This paper is concerned with proper noun modifiers in the English noun phrase. The paper builds on the work of Rosenbach (2007, 2010), Koptjevskaja-Tamm (2013) and Schlücker (2013). The starting point is the observation argued in these studies that proper noun modifiers (henceforth PNMs) can have a classifying or an identifying function in the noun phrase. The goal of this paper is to further explore and develop the description of the identifying function. The claims proposed in this paper are based on a corpus study of 500+ proper nouns functioning as modifier in the noun phrase in the Collins WordbanksOnline Corpus (see also Author 2013).
Rosenbach (2007, 2010) focusses on English PNMs that refer to persons in definite noun phrases and proposes that the identifying function of these PNMs is equivalent to that of determining genitives. Schlücker (2013), mainly on German PNMs, argues against semantic equivalence. According to her, the identifying strength of PNMs is weaker than that of genitives. I agree with Schlücker that Rosenbach’s proposal of semantic equivalence is untenable, but will argue that the argument can and should be taken further.
A first point is that the corpus data show that, contrary to genitives, identifying PNMs are not restricted to definite noun phrases but also occur in indefinite ones, e.g. a Christmas day tragedy (meaning a tragedy that happened on Christmas day), an IKEA spokesperson,. Semantically, this entails that PNMs, in contrast to genitives, do not necessarily provide exclusive identification.
A second point follows up on Rosenbach’s observation that not all PNMs have a genitive alternate, which she takes as evidence for the fact that PNMs can express a wider range of relations than genitives. My corpus data confirm this, but also show that PNMs tend not to be used for core-genitive semantic relations such as possession in English. With this in mind, I reconsider Schlücker’s proposal to analyse PNMs as having an “anchoring function” (Rijkhoff a.o. 2008, Zifonun 2010). This function is, in the prenominal zone, closely tied to the genitive both in terms of its semantic description and in terms of its positional specification (following the determiner but preceding any qualitative modifiers). In addition to the semantic discrepancies already indicated, the corpus data also provide examples that invalidate the positional claim, i.e. where the PNM is preceded by a qualitative modifier, the on-loan Portsmouth striker.
As an alternative analysis, I propose that the identifying PNMs are functionally similar to “identifying epithets” (Halliday 1994): descriptive modifiers that convey a property that is sufficient to single out and thus identify the referent in the appropriate context, e.g. red in the red car when picking out one car in a row of cars of different colours. I will argue that this solves both the positional and semantic issues connected with the anchoring analysis. I will also discuss how the analysis has wider-reaching implications for the model of the English noun phrase, as it entails that the function of epithet (or qualitative modifier) is not restricted to adjectives.
ReferencesAuthor. 2013. A new analysis of proper noun modifiers in PDE NPs. Conference paper.
Proper nouns are categorized as a subclass of nouns (or nominals), however, they often behave differently from common nouns in terms of overt marking of grammatical categories. While the morphosyntactic differences between pronouns and common nouns have been investigated extensively, hardly any studies have dealt with the differences between proper nouns and common nouns. I will present a crosslinguistic survey that compares the marking of the two categories definiteness and case for proper and common nouns. For the overt marking of a nominal category, there are five logical possibilities:
1. the category is not overtly coded on both common nouns and proper nouns
2. the category is overtly coded on common nouns only and not on proper nouns
3. the category is overtly coded on proper nouns only and not on common nouns
4. the category is overtly coded on both proper nouns and common nouns, but the two types combine with distinct markers (and possibly distinguish a different number of feature values)
5. the category is overtly coded on both proper nouns and common nouns, identical markers are used for both types of nouns
The most interesting scenarios are 2 – 4, i.e. the ones that differentiate between proper nouns and
common nouns. I will concentrate on those three types in my paper, but will also give an overview of
the frequency with which definiteness and case are overtly coded on proper nouns. Scenario 2 (lack
of marking on proper nouns only) is exemplified by the definite article in English, which does not
combine with proper nouns (but see below for the discussion of a special case). I am not aware of
any instances of Scenario 3 (lack of marking on common nouns only). Further research is needed,
but, as of yet, there appears to be a universal tendency that proper nouns do not distinguish more
categories than common nouns. Scanario 4 can be found in Tagalog (Central Philippine) and Libido
(Eastern Cushitic). In Tagalog (Schachter & Otanes 1982: 74, 79), distinct forms of the role markers
are used for common nouns (ng ‘non-topical actor/undergoer’, ang ‘topic’) and proper nouns (ni ‘nontopical
actor/undergoer’, si ‘topic’). Libido distinguishes no declension classes for common nouns
(but feminine nouns have an overt gender-marker), proper nouns on the other hand are grouped into
five declension classes (three for masculine and two for feminine ones). While the form of a noun
used as direct object (Accusative) is identical to the Citation Form for common nouns, Accusative
and Citation Form are distinct for the majority of declension classes of proper nouns (Crass 2014).
