pran, on 2019-November-09, 02:14, said:
which refers to the meanings attributable to a call, not to the suits involved as such.
In plain text this means that the possible meanings for the withdrawn call (had it been legal) must include at least every possible meaning for the replacing call.
Alternatively: The replacing call must not introduce (or add) any meaning that was not already included within the possible meanings of the withdrawn call.
Indeed. That is my point as well. Allow me to elaborate in this way.
For the purpose of reasoning, contemplate A, B,C all comparable calls. Further A is not equivalent to either B nor C; B is not equivalent to either A nor C; C is not equivalent to either A nor B.
Repeating. A, B, C being comparable are equivalent because by law they substitute without gain. And as you point out that reasoning is rubbish, and as I point out that is what the law is: substitute the three bullet points of L23 for A,B,C. QED
As for your antecedent regarding 23A2:
The author made an assertion as to what constituted a subset. I do not know if he was asserting how to untangle L23, but what I was doing was responding to his use of subset.
I think I can see my way to making sense of 'possible meanings attributable to the withdrawn call defining a set' but not 'a call defining a subset of possible meanings attributable to the withdrawn call '
To that end, I once had the epiphany of using Google to find such a usage. Six hours later I had the epiphany to never try it again. I guess I can't figure out how a call defines a subset of meanings. Now Richard Hills is a meticulous guy who meticulously proofed WBF2017. Perhaps he knows the meaning of the L23 language.