And I agree with you that “sculptural group is not a number of statues”.
What you seem not to understand is my point that OSM tag artwork_type=statue which (albeit similar to the English word “statue”) does not represent “a statue”.
Instead, it represents something else, which is a category (or a set of possibilities, if you will) which includes “statue”, “statues”, “sculptural group” (and quite probably few other things too).
It is clearly and explicitly defined in such a way. We may dislike its “OSM wiki definition” and its divergence from English dictionary definition of a word “statue”, but we can’t change what it means at this point in time - it is what it is, and people use it like it is defined (and was defined for many years).
We may however use additional tags to clarify more precisely what that artwork_type=statue actually is. We may provide that clarification in several ways:
- by popular human-readable tag description=
This sculptural group depicts dead Jesus in a lap of Virgin Mary(advantage: popular and very well supported) - by using structural wikidata tags: wikidata=Q107338575 (sculptural set) + subject:wikidata=Q302 (Jesus)
- by inventing machine-readable OSM-specific refinement tags like
statue:people_count=2orartwork_type:animal_count=0(or some other scheme which you find less ambiguous and more to the point). Disadvantage of this method is that ATYL will be mostly ignored by editors and other data consumers unless it becomes very popular and/or significant effort is put into its advertisement. - combination of above
If I was interested mostly in human consumption, I’d likely stick only with first point of using description=* and call it a day. If I however really cared about a subject and thought machine-readable e.g. querying is warranted, I might combine it with second point. (I’d probably also be more invested in wikipedia, wikimedia_commons and wikidata by that point!)