I should from posting when it is night and I am half asleep...

So I would like to see clear semantics for let(rec)-syntax/splicing that is in accordance with the rest of the R7RS. Possibly, let(rec)-syntax (non-splicing) won't be implementable with let(rec)-syntax/splicing then, meaning that we actually have too different binding constructs neither of which is expressible in terms of the other.

This is wrong. We can actually express let(rec)-syntax in terms of let(rec)-syntax/splicing but not with the transformation

(let(rec)-syntax <bindings> <body>) => (let () (let(rec)-syntax/splicing <bindings> <body>)

given above.

The correct transformation is

(let(rec)-syntax <bindings> <body>) => (let(rec)-syntax/splicing <bindings> (let () <body>))

This works even in the edge cases where <body> contains definitions redefining let(rec)-syntax/splicing, and irregardless of the particular semantics of let(rec)-splicing in these edge cases.

Note that I assume that let is inserted by a hygienic macro transformer and not shadowed by any of the <binding>s.

Actually, redefining let(rec)-syntax/splicing in the body would work for the transformation

(let(rec)-syntax <bindings> <body>) => (let () (let(rec)-syntax/splicing <bindings> <body>)

as well if let(rec)-syntax/splicing is hygienically renamed as well.

Marc