This looks like an excellent start!

Some suggestions toward addressing the character-encoding issue:

 * Change the API to distinguish between byte strings and character
   strings. (I think C code is as likely to need one as the other).

 * Where "char *" is used for strings (e.g., "expected_explanation" for
   a type error), define it to be an ASCII or Latin-1 encoding (I
   prefer the latter).

 * For Scheme characters, pick a specific encoding, probably one of
   UTF-16, UTF-32, UCS-2, or UCS-4 (but I don't know which is the right
   choice).

An additional request:

 * Distinguish between mutable and immutable strings, particularly in
   checking argument types. (C code that intents to mutate an argument,
   for example, should require a mutable one and reject an immutable
   one.)

Matthew