| Letterals words are pronouns with the mnemonically useful characteristic that they contain a letter (or some such thing) that ties in with the noun being pro'd. Any Lojban word followed by {bu} forms a new litteral, which may not have that connection or has a more remote one (Would {pabu}, aside from being malglico, be usable for the indefinite "one", as in "one might think that"?) through some other character (? mark? symbol?). (This contrasts somewhat with the Refgram which uses letteral words to refer to letterals, i.e., {a, b, c,...} and for which the pronominal use is secondary and derivative. That is, the Refgram regularly uses letteral words in spelling and creating acronyms, which, under the usage above, would turn out to be variables rather than names, since a string of letteral words is also a letteral word. But then it has to take this whole line back and say that, outside of spelling and acronyms (somehow contextually marked, one supposes) one can't use letteral words to talk about letterals, since this will conflict with the use as pronouns -- a practical decision, no doubt, but not totally coherent. The Refgram also notes a tendency to confuse letterals and letteral words -- use and mention -- which may be relevant here. ) {me'o}, whose first function is to convert a mathematical _expression_ (a formula?) into a sumti (how exactly did the formula? function before?), also converts letterals, which are already sumti ("sumti" here means somethiing that can be a sumti, not something that in fact is one -- roughly "proper noun or description"), into names for the [character? mark? symbol? letter?] with which they are associated (what the cmavo list mistakenly says they mean originally). Well, not quite, since at least the {bu}-formed letteral words are "more abstract" than (characters? marks? symbols? letters?). I am not quite sure what this means; it seems to go beyond token-type contrasts, even of the most general sort, but may not. I THINK it means something to do with functional equivalence (arguably still a token-type matter, like typefaces) {me'o zo'obu} would "humor" marker, whether a smiley-face or the Chinese character for "HEE-haw" or... And so, presumably, a Tengwar version of Lojban spelling might well contain a hook that was referred to in context as {me'o abu}. And, of course, we now have the problem of what to do about letterals when they are used in mathematical expressions: does {me'o} turn them into names of letters or mathematical sumti as unevaluated mathematical entities? Maybe we have to go yet another step to get to the names of letterals themselves? Test question: when you write out what is said as {abu prami by} is the written form exactly the same or is it {a prami b}? Why? What about the Lojban for "CIA"? I find this all rather muddled; can someone please clarify it a bit? |