Print

Print


On Fri, Jan 18, 2013 at 11:46 AM, Gary Shannon <[log in to unmask]> wrote:
> But I still stand by my insistence that there are not an infinite
> possible number of USEABLE sentences in English based on my engineer's
> limit of 45,000 words for the longest possible USEABLE English
> sentence. With a hard and fast upper limit on sentence length the
> number of possible sentences in English is finite, and even though it
> might be infinite "in theory" it cannot be infinite in practice.

Can we make a distinction between *potential* sentences of a given
language, those allowed by its grammar and semantics (which can
certainly be arbitrarily long if not infinitely long), and *actual*
sentences, that is, sentences produced at some time and place by some
fluent speaker of the language?  Maybe "potential" and "actual" aren't
the best possible words, but the distinction seems important, however
we describe it.  Intermediately, one may wish to add a third category,
of *realizable* sentences, which are not only grammatically and
semantically valid, but practically possible to say even if they
haven't actually been said or written.  I think what Gary is trying to
delimit with his estimate of 45,000 words for the longest sentence one
could utter in twelve hours or so is an upper bound on this
"realizable" category.


-- 
Jim Henry
http://www.pobox.com/~jimhenry/
http://www.jimhenrymedicaltrust.org