Oneness > Indefiniteness > Impreciseness in Numbers and Clock Time > Reciprocal Disorder in Bavarian, where else. Frans Plank (Universität Konstanz) - PDF

Description
Oneness Indefiniteness Impreciseness in Numbers and Clock Time Reciprocal Disorder in Bavarian, where else Frans Plank (Universität Konstanz) 1. From one to indefinite, as usual It is exceedingly

Please download to get full document.

View again

of 24
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Information
Category:

Leadership & Management

Publish on:

Views: 16 | Pages: 24

Extension: PDF | Download: 0

Share
Transcript
Oneness Indefiniteness Impreciseness in Numbers and Clock Time Reciprocal Disorder in Bavarian, where else Frans Plank (Universität Konstanz) 1. From one to indefinite, as usual It is exceedingly common for the cardinal numeral one to be reanalysed as an indefinite form, pronoun and/or article. Like often elsewhere, this also happened in all the Germanic languages. In Bavarian, an Upper German dialect mainly spoken in the German state of Bavaria and in neighbouring Austria, while arguably rich in old non- Germanic admixtures, one -derived indefinites are part of the heritage that is shared with the rest of German. Thus, corresponding to ein-s and ein- in Standard German, the numeral one in Bavarian is oàn-s as a counting form and oàn- otherwise, inflecting for gender, case, and number (if as a singulare tantum) in an adjectival manner. In many regional varieties of Bavarian /n/ induces nasalization on a preceding stressed vowel or diphthong and syllable-finally tends to drop; thus, the numeral (like the indefinites based on it) is [O å (n)] rather than [Oån]. In this same form, oàn- also serves as Bavarian s chief indefinite pronoun, and is inflectionally indistinguishable from the numeral used without a noun, except that, unlike the numeral, it also has plural forms: (1) MASCULINE NEUTER FEMININE PLURAL NOMINATIVE oàn-à oàn-s oàn-e oà ACCUSATIVE oàn-n oàn-s oàn-e oà DATIVE oàn-n/oàn-m oàn-n/oàn-m oàn-à oàn-e The inflectional endings transparently correspond to those of Standard German (oàn-à = ein-er, etc.), though Standard German lacks a plural (oà = ein-e, as in kein-e no ), using - 1 - the plural of the interrogative welch-e(n) in indefinite function instead (Plank 1994). Owing to the reduction of final nasal clusters, accusative and dative singular come out as oàn and oàm, with the distinctive dative form in -m under threat from the accusative. Somewhat straining the notion of indefiniteness since the relevant contrast is one of identity (or proximity) and alterity, oàn- also occurs in the company of the definite article, 1 typically in syntagmatic contrast with the other and inflecting essentially like a weak adjective: dà oà(n-e) the one, de oàn(-à) the ones, etc. The indefinite article is reduced in segmental substance, but also inflects for gender and case, showing more neutralizations, though: (2) MASCULINE NEUTER FEMININE NOMINATIVE à à à ACCUSATIVE à-n à à DATIVE à-n/à-m à-n/à-m à (àràn/àràm) (àràn/àràm) (àrà) The long dative alternants, augmented by extra segments /rå/, are synchronically and diachronically opaque (Standard German has ein-em, ein-er); probably, internal /r/ phonologically derives from stem-final /n/. In some southerly varieties there are regular plural forms of the indefinite article, too. Now, unlike in lots of other languages getting their indefinites from the numeral one, including contemporary Standard German and other German dialects, indefiniteness was not the end of the road in Bavarian. Not altogether unexpectedly, although native grammarians have found this hard to face up to, one might say it was going further downhill with indefinites, or rather with forms derivative of indefinites. But there are always lessons to be learnt from decline. The main lesson in such cases tends to be that there are more ways of declining than one might have thought. What we learn from the particular case at hand, on a more encouraging note, is that there are also more ways of hanging on. Capitalizing on natural conceptual affinities and with the play of morphological and syntactic reanalyses reined in none too strictly, there can 1 Perhaps as a nominalized pro-adjective rather than as a pro-noun: unlike in English and like elsewhere in German, noun phrases without a noun do not really need one as a head be more in store for a numeral one than to end its grammaticalization career as an indefinite Two numerals juxtaposed, plus (à)... à for approximation In comparison with contemporary Standard German and also with other German dialects, where numerical approximation of this sort is typically expressed through the asyndetic juxtaposition of two neighbouring numerals (3), Bavarian, apparently in all its regional varieties, goes to greater lengths, adding a form à in between the two numerals and optionally also in front of the first (4). 