I exclude from my investigation cases in which the marker of a nominal category has different functions for proper nouns and common nouns. Such a case is, for instance, the special pragmatic function of the English definite article when combined with proper nouns (cf. I saw THE Will Smith, roughly meaning ‘I saw the well-known actor Will Smith’). Another example, from a different nominal category this time, is the so-called ‘associative plural’ that is found especially frequently with proper nouns. Those cases are certainly very interesting from a pragmatic point of view, but they fall beyond the scope of my investigation.
ReferencesCrass, Joachim. 2014. Personennamen im Libido, einer markierten Nominativsprache in ¨Athiopien. Paper presented at LiFo, Universit¨at Regensburg on 14.05.2014.
Hoocąk is a North American Indian language of the Siouan language family, which is still spoken in Wisconsin and Nebraska. Hoocąk as well as the other 16 Siouan languages are morphosyntactically quite different from the more familiar European languages. They share the following grammatical traits: a very rich and complex verbal morphology, but no infinitives and participles, almost no nominal morphology, but a great variety of definite and indefinite articles and other determiners, no case marking of nouns at all, active/inactive alignment expressed on the verb by indexing the core arguments, SOV word order, and some others.
One of the interesting aspects of proper names in these languages is that they are almost never semantically opaque and that they are very often syntactically complex. It is not uncommon to have proper names that represent complete clauses. The goal of the proposed paper is to present a comparative study on the morphosyntactic properties of proper names – more specifically anthroponyms – in the Siouan language family. The questions that will be answered are:
1) Are there proper name formation rules (word or phrase formation) that are unique for proper names?
2) Which nominal categories occur with proper names in these languages?
3) What kinds of modification and determination of proper names are allowed, disallowed, or obligatory in these languages?
4) How are proper names treated syntactically, i.e. how are they indexed on the verb compared to NP with common nouns?
Adjectives derived by -isch (or its predecessor -isc) have ever since Germanic and Old High German times served to express affiliation and origin (OHG heidanisc ‘heathen’). Due to this function, the -isch-derivation has displayed a strong affinity to onymic bases from early on (OHG sirisc ‘Syrian’ < Siria).
The subsequent development of the deonymic -isch-derivation is highly interesting in that it gives a deep insight into the onymic status of the bases involved. This will be shown on the basis of dia-chronic corpus data.
One of the most striking results is obtained by graphematics: Capitalization coincides with highly specific bases, such as town names (Straßburgisch ‘of Strasbourg’) or family names (Fuggerisch ‘of the Fugger family’). Proper names with a wider scope, such as country or continent names, are capi-talized less frequently (Englaͤndisch ‘of/from England’). Finally, with tribal or people designations (tuͤrkisch), capitalization is found rarely − which supports their analysis as common nouns.
The use of upper vs. lower case initials hence provides a scale of increasing “onymicity”, starting with the name-like, yet appellative tribal or people designations. At the other end of the scale, we find the most specific among the occurring onymic bases − town names and family names. Interestingly, their high specificity has consequences not only in spelling, but also for their morphological behavior: These two types of proper names have experienced the most remarkable changes in the history of German -isch-derivation. The centuries investigated (14th−18th century) witness the rise and fall of town names and family names as bases of the -isch-derivation. The rise is quite explicable by the traditional functions ‘affiliation’ and ‘origin’. The decline of the two patterns invites a closer look.
When probing the derivatives in question (like Straßburgisch, Fuggerisch), we find that they are no proto-typical adjectives: They don’t convey a quality, but are purely relational. This relation could alternatively be expressed by a genitive construction (Glasgow’s population, the mayor of London). Interestingly, the -isch-derivatives of town names are indeed replaced by a construction originating in inflexion (Straßburgisch −> Straßburger). The -isch-derivation of family names is replaced by (or mu-tates into) the shortened variant -sch, which then soon conventionalizes for forming relational adjec-tives (Fuggerisch −> Fugger’sch). Interestingly, -isch is sometimes retained when a quality reading is aimed at: Lutherischer Glaube ‘Lutheran faith’. Presumably, lesser content (no quality, just relation) is expressed by shorter means − quite in accordance with iconicity.
