case/adposition term

[Bill Croft]

Dear Martin et al.:

    In "Typology and universals" (both editions), I used the term 
"case marking" (NOT "case", which I tried to use in the old-fashioned 
way) to cover both bound and free morphemes to encode participant 
roles (as I would call the function: note that in the argument 
structure literature, "participant" is used very broadly, even 
including circumstantials depending on one's theory of participant 
roles). In "Typology and universals", I also used the term 
"relational (encoding strategy)" to cover about the same range of 
semantic relations as Christian describes for "relator". (I 
contrasted it to "indexical [encoding strategy]" - a term used to 
avoid the problems with the term "agreement".)

    "Case marking" vs. "case" is hardly an ideal solution but at least 
it makes the distinction. If a satisfactory new term catches on, that 
would be better.

Best wishes,
Bill Croft