3 (3) Ambros säuft jeden Tag sieben acht Bier, Benedikt bloss vier fünf Ambros downs every day seven eight beers, Benedikt only four five Ambros downs some seven or eight beers every day, Benedikt only some four or five (4) Dà Ambros saufd jedn Dåg (à) sim à achd Bià, dà Bene bloß (à) fiàr à fimf(e) the Ambros downs every day (?) seven? eight beers, the Bene only (?) four? five Ambros may be drinking seven beers per day, or eight, or perhaps also only six or as many as nine, and mutatis mutandis for Bene(dikt): the speaker does not vouch for any precise number, but only for some number in the region of the two numerals given, not including the next round numbers on either side. 4 The identity of this extra à is a problem. One suggestion, probably originating with Schmeller s pioneering Grammatik (1821: 775), is that it is a reduced form of the disjunctive connector oder or. This would make semantic sense, since the non- 2 The origin of indefinites has received far more attention in the grammaticalization literature than their further progress; noun phrase markers of some sort is what they are known to be able to be turned into. Still, though not widely reported, it is unlikely that scenarios like those to be sketched presently are unique to Bavarian. 3 In spelling Bavarian, in a makeshift orthography which somewhat exaggerates the closeness to Standard German (the dialect is hardly written for any practical purposes), what is relevant here is that à is used for [å], contrasting with a back a [A], and also with mid back å [O]. 4 The juxtapositional construction also permits neighbouring ROUND numerals, such as fifteen, twenty, twenty, thirty, or two, three hundred exclusive disjunction of two or more numerals is a crosslinguistically frequent mode of expressing numerical approximation. 5 However, even the most casual of Bavarian speakers would not succeed to reduce [ o.då], with stress on the first syllable, to its unstressed second syllable minus its onset, i.e., to [å]. 6 An even bigger mystery on this interpretation is why à = oder should accompany also the first numeral: oder occurs BETWEEN disjuncts, never PRECEDING a FIRST disjunct. Nor can it be argued along such lines, seeking support for the disjunction analysis, that the asyndetic juxtapositional construction, as in (3), is not really so different from the disjunctive construction, but is only a more extreme variant of it, with the connector oder not only segmentally reduced but elided altogether. The asyndetic juxtaposition of numerals represents a wholly different strategy of conceptualizing numerical approximation, unrelated to disjunction: its rationale is not to give alternatives but an (abbreviated) enumeration of those numbers falling within the admissible numerical range. There is another interpretation of à in such constructions which is semantically equally plausible and formally entirely unproblematic, and which has plausible ramifications also elsewhere, as will be seen below. On this interpretation, unlike contemporary Standard German and probably unlike other contemporary German dialects, Bavarian has seized on the indefinite article, itself deriving fom the numeral one but synchronically distinct from it, and standardly employs it or rather a form deriving from it with a sequence of numerals to indicate that the numerical value is only approximate: (5) [(à) NUMERAL=à NUMERAL] NOUN The à which precedes the first numeral becomes omissible when another numeral follows, presumably owing to this mode of expressing numerical approximation having gotten grammaticalized and hence lending itself more readily to formal simplification. Originally coming associated with the second numeral in a sequence of two, the construction as a whole is tight enough for the second à to encliticize onto the first numeral, in line with a general preference in Bavarian (as in 5 On this whole domain of numerical approximation see further Plank (2003a, b). Support of this research within the Sonderforschungsbereich 512 ( Variation und Entwicklung im Lexikon ) at the Universität Konstanz is gratefully acknowledged. 6 Self-evidently, another semantically plausible connector, bis to, through, would not reduce to à either German in general) for a leftward association of clitics. In fact, there is little evidence to militate against an analysis of the second à as a genuine suffix of the first numeral, triggered by the approximative construction of two numerals in a row. Neither à is in immediate syntactic construction with a noun (plus perhaps modifiers) within a noun phrase, as is the indefinite article; their co-constituents are the two numerals. This enables the indefinite articles reanalysed as in (5) to appear in noun phrases with a definite article or other definite determiners: (6) a. de (à) sim=à achd Bià, de dà Ambros jedn Dåg saufd the (APPROX) seven APPROX eight beers which the Ambros every day downs b. Dà Ambros saufd jedn Dåg seine (à) sim=à achd Bià the Ambros downs every day his (APPROX) seven APPROX eight beers Although Bavarian has innovated a regular plural for the indefinite pronoun (and in some varieties also for the indefinite article), the form used for approximation in construction (5) is singular. (With the possible exception of typical counting units, nouns