11.05.2013 Views

Prefixation in English and Catalan - Departament de Filologia ...

Prefixation in English and Catalan - Departament de Filologia ...

Prefixation in English and Catalan - Departament de Filologia ...

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

. Ell encaputxà la Maria.<br />

He EN-hoo<strong>de</strong>d the Mary<br />

‘I put the hood on Mary’s head’ / ‘I covered Mary’s head with a hood.’<br />

(36) a. raiN ‘raft’ R<br />

b. enraiarV ‘to make a N’ [+c] [-c-m]<br />

a. Els homes enraiaren els troncs.<br />

The men EN-rafted the logs.<br />

‘The men created the raft out of logs’ / ‘The men tied logs together to create a raft’<br />

Although one could th<strong>in</strong>k that the semantics of the Vs <strong>in</strong> (34) <strong>and</strong> (35) looks quite<br />

dist<strong>in</strong>ct from each other, the division between the two may get blurred <strong>in</strong> some cases.<br />

This is ma<strong>de</strong> evi<strong>de</strong>nt by Vs like env<strong>in</strong>agrar ‘EN-v<strong>in</strong>egarV’, which can have both<br />

read<strong>in</strong>gs. That is, env<strong>in</strong>agrar can either mean ‘to soak someth<strong>in</strong>g (e.g. pickles) <strong>in</strong> N’ or<br />

‘to pour N over someth<strong>in</strong>g (e.g. food)’. (See Appendix B, section A/B for other words<br />

like env<strong>in</strong>agrar). 20 This suggests that the division between the two groups may not be<br />

l<strong>in</strong>guistically relevant after all, leav<strong>in</strong>g us with two semantic patterns for <strong>de</strong>nom<strong>in</strong>al enprefixations:<br />

one <strong>in</strong>volv<strong>in</strong>g a change of location (34, 35) <strong>and</strong> the other <strong>in</strong>volv<strong>in</strong>g the<br />

creation of the N (36).<br />

S<strong>in</strong>ce Vs of creation have the same semantics as en+A Vs, it is logical to<br />

assume that similar Θ-percolations take place. Although the [+c] role of enraiar can<br />

come from the conversion affix, the [-c-m] role has no apparent source, s<strong>in</strong>ce the N only<br />

has an R-role. However, Williams (1981b) argues on semantic grounds that the R-role<br />

could be <strong>in</strong>terpreted as a theme, a view I adopt, given that it is <strong>in</strong> accordance with the<br />

feature specification we would expect ([-c-m]) from the N. 21 Once the [+c] role <strong>and</strong> the<br />

re<strong>in</strong>terpreted R-role, [-c-m], are on the verbal no<strong>de</strong>, they will get <strong>in</strong><strong>de</strong>x 1 <strong>and</strong> <strong>in</strong><strong>de</strong>x 2,<br />

which will <strong>de</strong>term<strong>in</strong>e external <strong>and</strong> <strong>in</strong>ternal merger respectively (see 37).<br />

20<br />

Mateu (2001b, 2002) also groups location <strong>and</strong> locatum Vs together, <strong>and</strong> treats them as ‘change of state’<br />

Vs.<br />

21 Williams rejects the option of consi<strong>de</strong>r<strong>in</strong>g the R-role an external theme, because then there would be<br />

two themes <strong>in</strong> a s<strong>in</strong>gle Θ-grid. In this respect, Neeleman & Schipper’s (1992) remarks about Θ-role<br />

re<strong>in</strong>terpretation are illustrative. Accord<strong>in</strong>g to them, a Θ-role can only be re<strong>in</strong>terpreted as a role that is<br />

semantically close, <strong>and</strong> they also consi<strong>de</strong>r that that is the case for themes <strong>and</strong> R-roles.<br />

30

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!