The Psychological Foundations of the Grammar 53
To illustrate, imagine a man who has never tasted a papaya. Based upon
his community membership, he is able to recognize that papayas are
fruit and therefore, he assumes, taste sweet. This is his background belief.
Out of curiosity he buys two papayas and brings them home. Upon seeing
the papayas his partner comments that she adores their sweet taste. Based
upon the partner’s utterance the man’s belief that papayas are sweet is
strengthened. He has a shared belief with his partner. He offers her a
papaya. He takes a bite out of the other; he knows that it is sweet. The
woman takes a bite. It is clear that the man has a shared belief that both
he and his partner believe that papayas taste sweet. But it is not entirely
clear that the man and his partner hold shared or indeed mutual know-
ledge that papayas taste sweet.
Scholars who state that mutual knowledge is a prerequisite for successful
communication argue that mutual knowledge is feasible between speakers
and hearers. Others disagree. The primary objection^5 found in the liter-
ature to both the existence and the necessity of mutual knowledge is the
Mutual Knowledge Paradox. The paradox states that in order to be mutually
certain that they possess mutual knowledge, a speaker and hearer must
carry out an infi nite series of regressive checks to confi rm their mutual
knowledge. Each check takes a fi nite though miniscule amount of time. As
communication cannot take an infi nite amount of time it is impossible
for speakers and hearers to carry out the infi nite series of regressive checks
required to secure mutual knowledge (Clark and Marshall 1981: 15;
Sperber and Wilson 1995: 15–21).
Scholars who believe in the essentiality of mutual knowledge posit two fi xes
which they claim circumvent the paradox. The fi rst truncation heuristics is
that speakers and hearers do not engage in an infi nite series of regressive
checks to secure the mutuality of their knowledge. Rather speaker/hearers
only check regressively to a certain fi nite level. Bach and Harnish (1979: 309)
Table 3.2 Classifi cation of knowledge/beliefs in terms of certainty
100% self-certainty
of truthLess than 100% self-certainty
of truthKnowledge Belief
100% certainty shared with hearer Mutual Mutual
Less than 100% certainty shared
with hearer
Shared SharedLess than 100% certainty shared
with hearer
Common/Background Common/Background