Overall, the derivation of the most specific proper names has not ceased, but rather separated from the generic -isch-derivatives by using constructions that are as specific as their onymic bases.
Maria Koptjevskaja-Tamm with Robert Östling (Dept. of linguistics, Stockholm university)
We investigate creativity, productivity and frequency of Swedish proper-name compounds following in the steps of Dahl (2003, 2008) and Kajanus (2005). These studies described several examples of Swedish compounding patterns based on a particular proper name that have manifested a gradual diachronic rise in the frequency of both types (by spreading to further stems) and tokens, i.e. have been gradually entrenched. Dahl’s most striking example is the explosive development of Swedish PropN-compounding with Palme as the first component, following on the important and highly salient event in the modern Swedish history, the murder of the Swedish prime-minister in 1986. In fact, many Palme-compounds are related to the “murder script”, with Palme often metonymic for the Palme murder and also for further compounds derived from it (by means of metonymical chains), cf. Palme+kulorna — ’the Palme bullets, i.e. the bullets found at some distance from the place of the Palme murder’, Palme+misstänkta — ‘Palme suspects, i.e. persons suspected of having committed the Palme murder’, Palme+utredningen ’the Palme investigation, i.e. the investigation of the Palme murder’, etc. In all these previous studies the data come from the Swedish press and novel corpus (86 mln words). Our research uses the Swedish Blog Sentences corpus containing 6 mlrd tokens from 46 mln blog posts in the period of 2010-2014 (Östling and Wirén. We focus on creativity, productivity and frequency of compounds based on several proper names that have been particularly salient in the discourse during the relevant period . We consider how the fluctuations in the type and token frequencies of the proper-name compouns correlate with the rises and falls in the frequency of the relevant proper names. Interestingly, there are very few highly frequent compounds – in fact, 1-2 for each of the proper names considered (e.g., Putinregimen ‘the Putin regime’, Zlatanboken ‘the Zlatan book’, Obamaadministrationen ‘the Obama administration’). On the other hand, each of the proper names ”generates” a high number of unique compounds, i.e. compounds that have only one occurrence in the whole corpus. Finally, there are also proper name compounds that are in-between the unique and the highly frequent ones, but this group is quite restricted.
ReferencesDahl, Östen 2003. Palmesyndromet. A talk given at the First Swedish Linguistic Meeting, March 27-28 2003, Uppsala University.
As is well-known, in many cases the relationship between the ‘official’ first name and its ‘clipped’ variant(s) is far from transparent. It is this factor, of course, which underlies the fact that the clipped variants tend to develop into first names in their own right, meaning that, form a synchronic point of view, they can no longer be considered a variant of their non-clipped counterpart. The little transparent character of the relationship between the ‘full’ first name on the one hand and its clipped variant is illustrated by e.g. English Richard vs. Dick, Dutch Ineke vs. Pien, and (be it in a different way) Dutch Cornelis vis-à-vis Kees.
This paper deals with the question to what extent the relationship between the clipped first names and their ‘full’ counterparts is governed by systematic principles. I will discuss this matter by concentrating on Dutch, a language in which, in the realm of name formation, clipping in still very much alive. On the basis of Vincent (with initial stress), for instance, children coined spontaneously both Vin, Vins and Vis, whereas Vincent (with final stress) is paralleled by Cent.
It will be shown that, irrespective of the bewildering diversity that the clipped names exhibit, the formation of clipped first names is governed by a handful of principles which, as such, are fairly transparent. In addition, I will go into the difference between ‘name formation’ and ‘word formation’. To a certain degree, name formations may be characterized as ‘word formation by phonological means’. In this respect, name formation in Dutch is crucially different from word formation proper, although some parallels do exist!
However this all may be, the general conclusion reads that Dutch name formation appears to be much more systematic than one might expect on the basis of a superficial inspection of the data. At the same time, it is very much different from Dutch word formation in general.
