one or more pieces of evidence that imply that the CLAIM is supported by evidence. The WARRANT is scientific evaluation of empirical evidence.

The inference is an example of the Argument from Evidence To Hypothesis scheme for presumptive reasoning.

Notes

Supporting evidence should be provided as one or more arguments (such as references to published scientific papers) that support DATA.

There are some things to think about here. For example, is drawing a conclusion from an academic paper an inference (RA-node) or a rephrase (MA-node)? Is drawing such a conclusion a locution? Should the paper itself be considered a locution (or perhaps a locution event)? The details can be hidden under the covers of this template, so can make some assumptions for now, and firm up detail later in the light of experience.

Tentatively, we model support for DATA as an MA-node. The URI for a referenced academic paper should ideally be a DOI or something like it. A DOI can represent an I-node or an L-node (which is a sub-class of I-node). It can also be resolved to get any content, authors and dates that would be needed to construct a locution event. The scheme template just needs references as DOI's. Expansion into AIF could be part of some process that operates on the template (and might do different things in different circumstances).

It'll often be the case that drawing a particular conclusion from an academic paper is itself a presumptive inference (expert opinion say) that should be open to challenge. This gets into dialogue and dispute, and we need to think about how we can fit the dialogical modelling of AIF into scheme templates in general. For this specific scheme, we might have (or ask for) the speaker, and model that speaker as asserting the claim using a YA-node. This provides an anchor point for further dialogue.