Re: Mapwords
From: | Ian Spackman <ianspackman@...> |
Date: | Monday, July 21, 2003, 22:05 |
At 12:52 21/07/03, Christophe Grandsire <christophe.grandsire@...> wrote:
>En réponse à Ian Spackman :
>
>
>>Though as you say, with subclauses and the like it can get seriously
>>complicated. In practice it may be necessary to have subordinate mapwords,
>>though this detracts from the idea somewhat, I think.
>
>Not in my opinion. Sentence mapwords would just be impossible. The human
>memory has its limits. Moreover, sentence mapwords would mean that function
>words at different levels would get mixed up, something which is not really
>humanly possible (English has some raising of this kind, like "I want you
>to clean the floor" corresponding to "I want that you clean the floor",
>with the subject of the subclause raised to object position in the main
>clause, but this is about as much as you can do). We can handle a limited
>amount of complexity, and mixing clauses and subclauses is next to
>impossible for us.
Oh, I should have said that the whole-sentence mapword idea was intended
for an alien language (I almost brought it up last week when possibilities
for alien languages were being discussed). A *truly* alien language might
be unthinkable to us, but we can certainly come up with languages that we
can't handle, at least in real time!
I agree, for humans I'd expect clause level mapwords (and probably short
clauses at that, with a tendency to relative clauses rather than
complicated noun phrases within in main clause, for instance).
Ian
Replies