Hebrew personal names have been studied by various researchers, especially from the pragmatic and cultural points of view. Their studies describe names used among religious or secular communities (Birnboum 2000), modern trends in the use of names compared to the past (Rosenhouse 2002, 2013), typical names used among a specific community (Schwarzwald 2010), or the probability that certain names remain in use for a long time (Landman 2014). The grammatical aspect of names was mentioned as part of these studies, but no wide morpho-syntactic description of personal names has been offered, and the purpose of the present study is to fill this gap.
When asking whether there is a grammar to personal names in Hebrew, we can compare it with settlement names, which constitute another category of proper names, and were analyzed by Ephratt (1986). According to her description, these names are clearly governed by grammatical rules, and it appears that similar rules can be applied to personal names.
Proper names are regularly unspecified as to word-class (Anderson 2007). Most personal names in Modern Hebrew are nouns, and many of them are identical to common names. The basic formation of these names is the fusion of roots with patterns, typical of Semitic languages. Thus, for example, the name baraq is formed by the root b-r-q and the pattern CaCaC.
Gender unmarked nouns, regularly used as masculine, may also serve as female names (e.g. rotem), as part of a recent trend of using masculine generics (Muchnik 2015, in press). However, feminine marked forms presenting the suffixes –a, –t, –it are almost exclusively used for female names. We also find derived feminine forms from existing masculine names, such as yo’ela from yo’el. Another suffix found in personal names is the possessive –i, as in karmi (ʻmy vineyardʼ).
Not only nouns, but also other parts of speech may serve as Hebrew personal names. Thus, we can find names with adjectival forms, such as ʼadir (ʻgreatʼ), as well as verbal forms, such as yaniv (ʻhe will yieldʼ). Verbal names do not regularly present parallel forms for female names, although this could be morphologically possible by changing the initial y– into t–. An exception is taʼir (ʻshe will illuminateʼ), which is marked as grammatically feminine, and therefore never used as a male name.
Many personal names present a compound form, whether abbreviated (nili = netsaħ yisrael lo yešaqer (ʻthe Glory of Israel will not lieʼ), blended (linoy, ʻto me [there is] beautyʼ), hyphened (mei-tal, ʻdew-waterʼ) or separated words (bat ʻami, ʻdaughter of my folkʼ). According to Anderson (1997) proper names cannot regularly be predicates unless adding a copula. Hebrew compound names, however, may constitute predicate phrases, e.g. ʼorli (ʻ[there is] light to meʼ).
The study presented in this lecture comprises a full morpho-syntactic analysis of masculine and feminine names offered for baby boys and girls in this website: https://www.materna.co.il. It will show that personal names in Modern Hebrew present grammatical characteristics similar to common nouns, including different parts of speech, inflections, derivations and compound forms.
ReferencesAnderson, John M. 1997. A Notional Theory of Syntactic Categories. Cambridge: Cambridge University Press.
In both Ancient and Modern Indo-European languages, proper names can be involved in
derivational processes, like common nouns. In both Italian (It.) and Ancient Greek (AG), for
instance, proper names and common nouns (and also adjectives, verbs, prepositional phrases
and entire propositions: see Dardano 2009: 121f. and Schwarze 1995 on It., Chantraine 1933:
138f. and Schwyzer 1953: 493 on AG) can be combined with the suffixes It. -ismo/-ista, AG
-ismós/-istḗs to form common nouns such as It. bonapartismo ‘Bonapartism’, bonapartista
‘Bonapartist’ (from Bonaparte), AG. puthagorismós ‘Pythagorean doctrines’, puthagoristḗs
‘follower of Pythagoras’ (from Puthagóras). In both languages, the processes can be related
to each other thanks to Latin evidence: some -ismós/-istḗs items were borrowed from Greek
into Latin -ismus/-ista formations and via Latin they spread into Romance and other European
languages. Lexical borrowing explains the development of some Italian items in -ismo/-ista,
but it cannot account for their extremely high productivity in Italian, which is comparable to
that of AG but not at all to that of Latin, where these suffixes are much less productive.
This paper analyzes both Italian and Ancient Greek nouns having proper names as lexical bases and aims at describing the similarities and the differences between the two languages in terms of the morphological processes involved and the lexical meanings of the derived nouns with respect to the proper names (on the “meaning” of proper names cf. Jespersen 1924, Pulgram 1954, Jakobson 1957, Kuryłowicz 1966, Gary-Prieur 1994, Vaxelaire 2005 and Anderson 2007). Among the important questions discussed in the workshop, this paper tries to give an answer to the following: How does deonymic word formation differ from word formation with the same derivational affixes? Which patterns of onymic word formation can be observed, both language-specifically and cross-linguistically? The general topics dealt with in this paper are thus i) how the difference between a proper name (i.e. the derivational base) and a common noun (i.e. the result of the derivation) can be accounted for in a functional perspective and ii) how the meanings of derived nouns arise from the morphological processes involved. In our perspective the processes of antonomasia and metonymy are essential to explain these formations and their respective meanings (cf. La Fauci 2006, 2008 and 2010 on -eggiare verbs).
