Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Posting-Version: version B 2.10.1 6/24/83 (MC840302); site boring.UUCP Path: utzoo!watmath!clyde!burl!ulysses!mhuxr!mhuxt!houxm!whuxl!whuxlm!akgua!gatech!seismo!mcvax!boring!lambert From: lambert@boring.UUCP (Lambert Meertens) Newsgroups: net.nlang Subject: Re: Call for Proper Noun Idioms Message-ID: <6725@boring.UUCP> Date: Mon, 13-Jan-86 16:21:06 EST Article-I.D.: boring.6725 Posted: Mon Jan 13 16:21:06 1986 Date-Received: Wed, 15-Jan-86 03:10:47 EST References: <161@aero.ARPA> <868@kuling.UUCP> <869@kuling.UUCP> <6716@boring.UUCP> <4122T3B@PSUVM> <1027@lsuc.UUCP> Reply-To: lambert@boring.UUCP (Lambert Meertens) Organization: CWI, Amsterdam Lines: 158 Apparently-To: rnews@mcvax >>> An expression for "gibberish" in Dutch is "Koeterwaals". This could mean >>> something like "the language spoken in Koeterwaal" (several Dutch place >>> names end in -WAAL) except that there is no such place. >> This had never occured to me before, but reading Lambert Meertens's >> posting seems to lead to the conclusion that the English word >> CATERWAUL comes from the same source, and that the etymology and >> definitions in NEW AMERICAN HERITAGE DICTIONARY (1981) may be >> partly "folk linguistics." ... [it says] perhaps from Low German >> katerwaulen : kater, tomcat . . . + waulen, to screech." > Why couldn't "Koeterwaals" be a play on "caterwaul/katerwaulen" > and the place-name-language ending "waals"? Sounds good to me. Kinship between words in different languages can be of several types. The main two are: (i) the words are "cognates" (which for the purpose of this article I restrict to the case when the languages have a common ancestor that already had a version of the word, and both took the word with them on their evolutionary paths); and (ii) the word in language A is a "loan word" from language B (and in the course of language evolution the word may have been changed to conform to the phonology, morphology and spelling of A, may have shifted its meaning and may have disappeared from B, so that it only survives in A). For example, English GOOD, Dutch GOED and German GUT are cognates; English FURLOUGH was taken from Dutch VERLOF. (Since the final F is not mute, the GH was probably earlier pronounced as in ROUGH. There is a slight change of meaning, since Dutch VERLOF often means FURLOUGH, but is used also for other kinds of LEAVE--a cognate of the component LOF--than those of absence.) In DOUBLEGANGER from German DOPPELGAENGER we see a partial replacement by a cognate word; EIGENVALUE from EIGENWERT is much weirder (EIGEN is cognate to OWN, but VALUE has no kinship or resemblance to WERT). It may also happen that a word in language A is a neologism modelled after a word in language B; if A and B are relatives, this may result in something indistinguishable from cognates or loan words, except by historical research. An example is Dutch VERLOF, which was modelled after German VERLAUB. Although both words are built from cognate components, they are not considered cognates in their entirety since VERLOF once was a neologism. It is not considered a loan word either, because at the time the word was formed (and also now) the meaning of the newfangled word was not so much VERLAUB, but rather URLAUB (or ERLAUBNIS), VERLAUB being only used in the deferential phrase MIT VERLAUB = WITH [YOUR] PERMISSION. However, the distinction with a loan word is tenuous in this case. Other possible influences are that an existing word in language A is modified under the influence of a similar word in language B; this may involve the spelling, the meaning, or both. This host (from Old French [H]OST, cognate with GUEST) of possibilities makes etymology a difficult science, fraught (from Dutch VRACHT, cognate with FREIGHT) with pitfalls for amateurs like you and I. Before professional etymologists will admit so much as the plausibility of cognation, the words in question must be derivable from an older common form, for each of the languages involved in accordance with the developmental laws for that language. Furthermore, if one has to go back as far as a reconstructed common ancestor X which has not been recorded, than one wants to find the word attested in other descendants of X as well. However, there are cases in which the cognation is so obvious as to be unquestioned, but with quirks in the evolutionary development or with an unexplainable lack of attesting forms in some descendants of the common ancestor. In the end, it all is a question of more or less plausible, in the context of a whole web of plausibilities. There is no doubt in my mind that the WAALS in KOETERWAALS and the WELSCH in KAUDERWELSCH are related, and the most plausible is that they are cognates, as is English WELSH. In equating KOETER with KAUDER, however, there is a huge problem: Dutch (and English) T can correspond to German Z (TWEE/TWO/ZWEI) or SS (WATER/WATER/WASSER), and in later common loans, often from Romanic languages, to German T (TOREN/TOWER/TURM), but not to German D. The converse would not be a problem (DEUR/DOOR/TUER, VADER/FATHER/VATER, DUIZEND/THOUSAND/TAUSEND). German