While there has been a substantial amount of research into the morphosyntax of non-finite clauses in child language, especially root infinitives (RIs), far less is known about the interpretative properties of these sentences. There are two striking observations: First, non-finite clauses, such as RIs, English bare verbs, and others often have an interpretation that is different from their finite counterparts. For example, Dutch RIs, as in (1) typically have a modal interpretation expressing volition, direction or intention, rather than the present or past tense meaning that would be expressed by a finite clause uttered during the same stage (e.g. Wijnen 1997; Blom 2002). Similarly, the English bare verb, as in (2), can have ongoing meaning, in marked contrast to the adult or child finite present tense clauses (Deen 1997; Madsen & Gilkerson 1999; Torrence & Hyams 2002).

1. Papa ook boot maken
   Papa also boat make
   'I want Papa to make a boat' or 'Papa must make a boat'
   
2. Cowboy jesus wear boots
   'Cowboy Jesus (a doll) is wearing boots'
   (cf. Cowboy Jesus wears (=habitual) boots)
   

A second observation is that the interpretation of similar non-finite forms can vary across languages. Thus, Dutch (German, Swedish, etc.) RIs are modal, while Russian RIs most often have a temporal meaning (Brun et al 1998), denoting ongoing or past events.

In this talk, I will focus on the crosslinguistic differences (and similarities) in the interpretation of non-finite clauses in early language. I will take as a point of departure the observation that in bare participle clauses in child language, as in (3), the temporal interpretation of the clause is determined by the aspectual value of the participle. So, (3a) denotes a terminated/closed event, while (3b) denotes an ongoing/open event.

(3) a. Presa Checco campana
      taken Francesco bell
      'Francesco has taken the bell'

    b. He chewing it
       'He is chewing it'

I will propose that the temporal interpretation of all non-finite clauses in early language, including RIs and bare verbs in (1) and (2), has an aspectual basis: The crosslinguistic differences will be shown to follow from general principles of aspectual interpretation (especially Giorgi & Pianesi's 1997 Punctuality Constraint), in concert with the specific aspectual properties of the target language. The results cast a new light on the so-called 'aspect first' effects in finite clauses (e.g. Wagner 1998; Olsen & Weinberg 1999) and also suggest that children quickly converge not only on the morphosyntax of the target, a well-established finding, but also on the often subtle aspectual properties of the target.