A definition for a class introduces a class scope rather than doing that by a declaration that is not a definition

[basic.scope.class#1]

Any declaration of a class or class template C introduces a class scope that includes the member-specification of the class-specifier for C (if any). For each non-friend redeclaration or specialization whose target scope is or is contained by the scope, the portion after the declarator-id, class-head-name, or enum-head-name is also included in the scope.

The declaration of a class or class template does not definitely specify that the declaration must be the definition, which could be inferred from [dcl.dcl#dcl.pre-5]

In a simple-declaration, the optional init-declarator-list can be omitted only when declaring a class ([class.pre]) or enumeration ([dcl.enum]), that is, when the decl-specifier-seq contains either a class-specifier, an elaborated-type-specifier with a class-key ([class.name]), or an enum-specifier.

That hints struct C; is also a declaration of class C, however, the declaration does not introduce a class scope. I think we should reduce the extent of "declaration" to use "definition", which means only a class-specifier can introduce the class scope.

Start with [basic.scope.class#1] and compare its use of "declaration" with [dcl.dcl#dcl.pre-5] and the class declaration rules. Check how a forward declaration such as struct C; differs from a class definition, then update the wording so only the intended declaration introduces a class scope. Done means the clauses are consistent and no longer imply that a forward declaration creates one.