Characterizing ICEs, propositions, and images

[bdonohue29]

According to the definition of CCO_0000002 (Descriptive ICE), every entity in this class "consists of a set of propositions that describe some Entity." Similarly, CCO_0000003 (Directive ICE) is defined as "a set of propositions or images ... that prescribe some Entity".

I have a few questions and concerns about this characterization of ICEs.

Propositions in BFO? Within standard analytic philosophy, propositions are taken to be (among other things): the meanings of sentences; the objects of our mental attitudes, especially belief; and the bearers of truth or falsity. Stanford Encyclopedia of Philosophy has a nice entry summarizing the many common uses of "proposition" within analytic philosophy, as well as the roots of this idea of propositions in Greek and medieval philosophy.

But does the notion of proposition really fit in the BFOverse? I tend to think of ICEs as an alternative to propositions:

• Whereas propositions are typically taken to be self-standing abstract entities, ICEs generically depend on their bearers.
• Whereas propositions somehow stand in for or represent states of affairs (things capable of being true or false), ICEs are more granular "representational units" (cf. the 2015 "Aboutness" paper). For instance, the word "arm" can bear an ICE that is about the type arm or some particular arm, but it's not a proposition, and it's neither true nor false.

There's more we could pick at here (Do propositions as traditionally conceived stand in an "aboutness" relation, whether in a primary or derivative sense? Are propositions bound up with the metaphysics of "Fantology" criticized by Smith?) But I take it the concern is clear enough: should the major types of ICEs really be defined in terms of "sets of propositions"? I also don't see why this would be all that necessary to, say, differentiate Descriptive or Directive ICEs from other types. Those are really differentiated by "direction of fit" (describe vs prescribe). Thus, it doesn't look like the notion of proposition is doing any "work" in the definitions.

Propositions vs. Images? CCO_0000086 (Representational ICE) is defined as "A Descriptive Information Content Entity that consists of a set of propositions or the content of an image that represents some Entity." And again, CCO_0000003 (Directive ICE) is defined as "a set of propositions or images ... that prescribe some Entity".

I'm a bit confused about how to interpret this language of "propositions or (content of) an image". Should it be read as 1) a logical union (x is a set of propositions OR an image OR both), or does this mean 2) the content of an image IS a set of propositions?

The general thrust of the language suggests to me that propositions are being distinguished from images/image content here, and so accordingly we're defining these in an inclusive way as: a Directive ICE might be a set of propositions, or it might be (the content of) an image.

Then, however, there would be an issue with placing CCO_0000086 (Representational ICE) under CCO_0000002 (Descriptive ICE), as currently defined. The union dog or cat can't be a subclass of dog. That'd be logically backwards. Analogously, Representational ICEs (propositions or images) logically can't be a subclass of Descriptive ICE (propositions, full stop). Otherwise, the subclass would be broader than its parent class.

One might propose updating the definition of Descriptive ICE to similarly be "propositions or images," but then there'd be no differentia between Descriptive ICE and Representational ICE.

Alternatively, the intent might be: "the content of images are propositions," and thus the "propositions or images" language is just clarifying the scope (i.e., "propositions" includes the content of images). If so, then we run into the same issue as above: if image content = propositions, how do we differentiate Descriptive ICEs from Representational ICEs? Both would end up being ICEs that consist of propositions and describe some Entity...