|
At the same time that the new Frankish inhabitants were picking up these local words (discussed in the previous chapter), the native Frisians were shifting to Hollands, and in the process leaving traces of their original language in the phonology of that Dutch dialect thereafter spoken in that area. (footnote 12) This is reflected in words that show a long a as a descendant of original Germanic *ai instead of the expected long e, such as Hollands klaver < *klai-bro-n 'clover', vs. kle-ver in the Middle Netherlandic dialects. Also this group is Hollands ladder 'ladder', compared to standard leer. Next there are words that show monophthongization of Germanic *au to long a instead of ou or long o, such as Vlaming < *flauming 'Fleming' or baken < *baukna- 'beacon'. (The long a was later shortened in closed syllables, hence the modern pronunciations of words such as ladder.) In some dialects of Noord-Hollands, the long a then shifted to long o, which was taken up into ABN in words such as moot 'slice, portion', toon 'tone', and flikflooien 'to cajole, wheedle'. Another Frisian characteristic is ie as i-umlaut of long o, long u, or fronting of long a. These words often exist alongside those showing 'proper' Dutch developments, so that there is vliering 'attic' beside vloer 'floor' (where the slightly divergent forms serve to distinguish slightly divergent meanings), and in Middle Dutch lieke 'leech' beside leke and lake. Some influences can be traced to either Frisian or Saxon (another Ingvaeonic language, which abuts on and has influenced the development of ABN). This is the case with long * u- > /u/ (spelled <oe> in ABN) instead of the expected uu or ui, in words such as poes 'cat', snoet 'snout', boer 'farmer'. The Ingvaeonic loss of n before fricative with compensatory lengthening shows up in *unð > uð, which occurs in both Frisian and Saxon as mu-th 'mouth.' This appears in the suffix -muiden in the names of many towns, especially those located at the mouths of rivers: IJmuiden, on the North Sea coast in Noord-Holland and hence ancient Frisian territory; but note also Genemuiden, near Zwolle in solidly Saxon territory, and Dixmuiden (Dixmude/ Diksmuide) in West Vlaanderen, Belgium. The influence of Hollands on ABN did not end with its official codification in the 19th century, but continues into present times. A recent change in ABN is, I claim, also traceable to originally Frisian interference. It is a characteristic of Hollands dialect that word-initial v- and z- are devoiced (Donaldson 1983:13). In Old Frisian, as in Old English (footnote 13) , the f/v and s/z distinction was not phonemic, and the unvoiced partner always appeared word-initially. This is not the case in Middle Dutch, or even all modern Dutch dialects. Compare modern English foot and modern Dutch voet, which is spelled with a v-, and so pronounced by speakers of the more southern dialects and by most educated older people. Under the influence of Hollands, an unvoiced pronunciation is becoming acceptable for this and similar words. Although older grammars proscribe it, recent Dutch textbooks for foreigners even encourage a "half-voiced" pronunciation in order to help the student sound less foreign (Shetter 1988:14). Devoicing of initial z-, however, is still generally unacceptable in ABN, in spite of Hagen and Giesbers' claim that the neutralization of this /s/ - /z/ opposition is accepted in the standard language as part of the generalization of Amsterdams, a sub-dialect of Hollands (Hagen and Giesbers 1988:32). Van Haeringen suggests that words preserving initial f- or s- might also reflect a Frisian form; he contrasts the Frisian ethnonym Fries with the old Dutch pronunciation preserved in the name De Vries. He does urge caution, however, pointing out that some f- words, such as fladderen 'flutter' or fniezen 'sneeze' (which occurs beside niezen 'sneeze') are onomatopoeic, while others are later borrowings from French. An s- word probably from Frisian is sim 'line'. But here again many s- which occur where z- are expected are borrowings from French and German, or also onomatopoeic, such as souzen 'to buzz'. As in the case of borrowings, layers of time need to be looked at closely here as well: even though the same development may show up in both Hollands and Frisian, it could still be an independent development in each. One must search for factors which argue against such an independent development, such as a time separation between occurrences