Our corpus of investigation is constituted by the lexemes collected from the Italian dictionary Zingarelli (2003) and the AG dictionary Liddell, Scott & Jones (1996 ), for a total amount of ca. 1570 It. nouns with -ismo suffix (175 with proper names) and ca. 1380 It. nouns with -ista suffix (90 with proper names) and ca. 860 AG nouns with -ismós suffix and ca. 530 AG nouns with -istḗs suffix. Corpus analysis is complemented with textual analysis. Furthermore, the study integrates recent neologisms (not yet accounted for in actual dictionaries as e.g. merkelismo ‘Merkelism’) collected from the website of the Italianlanguage Treccani encyclopaedia www.treccani.it.
ReferencesAnderson, John M. 2007. The Grammar of Names. New York: OUP.
The history of the German proper name inflection is a story of a profound change: In the beginning (in Old High German), proper names inflected like common nouns sharing the same declension classes. Later they reduced their inflectional endings first by gradual removal of most allomorphs followed by deflection on the syntagmatic level. Today, proper names belong to a distinct declension class. This development can be explained by the high amount of borrowed names whose word shape needs to be preserved by agglutinative suffixes for reasons of recognition. Until today, proper names contain most foreign linguistic material. Accordingly, this process is reflected in writing: the apostrophe (e.g. Goethe's Werke 'Goethe's oeuvre') exclusively occurs with proper names and marks the boundary between name body and suffix. On the morpho-syntactic level, the definite article in front of proper names developed completely different functions, and even gender assignment is now dissociated from common noun gender. The fact that in many respects proper names diachronically diverged from common nouns leads to the question whether they constitute a different word class.
Background: This paper explores Rothstein’s (2012) observation that proper names are infelicitous as complements in Construct State Constructions (CSCs). CSCs are ‘syntactic words’ characterised by a phonologically reduced head N, a complement directly adjacent to the head and definiteness marked only on the complement, (1). CSCs contrast with free genitive constructions where the head is unreduced, definiteness is marked on both head and complement, and the complement is introduced by the lexical preposition šel (2).
Rothstein (2012) shows that the definite in (1) cannot be replaced by a proper name (unless the CSC is analysed as compound proper name, e.g. bet ariela is the name of the public library in Tel Aviv). Only (4) expresses “Ariella’s house”.
Data: There are three kinds of exceptions to this generalisation: (i) Proper names are acceptable as arguments of derived event nominals. (5a) is felicitous only in the eventive reading, and not as an NP denoting an individual object (5b).
(ii) Proper names are acceptable as complements of ‘picture’ nouns;This use is highly restricted. The head noun must be plural, the distributive universal quantifier kol is impossible and the CSC is definite. The complement must denote the artist/creator, and the more famous the creator, the more acceptable the expression.
(iii) Proper names for places are freely acceptable. The CSC is indefinite, and the definite Accusative marker et is not allowed (7). The CSC can be the complement of distributive kol (8):Analysis: Following Rothstein (2012), CSCs have predicate NP complements at type
ReferencesChierchia, Gennaro and Ray Turner (1988) Semantics and Property Theory, Linguistics and Philosophy 11(3), 261-302.
Regarding gender assignment proper names behave completely different from common nouns, in that in the
long run proper names tend to establish a referentially assigned gender, i.e. ships for ex. are always feminine
(die Peter Pan), cars masculine (der Fiesta), hotels and restaurants neuter (das Hilton). Thus, as Fahlbusch/
Nübling 2014 have shown, genders contribute significantly to the classification of objects (die Admiral
→ ship, der Admiral → car, das Admiral → hotel/restaurant) (cf. also Nübling et al. 2012: 73-76). But until
now only little is known about the development of onymic genders and the gradual change from grammatically
assigned common noun gender to referentially assigned onymic gender.
