This dog is to guard the tobacco.
The dog who is to guard the tobacco arrived.
The dog to guard the tobacco arrived.

This analysis brings together what Huddleston & Pullum (2002) describe in terms of a ‘quasi-modal’ use of the verb be (p.113) and what they consider a ‘modal meaning’ associated with certain non-wh infinitival relative clauses (p.1068). The be verb has all of the ordinary auxiliary properties, shares with the modals the further property of only appearing in finite forms. H&P further characterize its meaning as ‘modal’, without further elaboration. In the ERG, the modal semantics is modeled with the eventuality relation, and this is associated with the infinitival clause, which may appear as the complement of (semantically empty) be.

Treating these two phenomena as semantically equivalent is motivated by the following pair of examples, which we judge to be paraphrases of each other and to which we assign the same semantic representation.

• The plumber to fix the sink is coming at 10.

• The plumber who is to fix the sink is coming at 10.

We differentiate this use of infinitival clauses from the superficially similar instrumental relatives like the following:

• Money to buy the dog arrived.

In instrumental relatives, as opposed to quasi-modal infinitivals, the head noun fills an instrumental role with respect to the relative clause.

  eventuality[ARG1 h1]
  h2:[ARG0 e]

  { h1 =q h2 }

• Is eventuality the right pred to use here, or should this be assimilated to something more clearly modal?

ERG 1212 (but there actually is no qeq there)

Huddleston, R., & Pullum, G. K. (eds) (2002). The Cambridge Grammar of the English Language. Cambridge: Cambridge University Press.

• ErgSemantics main page

• Inventory of semantic phenomena (to be) documented

• How to cite this work