accompanying such approximative numeral sequences are plural although that used for illustration in (4), Bià, happens to have a plural identical to the singular.) As shown above in (2), in nominative singular the indefinite article à neutralizes gender; but it does inflect for case, and in other cases genders are distinguished (never all three, though). Used in the approximative construction (5), however, à is essentially invariable. Thus, in construction (5), owing to its co-constituency and its lack of morphological variability for gender, case, and number, à can no longer be identified with the indefinite article. It has become reanalysed as a marker of enclitic or indeed suffixal status whose meaning is ill-defined numerical value in the region of the two numerals which the marker is in construction with. In fact, in some varieties of Bavarian, the approximative marker does show phonological variation, taking the form [ån] before vowel, and [å] otherwise; thus, (à) sim àn achd Bià, (à) fiàr à fimf(e) in example (4). Underlying postvocalic final consonants, notably nasals, are evanescent in Bavarian, but are retained in hiatus position; when there is no underlying final consonant to be retained, intrusive /r/ is - 5 - productively used to avoid hiatus. 7 Which goes to confirm the analysis of approximative à(n) as deriving from the indefinite article rather than from, say, oder or. And there is what looks like another case of formal variation of approximative à. After prepositions governing the dative, initial à, when present, may also take the form àrà (7b), with the second =à remaining invariable, though: (7) a. Mid (à) achd=à nein Bià fahrsd bessà mid à(rà)n Daxi, Walburga b. Mid (àrà) achd=à(*rà) nein Bià... with (APPROX) eight=approx nine beers go.you better with a taxi, Walburga With some eight or nine beers you better take a taxi, Walburga The longer form àrà looks like the longer feminine singular form of the indefinite article (cf. (2)): but this would be the wrong gender, Bià beer being neuter (like Daxi taxi ); and the long forms of the right gender (àràm/àràn) are impossible as approximative markers. As will be seen subsequently, there is evidence of morphological material including indefinite-derived à having been assembled to yield a clitic or suffix àr(-)à to express numerical approximation and related notions in the company of several kinds of hosts, including prepositions: in cases like (7b), we seem to have an instance of it. And it does not even need two numerals in a row to trigger this newly assembled form àrà as the sole marker of numerical approximation after dative prepositions: (8) a. Mid nein Bià... with (precisely) nine beers... b. Mid=àrà nein Bià... with=approx nine beers... One question that remains open is why the basic approximative marker à itself does not seem to have caught on also with single numerals, remaining confined to pairs: (9) *Dà Ambros saufd jedn Dåg (à) sim=à Bià 7 Far less frequently, hiatus is also avoided by intrusive /n/ in Bavarian the Ambros downs every day (APPROX) seven=approx beers As the pattern is now, =à is a marker that comes with an entire construction, (5), consisting of two numerals in a row, rather than expressing a meaning on its own, independent of its construction. But gaining independence would, perhaps, be the next step in its grammatical reanalysis. Taking this step in fact seems to be facilitated by the presence of another word secondarily utilized for purposes of numerical approximation in Bavarian and elsewhere in German(ic), 8 on its own or in combination with other forms (including possessive pronouns): the manner and amount pro-adverb so. (10) Dà Ambros saufd jedn Dåg so(=à) sim Bià the Ambros downs every day so(=approx) seven beers Like the obligatory à in sequences of two numerals in construction (5), the optional à here has a preceding word to formally associate with, whose meaning, moreover, bears some resemblance to numerals insofar as it is quantificational as least in its deictic use (e.g., so gross that big ). Arguably, thus, à still comes with a whole bipartite construction, [ADVERB=à NUMERAL], rather than applying to a numeral as such. An alternative approximative marker with this adverb, incidentally, is the composite form that we have just seen with dative prepositions, àr(-)à: (11)... so(=àrà) sim Bià And this is not the last context that àr(-)à has been spreading to. Whatever will be its future, tracing approximative à to the indefinite article, itself derived from the numeral one, leaves a fundamental question unanswered about its past: Precisely HOW did à get reanalysed in this function in Bavarian, of all German(ic) varieties? A comparative look at (West) Germanic in the next section will help in piecing together a plausible full story. One part of this story is arguably a tale of metaphorical transfer, and it goes as follows. 