The LINGUIST List is dedicated to providing information on language and language analysis, and to providing the discipline of linguistics with the infrastructure necessary to function in the digital world. LINGUIST is a free resource, run by linguistics students and faculty, and supported primarily by your donations. Please support LINGUIST List during the 2016 Fund Drive.
|Title:||The Lexical Level of Nominal Compounds of the Noun+Noun Type in Contemporary English||Add Dissertation|
|Author:||Stanimir Rakic||Update Dissertation|
|Email:||click here to access email|
|Institution:||University of Belgrade, Faculty of Philology|
|Abstract:||Introducing lexical phonology Kiparsky (1982) has proposed
the following three lexical levels in English:
Level 1: the first order affixes and irregular inflection
Level 2: the second order affixes and compounding
Level 3: regular inflection
The analysis of the corpus of 6,270 English N+N compounds excerpted from
Longman Dictionary of Contemporary English (2003) should shed some light on
the distribution of affixes in the compound components and on the question
whether English compounds can be assigned to a particular lexical level.
Morphological analysis should show the distribution of affixes in both the
first and the components of compounds because, as of yet, the distribution
of plural forms in the second components of compounds has been gratuitously
neglected. The dissertation addresses the question whether two lexical
levels sufficient, or the third one must also be introduced?
As far back as the eighties Selkirk (1982) noted that the examples such as
parks commissioner and programs coordinator violate level ordering in (1).
To avoid such counterexamples some linguists have suggested that
compounding and regular inflection both belong to the second lexical level
(Selkirk 1982, Borovsky 1986, Katamba 1993). The analysis of compounds
shows that only very frequent plural forms may appear in the first, as well
as in the second component of compounds. The concurrence of the genitive in
‘s, cited by Hayes et al. (2002) and Murphy and Hayes (2010) as a decisive
factor, cannot explain these relations. There is only a small number of
compounds with plural forms in the first component. This holds true even in
cases where the first components denote inanimate objects with which the
genitive in 's is normally not compatible. Similarly, there is also a
relatively small number of compounds with plural forms in the second
component although no concurrence of the genitive in 's is possible there.
This implies that there must be some other restrictive factors different
from the genitive in 's.
Examination of the compounds shows that only very frequent regular plural
forms may appear in the first or second component. Most often, these are
the nouns whose plural forms exceed the threshold of six examples in a
million of words postulated by Alegre/Gordon (1999). The nouns whose
frequency is lower than this threshold can sporadically appear,
particularly as second components. The examples of such nouns suggest that
the threshold may be lower for the second components than for the first
components as the plural of the first component must be licensed by the
meaning of the second component. There is therefore an asymmetry in English
between the first and the second component because plural forms of the
first component must fulfill one condition more than the plural forms of
the second component.
We can therefore conclude that N+N compounds in English may be assigned to
the second lexical level, and regular inflection to the third lexical
level. This ordering however applies only to the lexical items which are
not frequent. Very frequent lexical items are lexicalized as a whole, and
are not parsed into constituent elements. The ordering of these items does
not depend on the classification of affixes into different classes. The
suggestion of some linguists that regular inflection belongs to the second
lexical level only covers up these fundamental relations. This insight
resolves the problem of plural forms in the theory of level ordering, but
at the same time strictly limits its domain of application.