Appendix I: Invited Talk 
From lexical-aspectual components to syntax. 
Nomi Erteschik-Shir & T.R. Rapoport 
Dept. of Foreign Literatures and Linguistics 
Ben-Gurion University of the Negev 
Beer Sheva 84105, Israel 
We present-a-fheory:o~f~'syntaetie<.projeetion,~: arguinff.that*aspeetual:syntaefie~structure is 
directly projected from a limited inventory of iexical semantic components. Only lexical 
meaning components, necessary in any case for interpretation, can project structure. In our 
approach, the lexicon is not a separate level of representation with its own restrictions; we 
offer a structural account for what has hitherto been accounted for in terms of argument 
(number) specification, theta-roles, and mapping conditions. 
Each lexical semantic component projects a V-complement structure, the complement 
typically realizing the categorical type of the component, as shown: 
V V V 
/ \ / \ / 
V N V A V 
I I I VM VS VL 
M = manner/means/instrument 
\ S = state 
P L = location 
The full projections correspond roughly to the Vendler-Dowty aspectual classes: 
process/ change-of-state/ 
activity achievement 
cause + change of state/ 
accomplishment 
V V 
I \ I \ I 
D V D V 
I /\ /\ I \ 
Jane V N the vase V A 
I I I I 
laugh \[laugh\]M break \[break\]s 
\[break\]. 
V 
/ 
D 
I / 
Jane V 
/ 
\[break\]M 
I 
\ 
V 
\ 
V / \ 
D V \ / \ 
the vase V A 
I I 
break \[break\]s 
Projection is free. Transitivity thus follows from the number of projected meaning 
components and crucially from the availability of an interpretation for each component in a 
particular projection. 
Unprojected components are aspectually defocussed (backgrounded) and are interpreted as 
modifiers. Such aspectual defocussing also accounts for the variety of constructions in which 
different verbs can occnr, e.g. the contrast between break and cut, which does not allow *The 
cake cut. 
We demonstrate how a combination of semantic component projection and aspectual focus 
accounts for the possibilities of verbal distribution and interpretation. 
272 