of such a development, which would strongly suggest it was borrowed. As Hickey (1995:116) points out, "[g]oing on the assumption that a language is a self-contained structural work, the possible addition of elements from a further language is most likely in a situation where the speakers of one language are learning another and mix elements of their first language into that which they are acquiring." The process of interference to which he refers is not limited to phonology; and indeed there are also suggestions of Frisian interference in Dutch morphology and syntax. The ge- prefix occurs on Dutch nouns (as in de gebroeders Grimm 'the brothers Grimm') and verbs and deverbal adjectives (Het boek is gestolen / het gestolen boek 'The book is stolen / the stolen book'). In Frisian, as in English, this prefix has almost entirely disappeared, even in nouns where the collective function is not immediately obvious (compare Dutch geschiedenis and Frisian skidnis, 'history'). Use of the prefix in Dutch collectives is now less frequent then it once was. Words such as gebroeders 'brothers' are now fossilized forms, and the prefix is only productive for nouns with verbal stems and often has a belittling implication, as with schreeuwen 'to scream, to screech', geschreeuw 'screaming, screeching' (personal communication with Dr. Hilda Koopman, UCLA Department of Linguistics). Some words which formerly included this prefix now appear without it: (ge)makkelik, (ge)beuren (Van Haeringen 1923:4). This loss of this prefix in ABN may possibly be traced to Frisian influence on Dutch morphology, a suggestion supported by dialectical distributions of its participial use. In the Dutch (Flemish) speaking Belgian provinces, as well as the Netherlands' southern and middle provinces of Zeeland, Limburg, Zuid-Holland, Utrecht, Gelderland, Overijssel, and Noord-Holland up to the river IJ, there are various realizations of this prefix ranging from /x-/ to /h-/ and even /e-/. The northern portions of Noord-Holland, as well as Friesland, Groningen, and Drenthe, on the other hand, do not have any form of the prefix in this position (Schönfeld 1959:136). This may be merely a matter of phonological reduction and eventual loss, but the distribution, combined with that of other Frisian (or at least Ingvaeonic) characteristics is suggestive. There is also possible syntactic interference in the word order of subordinate clauses: Goossens (1977:42) presents a map showing that in the provinces of Noord-Holland, Friesland, Groningen, and most of Drenthe, the accepted way of saying 'Auntie wants (me) to say that she's not coming today' is Tante laat zeggen, dat ze vandaag niet komen kan. In contrast, the Dutch (Flemish) speaking part of Belgium allows only Tante laat zeggen, dat ze vandaag niet kan komen. The areas in between - that is, the southern provinces of the Netherlands - find both ways acceptable. (ABN, being based upon Noord-Hollands, also prefers the niet komen kan option, which parallels Standard German's dass sie heute nicht kommen kann.) This distribution may be due to Ingvaeonic influence in the north, or French influence in the south, or other factors entirely, but it suggests another avenue of research. As Thomason and Kaufman point out, "word order seems to be the easiest sort of syntactic feature to borrow or to acquire via language shift" (Thomason and Kaufman 1988:55), and it is too much to base my entire argument on this point. In fact, both the ge- prefix discussed above and the word order just mentioned are presented as tentative suggestions which require further research. To repeat, once again: borrowing and interference may go on simultaneously in the two groups of speakers (Thomason and Kaufman 1988:69, 71). In the case of Frisian and Dutch, while the speakers of medieval Hollands are borrowing these terms from the local Frisian-speaking population, that Frisian-speaking group is shifting to Hollands. And this shifting group is maintaining those parts of its native vocabulary which the target language, Hollands, has no words for, such as klunen, along with at least some of their native phonology, which spread through Hollands and thence into ABN. And just as the Cushitic and Ethiopic Semitic languages discussed above shared some morphological and syntactic categories (Thomason and Kaufman 1988:134), so, too, similarities between Hollands and Frisian may have increased the likelihood of the concurrent operation of the two processes.
|