The meaning of -ize derivatives

The theoretical framework: Lexical conceptual semantics

For the semantic analysis of derived verbs in this topic, I will use the theory of lexical conceptual semantics developed by Jack-endoff (1983, 1990, 1991). The reason for adopting this framework is that it is particularly useful for the description of the semantics of verbs in general and the analysis of verbal derivation in particular (e.g. Lieber and Baayen 1993, Lieber 1996). For readers unfamiliar with this theory, this topic provides a short introduction to the main concepts and technology of lexical conceptual semantics relevant for the investigation to follow.¹ Some problematic or controversial aspects of the theory are ignored in this necessarily rough overview but will be taken up later, as our investigation proceeds.

The theory of lexical conceptual structure as developed by Jackendoff assumes that there is a form of mental representation, so-called 'conceptual structures', that is common to all languages and that serves as the 'syntax of thought' (Jackendoff 1991:10). These conceptual structures can be described in terms of semantic primitives on the one hand and principles according to which these primitives can be combined on the other. The primitives of the theory are at least major conceptual categories such as Thing, Event, State, Property, Path, or Place, which are combined by functions that operate on these primitives. Verbal meanings, for example, can be represented by hierarchically organized structures containing semantic functions like GO, CAUSE, TO, and arguments on which these functions operate, namely the major conceptual categories just mentioned. Let us see how this works. Consider the sentence in (1), which corresponds to the conceptual structure in (2) (see Jackendoff 1991:13):

GO is a two-place function that maps a Thing (here: 'Bill') and a Path (here: 'to the interior of the house') into an Event consisting of the Thing traversing the Path (the expressions BILL and HOUSE are not decomposed for the present purposes). The entire expression in (2) can be read as "Bill traverses a path that terminates at the interior of the house". The lexical entry of the verb go can (roughly) be represented in this theory as in (3):

The subscript A indicates the argument status of Thing and Path for the function GO. Such arguments must be expressed by a syntactic argument, with the linking of conceptual arguments to syntactic arguments being done by general principles of linking, which need not concern us here any further (see Jackendoff 1990: chapter 11 for discussion). Thus, in a sentence like (1) the empty argument slots in the lexical conceptual structure (LCS) of go are filled by the appropriate arguments [Bill] and [into the house]. The preposition into (with its complement [the house]) provides an appropriate Path argument for the function GO because the LCS of this preposition is a Path, and [the house] is an appropriate argument for the function IN, which is also part of the LCS of into. Consider (4):

As we will see, the LCSs of -ize verbs can be represented in a similar fashion. In essence, the theory of lexical conceptual structure is a theory of semantic decomposition, a fact that is particularly useful for the description of derived verbs, because it allows the teasing apart of the respective contributions to the meaning of the derivative by the different morphological elements that are combined to derive a complex form. I believe that the main insights of the analysis to follow do not hinge on the formalism of this particular decompositional theory, but that they could equally well be expressed in the terminology of other existing frameworks.

We are now in a position to turn to the problem at hand, the meaning of -ize derivatives.

¹ For a short first-hand introduction to the basic concepts and functions used in lexical conceptual semantics, see Jackendoff (1991:10-15).