This research notes gives details of an atypical response to a 'give a
number' task by a 34-month-old child. The child clearly demonstrated
that he understood the word 'two' to refer to an object rather than to
a quantity. This incident offers support to Fuson's (1988) theory of
the functionally specific initial acquisition of number word meanings
and suggests that Gelman & Meck's (1983) 'principles first' account of
the development of counting gives insufficient weight to the role of
language and social interaction in the development of early number und
erstanding. The incident also shows that our knowledge of this area of
development would benefit from a greater emphasis on longitudinal stu
dies, studies of social context and studies of individual differences
in the processes of acquisition.