Ontology just ain't linguistics, but there are a large number of detailed reasons we can give for this, including:creativesoul wrote:Why would ontological arguments not be required to pass this substitution test?
(A) The linguistic conventions of referring to one thing might not at all grammatically parallel the linguistic conventions of talking about something else. So if we substitute the term normally found in sentences about the thing in question in context A with the conventional terms for talking about the thing claimed to be ontologically identical to it (but not normally found in sentences in context A), the substitution might not work per the grammatical conventions of that language. It would require particular grammatical permutations that might necessitate neologisms (if the term in question doesn't conventionally exist in the language in question in a parallel grammatical form).
(B) As was one of the points of the lightning/electrical state example, if typical language-users of the natural language in question do not at the time in question accept the ontological identity, the resulting sentence is going to "sound funny" to them no matter what you do with the terms grammatically and syntactically. It won't sound right to them until they accept the ontological identity, the words are in a parallel grammatical form, and those typical language-users basically accept the terms as synonyms.
(C) As was another point of the lightning/electrical state example, while saying that lightning is an electrical state is correct, that doesn't amount to saying that lightning is "just any old electrical state." So when we make the substitution, if we use the term "electrical state," people who are prone to reading things like an Aspie--and many people are--will think it sounds funny because they read it as "just any old electrical state," which wouldn't be what anyone is saying. For the Aspie-prone, we'd need to spell out just what sort of electrical state in some detail, although then the sentence is going to sound funny to typical language speakers of that natural language because obviously "lightning" is a lot shorter than specifying just what sort of electrical state in some detail. In other words, the syntactical difference, even if the longer phrase involving "electrical state" is gramamtically parallel, will make the sentence sound funny to most folks.
(D) If an ontological identity statement uses terms that are conventionally closely associated with each other in some context, the substitution might not work unless work is done to more than just one of the terms in the sentence in question, where some terms might require an "analytic description" substitution (that will result in the sort of syntactical difference described immediaely above, in (C)). I know that reads confusingly, so let me give an example. I run into this frequently when the ontology of time comes up, because in my view, time is ontologically identical to motion or processual change. So what always ends up happening is some physics-oriented person who can't think much beyond what I'd call the script they've memorized will go, "That doesn't make sense because then acceleration, say, would have to be defined as 'the rate of time of velocity per unit of time.'" (Conventionally, acceleration is "the rate of change of velocity per unit of time.") Well, what's going on there is simply that we're measuring one motion/change against another. What's going on ontologically with "per unit of time" is that we're looking at the motion/change of something like a clock--a digital meter changing numbers, or a second hand moving--and comparing the change of one thing--velocity (which is speed + direction) to the change of another thing--those changing numbers on the digital readout, or that changing second-hand. Additionally, the substitution of "time of velocity" for "change of velocity," as well as "per unit of change (in a clock)" for "per unit of time" both have the problem outlined in (B) above.
So one thing--language--is about conventional ways of referring to and talking about things, as well as conventional ways of thinking about/conceptualizing things, while another thing--ontology--is about what things really are, where that might not at all substitutionally match conventional ways of referring to and talking about things, OR conventional ways of thinking about or conceptualizing things. But things aren't what they really are just because people don't conventionally talk about them or think about them that way, or just because it might sound funny to try to substitute one thing for another in a sentence.