[FOM] Why inclusive disjunction?

Neil Tennant neilt at mercutio.cohums.ohio-state.edu
Thu Jan 11 16:08:18 EST 2007


Back in 1975 there was a conference in Canterbury, Kent, where the last
stage of the breakfast cafetaria line had a sign saying "Cereal or orange
juice". I put both a glass of (rather synthetic-tasting) OJ and a bowl of 
(rather tired-looking) All Bran flakes on my tray, and the check-out lady
upbraided me. My appeal to inclusive "or", and the authority of several
amused logicians in the line behind me, was to no avail. Exclusive "or"
prevailed. I think I opted for the juice.

This is analogous to the form of the legal provisions brought to the
list's attention by Francis Davey. "The court may do X if it appears to
the court that either A or B or C", as Davey points out, invites the
inclusive reading, since one can easily see that (say) A, B and C are
perfectly compossible.  But of course counsel hoping to get the court to
do X would be well-advised to focus on establishing whichever of A, B or C
is most plausible on the evidence available. To aim to do more would be
belaboring the point, probably to the court's vexation. Though any one of
A, B or C will be sufficient, counsel ends up giving the
non-logical but reasonable bystander the impression, however unintended,
that the disjunction is to be interpreted as exclusive.

In the breakfast line, it is clear that the caterer, in saying "A or B",
intends it to be understood that she is obliged only to do a local
minimum to make "A or B" true. The greedy customer (or starving student,
as I was back then) who tries to get both A and B is trying to make the
caterer part with more than a bare minimum in order to make good on her
promise "[You may have] either A or B".

Clearly there are both semantic and pragmatic factors at work here.
Perhaps inclusive "or" is a good default semantic interpretation, and
Gricean considerations should be brought in whenever that interpretation
appears to be incorrect.

The Romans, of course, marked the distinction by using "vel" and "aut";
too bad we haven't followed suit.

Neil Tennant




More information about the FOM mailing list