8 Cf. English ten or so beers, ten beers or so One thing that is made clear by the vocalic form of the approximative marker is that it is indeed the indefinite article (à(n)-), rather than the numeral (oàn-), which must have been its immediate source at least in Bavarian. Now, indefiniteness is to do with the reference of a noun phrase: essentially, an indefinite article or also pronoun gives overt recognition of the speaker s assumption that the referent at issue cannot be uniquely identified by the addressee. 9 On notional grounds, it is not really self-evident that forms for indefiniteness are naturally predestined to also serve in the domain of quantification, specifically for expressing numerical approximation. (Though quantificational, the numerical value 1 as such would be even less obvious as a suitable direct source.) As seen rather close to home, namely in (standard) Modern High German, numerical approximation is not inconsistent with the very opposite of indefiniteness; when local adverbs were reanalysed as approximative markers, they came firmly joined with the DEFINITE article, in plural form but now formally invariable: an/um die fünf Liter around/about the five litres. In Bavarian itself, incidentally, the only local preposition thus reanalysed as a genuinely dialectal approximative adnumeral is um around, and it is NOT combined with the definite article, but once more with àr(-)à (less commonly also just à), already encountered as an approximative marker with dative prepositions and the deictic adverb so: (12) Nach umm=(àr)à fimf Mass schlaffd dà Schorsch immà ei after around=approx five litres sleeps the Schorsch always in After some five litres Schorsch always falls asleep Despite the clear notional distinctness of approximation and indefiniteness, it would not seem too far-fetched, however, to recognize a relationship between them that is sufficiently close to encourage the forging of a metaphorical link. Non-unique identifiability of a precise numerical value would thus be metaphorically derived from non-unique identifiability of a referent. And Bavarian would not be the only language thus to reanalyse a one -derived indefinite article as a marker of numerical approximation. Close by, there is Italian (as in fact observed by Schmeller 1877: 730), where the indefinite combines with single 9 In Bavarian, unlike in Standard German, the indefinite article à(n)- accompanies mass nouns as well as singular count nouns, and in some varieties also plural count nouns; but this seems immaterial for the present issue numerals (preferably round ones: un dieci, un cinque cento approximately ten, five hundred, literally a ten, a five hundred ) or also with a sequence of numerals (un cinque o sei metri lit. a five or six metres ); or even closer, there are its immediate southerly neighbours and old contacts, the Rhaeto-Romance languages (where, however, indefinite-based approximatives, such as en trenta meters some thirty meters in Surmeir, are marginal, with derivatives of Latin vere truly or bene well as approximative adverbials of numerals being far more common). Within (West) Germanic, there is Dutch, where the indefinite article een, identical to the numeral one, can optionally be added in front of two numerals joined by the specifically approximative connector à ((een) tien à vijftien biertjes (some) ten to fifteen beers ), or where een on its own renders a single round numeral approximative (een tien biertjes some ten beers ); and there is also English, which goes for a plural indefinite (some seven (or eight) beers; but cf. also a hundred books vs. (one) hundred books, with a implying that the number is approximate). 10 In fact, earlier Standard Modern High German itself had the indefinite article as an approximative marker too, preceding single numerals (ein drei Jahre warten to wait some three years ; see Grimm & Grimm 1862: 137). 11 None of these languages, however, has two (ex-) indefinites in sequences of two numerals in their accustomed position preceding the noun, alla Bavarese (5), but either uses a disjunctive or some other connector between the two numerals (INDEF NUMERAL or/to/*indef NUMERAL). Pointing out such parallels and divergences is not to imply that Bavarian inherited or borrowed and then adapted. Rather, innovating at some point vis-à-vis the rest of German, perhaps some two or three centuries ago, the indications are that Bavarian has independently reanalysed its indefinite article as an approximative marker in construction (5). Especially when certain predispositions are shared, this does not seem such an unnatural thing to do that one would perforce need a model. 10 Not such a far cry from indefinites, the mid-range quantifier einige can also function as an approximative marker in Standard German: Ambros säuft jeden Tag einige sieben Biere Ambros downs some seven beers every day. 11 In such uses, ein is is almost invariably invariable. Sanders (1876: 353) records one case of an inflected form: ein-e 2 Stunden some two hours, where -e can be interpreted as feminine singular, agreeing in gender as if the noun where singular, or also as an innovated plural (otherwise only attested in Bavarian), provoked by the noun
Related Search
Similar documents
View more...
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks