tributed goals to further sub-goals. In particular, theory-theory should
deny that our conceptions of mental state types are provided by simula-
tion, as well as denying simulation any role in the initial attribution of
thoughts to others, and denying it any role in the prediction of action from
intention and the prediction of intention from desire.
4 Developmental studies
During the last two decades there has been much ingenious and revealing
research into the development of mind-reading in children. Ironically, it
was an attempt to assess whether chimpanzees have mind-reading abilities
which provided the spur to progress in the human case. Premack and
WoodruV(1978) advanced evidence which might be interpreted as in-
dicating that chimpanzees do have such abilities. Although there is con-
siderable evidence – from both studies in theWeld and controlled ex-
periments – which is at least suggestive of mind-reading abilities in some
other primates besides ourselves (particularly chimpanzees and gorillas),
the case remains far from conclusively made out (see Byrne and Whiten,
1988; Whiten and Byrne, 1988; Gomez, 1996; Povinelli, 1996). The prob-
lem with imputing mind-reading to chimps is that they might just be very
adept at exploiting knowledge of correlations between situations, bodily
cues (such as direction of gaze, bodily orientation and posture), and
behaviour – without having a capacity for thought about the contents of
another chimp’s mind. In comments on Premack and WoodruV’s seminal
article Dennett (1978e) and Harman (1978) pointed out that what was
required for a convincing demonstration of mind-reading was a test in
which an expectation was formed on the basis of attribution of afalse
belief.
The idea behind the false-belief test is that cases in which conduct is
appropriately related tohow things aremay be predictable simply on the
basis of knowledge of regularities linking situation and conduct. For
example, you may be able to predict that when there are ripe berries
around people will pick and eat them, because you have noticed they tend
to do so in that sort of situation. In general, where others have true beliefs
about a situation a correlation with the situation can be substituted for a
correlation with their belief about the situation. But if, instead, you predict
that they will go looking for berries in a place where you know there are no
berries, then that prediction is being made on the basis oftheir misrepresen-
tationof the situation, their false belief that berries are to be found there.
So far nobody has yet discovered a way of subjecting chimpanzees or
gorillas to a clear-cut false-belief test. (The closest anyone has come so far,
has been to devise anignorancetest – see Gomez, 1996; O’Connell, 1996 –
Developmental studies 91