Goeman, T. (2000e) Structurele aspecten van de morfologie van voegwoordvervoeging: mogelijkheden en beperkingen, morfologisch gewicht en MCGG. In: deTier, V. & M. Devos & J. van Keymeulen (ed.). Nochtans was scherp van zin. Huldealbum Hugo Ryckeboer. Een bundel artikelen aangeboden aan Hugo Ryckeboer voor zijn 65e verjaardag. Gent-Deinze : Vakgroep Nederlandse Taalkunde van de Universiteit Gent-Van Daele, 2000, pp. 269-294. ●Structural aspects of the morphology of Comp-agreement; possibilities and restrictions, morphological weight and MCGG●
[on Comp-Agreement and aspects of enclitics]
Download here (zipped)
Comp-agreement; inversion forms: monosyllabic host+enclitic; non-inverted order
(1sing lèèv alternates with lèève and lèèw)
(3sing/3plur a-se < at-e and < at-se respectively)
Summary
Structural aspects of the morphology of Comp-agreement: possibilities and restrictions, morphological weight and MCGG (Monosyllable in Clitic Group Generalization).
This article is the third part of three on Comp-agreement; the two other articles were concerned with the research-history (Goeman 1997b) and with typological linguistic observations, together with proof of the existence in earliest Middle Dutch (Goeman 1997a).
Bibliographical entries from those two articles are thus not repeated here.
The subject of this article are the structural aspects that may play a role in Comp-agreement. First, the possibilities for Comp-agreement are given. Second, a number of restrictions on Comp-agreement. Last, some consequences.
I make a reasonable case for a generalization. Comp-agreement shows the same affix forms as a monosyllabic, or a verb that became monosyllabic in the position before an enclitic:
Monosyllable in Clitic Group Generalization (MCGG). This is not so strange, because conjunctions and relatives are prototypically monosyllabic.
One of these monosyllabic verbs is doen
You are celebrating your birthday tomorrow, do you? Yes/No+Agr+Clitic
Special attention is payed to enclitics that form a strong and unbreakable unit with the verb. These are forms of verbs that became monosyllabic, like auxilaries and preterito-presents that open up new possibilities for Comp-agreement.
We have to discern four structural linguistic aspects: semantic; syntactic position; morphological form; prosodic-phonological weight.
My approach is mainly lexical, prosodical and morphological; other approaches are mainly syntactical. My lexically oriented proposition is that the morphological realization of Comp-agreement exclusively is the morphological form of those enclitic combinations that offer a monosyllabic base of attachment to the enclitic.
This is in the first place the verb doen
We find monosyllabicity, morphological weight of host and enclitic and the temporal/modal semantics also as characteristics of subordinators.
A clear classification of pronouns in terms of stressable, unstressed and fused is enlarged by me with zero-realization. This state of affairs follows Burzio's pattern of forms according to morphological weight. The precondition is that a second position has to be relatively unstressed compared to its base of adjunction.
Anderson (1993) does not join clitics syntactically but does the job by 'stray adjunction' in second position to a base or host by means of the same sort of word formation rules that is needed for the morphological adjunction of inflection-material. One of his arguments is that cliticization and inflectional adjunction as well often need to have phonological information from the base.
A comparable proposal, albeit without Anderson's phonological motivation has been made earlier (Goeman 1980) for Comp-agreement and cliticization. A further piece of evidence comes from the data for the Wackernagel-position - described by Wackernagel himself typically in terms of prosody: unstressed clitics, unstressed verbs. In more recent linguistic theories as Distributed Morphology clitics do not have a status as morphological primitives, the clitic behavior follows instead from other characteristics and structures, among which: syntactic ones.
From that follows, that in morphology the clitic-induced affixes have to be explained by syntactic principles (like movement).
It seems that syntacticians have to find a solution for phonologically sensitive syntactic phenomena and that phonologists/morphologists must have a solution for morphological rules that are sensitive to syntax. The division of work needs to be cleared.
Note (2012): Hoekstra & Smits (1998) do not share the idea of the MCGG. According to them, Comp-agreement is not bearing tense information and they hypothize that Comp-egreement occours when the inverted forms of present and past tense are the same; they point to Frisian for the other situation; in cases that present tense and past tense inverted personal forms of verbs are different there do not exist Comp-agreement forms, where they are the same (as in 2nd person singular) Comp-agreement does exist.
Hoekstra, E. & C. Smits. 1998. Everything you always wanted to know about complementiser agreement. Proceedings of Western Conference on Linguistics (WECOL 1998).
Hoekstra, E. (2006) Review of SAND Volume I