The present paper aims at filling the gap by taking diachronic changes as well as synchronic variation in gender assignment into account. One case in point are names of cities that are classified neuter in NHG (das mittelalterliche Hamburg ‘theneut. medieval Hamburg’) but exclusively feminine in MHG, cf. (1)-(2). The gender change from feminine to neuter must have occurred by the end of the ENHG period (16th/17th century?). Nevertheless, the exact time span as well as the driving forces behind are completely unexplored.
(1) Diu himeliſch Ieruſalem, diu bedarfe der ſvnnen lihtes niht, vnſer herre, der almehtige got, der erluhtet
ſi ſelb. (priest Konrad’s Book of homilies 20, 66, 13th century).
(2) Da inne ſint ʒuo creftige burge. Die eine heiʒet Adromeuſ, die andere biʒantium. Da bi lit ceuſis vnde die michelú Carthago. (Lucidarius 35, 5, 12th century)
Synchronically, in present day German, gender variation can be observed for names of mountains (der/die Annapurna) and local rivers (der/die Warnow) yielding also hybrid forms, cf. (3)-(4). The latter are of special interest and rise the question whether agreement hierarchy (cf. Corbett 1979, 1991) is important for the gradual consolidation of onymic gender as supposed by Fraurud 2000 for Swedish.
(3) Die Zugspitze, zu dessen Füßen das Bergdorf Grainau Besucher zum Verweilen und erkunden der Landschaft einlädt, bietet Erlebnisse wie kein anderer Berg in der Alpenregion. https://www.gaestehaus-am-kurpark.de/grainau-das-zugspitze-dorf.html (30.4.2015)
(4) (…) hier fließt die Warnow, an dessen Ufer Rostock liegt. https://www.schwarzaufweiss.de/deutschland/rostock-reisefuehrer/vicke-schorler-rolle.htm (30.4.2015)
In order to shed light on the emergence and the consolidation of onymic genders synchronic (Deutsches Referenzkorpus, Decow) as well as historical data (Bonner Frühneuhochdeutschkorpus, GerManC) will be taken into account.
ReferencesCorbett, Greville G. (1979): The Agreement Hierarchy. In: Journal of Linguistics 15/2 (Sep., 1979), pp. 203-224.
The structure of complex toponyms follows the linguistic rules of the language they belong to, but also the specific constraints for proper names. Therefore, the grammar of toponyms has to describe not only their internal structure, often quite complex, but also to take into account the presence of linguistic components belonging to different word classes and / or of various origins and the context in which they occur.
Working on the grammar of toponyms nearly always involves transparent complex toponyms, which have a structure which can be easily analysed by a native speaker like compounds Sandberg, Karlstadt, (Ge.), Sandhurst, Charlestown (Eng.), Sables-d'Or-les-Pins, Charlesville (Frz), Sandhamn, Karlstad (Sw.), or multiword syntactic constructions like Kap der guten Hoffnung, Cape of Good Hope, Cap de bonne espérance, and Godahoppsudden. This kind of toponym can be examined and classified by a traditional morphological analysis which distinguishes heads and modifiers, for example the heads ‘stadt, town, ville, stad, and modifiers ‘Karl, Charles’; but also by a functional analysis based on a double dichotomy classifying its constituents in appellative or proprial elements on the one hand and in specific and generic elements on the other hand. This method takes better into account the different status of the constituents as a proper or as a common name like in our example ‘sand, sables’ as specific appellatives (sA) and ‘Stadt, town, ville, stad’ as generic appellatives (gA) versus ‘Karl, Charles’ as specific proper names (sP).
Several studies have shown that a functional analysis also enables the relation between headwords and modifiers to be studied on a more abstract level than a classic structural morphosyntactic analysis, and facilitates the comparisons between different languages and different means of transposition as shows the example above: Kap der guten Hoffnung (sA = postponed genitive noun phrase), Cape of Good Hope and Cap de bonne espérance (sA = postponed prepositional phrase), and Godahoppsudden (sA = prepositioned nominal constituent of a close compound).