D is always D in Dutch cognate words and then TH (voiced or unvoiced) in English (DAT/THAT/DASS, DIK/THICK/DICK) or D in English in later common loans (DOZIJN/DOZEN/DUTZEND). (In the triple DOM/DUMB/DUMM a Low German version DUMM that kept the voicing of D like Dutch did superseded the Old High German version TUMP, now extinct.) Also, although Dutch OE can correspond to German AU (BOER/BAUER), this is a rare and irregular happening, possibly occasioned by the need to distinguish in Dutch between the meanings of neighbour (now Dutch BUUR) and peasant, whereas the German form was influenced by the verb BAUEN (to build); cf. Dutch LANDBOUW = agriculture, whence LANDBOUWER. Normally, Dutch OE corresponds to English OO and German U or UE (VOET/FOOT/FUSS) and German AU to Dutch UI and English OU (HUIS/HOUSE/HAUS). In summary, for Dutch KOETER we should expect a German counterpart *KU(E)SSER and for German KAUDER Dutch *KUIDER. A regular English counterpart to KOETER would be *COOTER and to KAUDER it would be *COWTHER or possibly *KYTHER. In view of the identical and specialized meanings, and an undeniable resemblance, it seems unlikely at first sight that there is no genetic relationship at all between KOETER and KAUDER. [A correction to my earlier article: I stated: "There is no Dutch word *KOETEREN." This is wrong. Although I did never encounter the word before, it is listed in Dutch dictionaries, with the same meaning as German KAUDERN, viz. to gibber/jabber, or specifically to talk in a foreign language not understood by the person who classifies this as KOETEREN. The -EN is the Dutch infinitive suffix to the stem KOETER.] In view of the difficulties mentioned above, it is highly implausible that both derive from one form in Germanic. In view of the fact that the pronunciations are well apart, later loans, either in one language from the other or from a common source, are also quite implausible. Now where does this leave us? It does not suffice to dispel the hypothesis of a genetic relationship, but makes it rather more implausible than it appeared on first sight. To save the hypothesis, we have to invent a more tortuous road. For example, in some Dutch/Low German dialects, the Dutch UI/High German AU vowel is represented by an /u/ sound (as in FOOT = /fut/); rendered in Dutch orthography (/u/ = OE) we have HOES instead of HUIS or HAUS. So there, at least *KOEDER would be regular. If the word is a later borrowing in "High" Dutch from such a dialect, this could explain the OE. And so on (maybe). More evidence would still be needed. As to CATERWAUL, cognation with KOETERWAALS or KAUDERWELSH is implausible, since huge problems in explaining vowel shifts are compounded with an a priori less likely relationship because of the difference in meaning. According to The Oxford Dictionary of English Etymology, the earliest attested cognate formation is CATERWAWED (Chaucer: Wife of Bath's Prologue 354; some mss. have -WRAWET). I don't have a copy of Chaucer; an interesting question is whether the meaning there is already unmistakably that of "noice (like that) produced by cats in rut". There are many reasons for not thinking KOETERWAALS to be a word play on CATERWAUL. In the first place, although there is a surprisingly large number of loan words in English from Dutch or Low German, there are hardly any loan words in English from Dutch, except for recent acquisitions that have kept English spelling and usually pronunciation: COMPUTER, pronounced /k)mpjut0r/; FOLKLORE, pronounced /f)lkl):r0/. In fact, I cannot find even one single example of a loan word from English that has been "Dutchified", although I am sure there must be some. Secondly, for a loan word such a huge vowel shift as from /ae/ to /u/ is unexplainable. To Dutch ears, /kaet0rw):l/ sounds like /ket0rw)l/ and would be rendered *KETTERWOL. A literate person might have turned the word into something like *KATERWAUWELEN (KATER = TOMCAT; WAUWELEN, which has the form of a frequentive of *WAUWEN, = BLATHER, TWADDLE). Thirdly, loan words usually arise because the donating language has a word for a concept in want of an embodiment in the receiving language (like CLOCK from Dutch KLOK filled the need for a word for this new invention). At the time of the loan, the meaning of the imported word is at least one of the meanings of the original (Dutch KLOK originally meant and still has as one of its meanings BELL). If CATERWAUL meant "wail like a cat" all the time, it would not have been borrowed, since Dutch has no problem in expressing the concept. Even if it had been borrowed nevertheless, the semantic shift from "cat wailing" to "gibberish" is huge. Finally, the evidence in the diagram Dutch German Danish Swedish WAALS WELSCH (Romanic) ROTWELSCH ROTVAELSKA (argot) KOETERWAALS KAUDERWELSCH KAUDERVAELSK (gibberish) in favor of -WAALS in KOETERWAALS being the "same" as stand-alone WAALS is just too strong. To return to the original subject, what about Double Dutch? -- Lambert Meertens ...!{seismo,okstate,garfield,decvax,philabs}!lambert@mcvax.UUCP CWI (Centre for Mathematics and Computer Science), Amsterdam