Grice said that poets are allowed to 'flout' the conversational maxims --
if they have a 'conversational implicature' in mind. His favourite couplet
was by William Blake,
"Never seek to tell thy love, love that never told can be".
For Grice notes that 'love that never told can be' can be interpreted
prosaically as "love that can never be told", or, given Blake's sophistication
as a poet, as a claim about a love that when told can cease to exist.
We are considering a similar outburst, one line now, by a somewhat later
poet, W. H. Auden, brought to the forum by J. M. Geary:
i. We must love one another or die.
Geary sees this as "an affirmative advocation", even if he guardedly goes
on to say that "the consensus
is that we all would rather die" than share our world or love one another.
Omar K. proposes a conditional treatment:
Auden's casual remark "should be taken", O. K. notes, as, 'If we do NOT
love one another, we will die', again, guardedly adding: Auden does not
"specify what will happen if we do love one another."
McEvoy notes that if Auden does not specify this (or 'explicaturise') it's
because he is implicaturizing. He calls this Auden's 'trick', which may be
missed on some.
In McEvoy's words:
"While "what will happen if we do love one another" is not made explicit,
it is implicatured that if we do love one another, we will do better on the
death front", i.e. not die.
McEvoy notes that 'we must die' has different usages (he speaks of
'senses'). Under one interpretation, it is possibly tautological. Auden does
NOT
mean the utterance in its tautological interpretation. To symbolise that,
McEvoy needs to retort to the use of time subscripts, t1 and t2.
The second disjunct of Auden's statement,
"We must die."
thus should not be interpreted as, to use McEvoy's words, "in the long run
we will die no matter what we do", i.e. in the scenario to hand, whether
we love one another or not.
There is a time, that McEvoy symbolises as 't1', "after which each of us
will be dead."
Auden does not mean "t1"
He means a different time that McEvoy symbolises as "t2"
where
t2 < t1
a time "prior to t1 when we are still alive and when we would NOT be alive
(i.e. but dead) if we did not love one another.
Auden's use of "must" complicates the issue, and Geary's paraphrase
simplifies it by omitting it. In Kripke's modal system, "must" gets symbolised
by
a square:
□
Thus, for
i. We must love one another or die.
there are at least two possible expansions:
ii. We must love one another or we do die.
(i.e. where the modal operator □ does not have scope over 'we do die').
Cfr.
iii. You must bring a female to the party or dress as a clown.
----
iv. We must love one another or we must die.
(i.e. where it does).
The use of quantifiers may simplify the issue. Since 'love one another' is
what Russell has as a 'reciprocal' (in his theory of relations, 'a
paradigm of analysis') we can use x and y for individuals, and
L
for the dyadic predicate 'love'
and
D
for the monadic predicate 'die'.
In symbols:
v. (□(x)(y)L(x,y) & L(y,x)) v (Dx & Dy)
A note on the first disjunct may be in order. To have 'universal' scope we
need universal quantification. It should not necessarily be interpreted
(sollipsistically) alla Queen Victoria:
vi. We must love each other or die (cfr. her "We are not amused").
In Queen Victoria's case (she used the plural for herself), to love one
another, symbolically, becomes
vii. L(v, v).
i.e. Victoria loves Victoria.
But it cannot be interpreted merely as per "Auden in love" to mean
viii. Wystan must love Chester and Chester must love Wystan.
Rather, Auden is referring to humanity as a whole, we grant, hence the x
and y (Since McEvoy's temporal indexes need also be under the range of
quantification, this has to be considered in the final formula).
i.e. it is necessary that for x and y, either x and y love one another or
x and y die.
Now we add the temporal subscript t1 (I prefer to use t1 to a time prior
to t2).
ix. (□(x)(y)L(x,y) & L(y,x)) v (∃(t1<t2)(Dx & Dy)
But of course, it's best to treat the modal operator -- our square -- to
apply to BOTH disjuncts:
x. □((x)(y)L(x,y) & L(y,x)) v (∃(t1<t2)(Dx & Dy))
The symbolisation of
xi. We must die.
as
xii. Dx & Dy
makes it manifest that x and y need not die at the same time, provided t1
is prior to t2 (in fact, Wystan and Chester loved one another but Wystan
died first).
So, the temporal indexes need to apply individually to each conjunct
xiii. □((x)(y)L(x,y) & L(y,x)) v (∃(t1<t2)Dx & (∃(t1<t2)Dy))
We know proceed alla Omar K. to 'eliminate' the disjunction in preference
to negation and conditionalisation, and we get what might by Auden's final
interpretant. A note on the negated conditional may be in order. As Omar
notes, the colloquial way to take this is "if we do NOT love one another".
Even if Auden writes, "We must love one another". Where did the Kantian 'must'
go?
The way out of this apparent paradox is to take a possible-world semantics
for "□": a sentence using 'must' applies to all Leibnizian possible worlds.
If we colloquially tend to use the indicative, it's because we are
sticking with the actual world, rather than all possible worlds. This is what
Witters may call a 'deep grammatical' feature of the final formula. We make it
explicit for adding quantifiers for possible worlds:
xiv. □(((∀w)(x)(y)L(x,y) & L(y,x)) v (∃(t1<t2)Dx & (∃(t1<t2)Dy))
and then proceed with negating the antecedent and getting rid of the 'or'
which is thus deemed NOT a primitive operator (even if Auden does use it):
xv.
~□(((∀w)(x)(y)L(x,y) & L(y,x)) ⊃ □(∃(t1<t2)Dx & (∃(t1<t2)Dy))
which might be more or less what Auden was having in mind (or 'meaning').
Cheers,
Speranza
------------------------------------------------------------------
To change your Lit-Ideas settings (subscribe/unsub, vacation on/off,
digest on/off), visit www.andreas.com/faq-lit-ideas.html
