> Well, doesn't it depend on whether VAX MACRO kept the macros as
> high-level entities when translating them, or if it processed macros in
> the familiar way into instructions that sat at the same level as
> hand-written ‘assembler’.  I don't think this thread has made that clear
> so far.

The Multics case that I cited was definitely in the latter category.

There was no "translator". Effectively there were just two different

macro packages applied to the same source file. 

In more detail, there were very similar assemblers for the original

IBM machines and the new GE machines. Since they didn't have

"include" facilities, there were actually two source files that differed

only in their macro definitions. The act of translation was to supply

the latter set of definitions--a notably larger set than the former

(which may well have been empty).

Doug