The articles by Glenberg and Robertson and by Roth try to advance radical p
rinciples about reference and meaning. Glenberg and Robertson show that ind
exing is required to disambiguate verbal instructions to the point where th
ey can be carried out. Roth's data show that even young students understand
that an indexical expression that ostensibly refers to an analog represent
ation in actuality refers to its referent. These are valuable insights, but
they do not add up to a radical new view of language.