Variation and change in Early Modern Dutch: Negation in Letters as Loot
Summary
In Early Modern Dutch, variation within writers is found between negative concord and single negation. This intermediate phase between negative concord and single negation is affected by both bottom-up change (the negative clitic started to erode due to functional redundancy) and top-down change (elite writers dropped the negative clitic completely). In literature, the frequency of the negative clitic has been described, pointing to an increase of the clitic in southern regions and in lower social classes. However, the theory of Jespersen’s cycle and finer mechanisms of generative syntax have not been involved in these studies. In the current corpus study, 17th century letters from people across regions and social classes were analyzed with respect the variation in negation, involving Jespersen’s cycle and notions from generative syntax. In particular, it was explored what types of grammars are shown between the varieties displayed by the writers of these letters. From a qualitative analysis of the data, it was found that many co-occuring varieties display different phases of Jespersen’s cycle in synchronic fashion: while some varieties display a conservative grammar exhibiting consistent negative concord (phase II), others display a progressive grammar containing consistent single negation (phase III). Furthermore, a dominant portion of the letters display a grammar with variation between negative concord and single variation. These findings contradict the assumed nature of Jespersen’s cycle, which portrays the phases of development in subsequent fashion. With respect to the syntactic features of the negative clitic – polarity features projecting PolP – four different derivational systems are found in at least one variety: (i) a high PolP as part of an extended CP, which indicates the C-parameter; (ii) a low PolP that attracts the finite verb to T, which results in V1 position in verb clusters; (iii) both high PolP and low PolP; and (iv) no PolP. Many letters with single negation show high PolP. From this observation I conclude that, although the negative clitic has been deleted in these varieties, its polarity features remain present. Hence the parametric hierarchy of the extended CP is not affected. Some of the varieties with single negation show evidence in favor of low PolP: the finite verb is attracted to V1 position of its verb clause in polarity-related sentences. This means that the negative clitic must have been present during the derivation, base-generated in low PolP and attracting the finite verb to T but was deleted artificially at surface structure. The result is then a dual grammar exhibiting progressive single negation at surface structure, but conservative negative concord at deep structure. This might have been an effect of the external upper-class prescription to drop the negative clitic on the one hand, and internal natural development on the other hand. A portion of the grammars with single negation exhibited total loss of the C-features. This is considered the most progressive grammar, showing Jespersen’s phase III at both surface structure and deep structure. With respect to the varieties showing variation between single negation and negative concord, it has been explored what factors drove the language user to insert the negative clitic. In literature, the negative clitic is assumed to express emphatic meaning: it emphasizes negative contrast or stress on the importance of the subject matter. For West-Flemish, the clitic has been proposed to emphasize the unexpectedness of the negation of a presupposition. In my dataset, I found the clitic to convey the latter function in all letters showing variation. A subset of these letters also showed the more general function of emphasizing negative stress. This extended emphatic function is captured in high PolP. I propose that, during language development, high PolP became available for the clitic to move to convey this emphatic meaning. Since this function is less specific, it also supports the assumed motive for the eventual loss of the negative clitic: functional redundancy. It follows that high PolP as part of the extended CP eventually loses ground. In the final step of this research, I applied the sociolinguistic variables of class (high/low) and region (Noord-Holland/Zeeland) to the letters. The most conservative variety, containing consistent negative concord, only appear in the low class of Zeeland, while the most progressive variety, containing consistent deletion of both the negative clitic and its polarity-related C-features, only appears in Noord-Holland. Since the prescribed grammar with single negation originated in the high class of Noord-Holland, it comes as no surprise that the most conservative grammar is found in southern regions and that the most progressive grammar is found in Noord-Holland. My data-analysis shows that in some varieties, the deletion of the negative clitic is merely a surface deletion, while in other varieties, which all happen to occur in Noord-Holland, both the clitic and its features are deleted.