Therefore, we will focus in this study, on the one hand, on the categorisation of complex toponyms in Germanic and Romance languages according to the nature and relation between the generic and the specific constituent. On the other hand, we will examine the different kinds of transpositions of toponyms, which occur by creating exonyms. Dependant on the language, those transpositions go from total adoption of an endonym, Rocky Mountains (Ge, Ita), to small differences in orthography, Straßburg (Ge.) Strasbourg (Fr), to partial translation of an endonym like North America (Eng), Nord|amerika (Ge, Swe), Amerique du Nord (Fr), América del Norte (Sp), and full translation of the endomyms as Montagnes Rocheuses (Fr) and Klippiga bergen (Swe) for Rocky Mountains. The transcription of an endonym written with non-Latin characters to an exonym written in Latin characters also called Romanisation constitutes a special challenge in standardisation. The formal adaption of exonyms or standardisation is one of the main concerns of normalisation, as it is done in the work of the group of experts of the United Nations.
ReferencesBack O., 2002. Übersetzbare Eigennamen. Eine synchronische Untersuchung von inter-lingualer Allonymie und Exonymie. Vienne, Edition Praesens.
Russian has full and short forms (or hypocoristics) of first names. Full forms are used in formal settings, while hypocoristics are used in informal settings. According to Stankiewicz (1968: 146), hypocoristics have the function of ‘familiar and intimate terms of address or reference’. They are used productively among friends and relatives. Hypocoristics are formed by trunctation of the corresponding full forms (1)-(4).
What is puzzling about hypocoristics is that they all belong to the inflectional class II (or the second declension) in Russian, regardless of inflectional class of the corresponding full name. In Russian, nouns of different inflectional classes differ in their morphological properties. For example, Class I nouns have the morphological ending -ø in the Nominative (Nom) case, while Class II nouns have the ending -a in this case(1).
In (1a) and (2a) above, the masuline full names Bor’ís and P’ótr belong to the inflectional class I, evident by the –ø ending in Nom case. After forming of hypocoristics, the class I changes for class II, evident by the –a ending (1b), (2b). In (3) and (4), the full feminine names are already in class II and there is no change in class when hypocoristics are formed. The question arises: Why do hypocoristics always belong to the inflectional class II, regardless of inflectional class of the full names?
I propose that the answer to this question lies within the morphosyntactic properties of Russian hypocoristics. Namely, I propose that hypocoristics are associated with a noun head of their own that occupies a particular hierarchical position in the syntactic tree, as shown in (5). The noun head triggers truncation and is associated with an inflectional class of its own [CLASS II].
Evidence comes from a change in inflectional class of masculine full names, as shown in (1) and (2), as only noun heads are capable of changing inflectional class of the base. A syntactic structure for (1) is proposed in (6).
Another piece of evidence comes from truncated common nouns. Such nouns are used productively in colloquial Russian (Zalizniak 1977). Truncated common nouns belong to the inflectional class II, regardless of inflectional class of the full form, as shown in (6) and (7). Similarly to hypocoristics, truncated common nouns involve a change in inflectional class from class I to class II (6b), (7b).
Thus, I propose that truncated first names (hypocoristics) as well as truncated common nouns are associated with a noun head that is specified for the grammatical feature [CLASS II].
(1) For an analysis of correlation between inflectional class and gender in Russian, see Steriopolo (2008).
ReferencesStankiewicz, E. (1968). Declension and gradation of Russian substantives. The Hague: Mouton.
In the extant literature on word-classes, place names are often classified as a sub-type of nouns, more specifically of proper nouns. This relatively common practice tacitly assumes that place names behave largely similar to other kinds of nouns morphosyntactically. However, the pair of Maltese sentences in (1) raises the question whether this assumption can be defended empirically.
In (1.1), the common noun requires the preposition fi 'in' to yield a grammatically acceptable sentence whereas in (1.2) the place name Malta blocks the use of the preposition although the nuclear predicate of the clause is the same motion verb as in (1.1). Besides Maltese there are many more languages world-wide which provide evidence of the (either optional or compulsory) special behavior of place names in contrast to other classes of nouns. In this context zero-marking of spatial relations is a prominent characteristic of place names - albeit not the only one.
In my talk I discuss a selection of morphosyntactic features which single out place names in a selection of languages from all continents. I argue that there is compelling cross-linguistic evidence of the special status of place names so that it seems to be justified to revise the traditional classification of place names under the rubric of nouns.
ReferencesStolz, Thomas & Lestrade, Sander & Stolz, Christel. 2014. The Crosslinguistics of Zero-Marking of Spatial Relations (= STTYP 15). Berlin, Boston: De Gruyter Mouton.
During the short-lived German colonial empire (1884-1914/9), literally thousands of partly or entirely German place names were introduced in the so-called Schutzgebiete (i.e. protectorates) which imperial Germany had acquired in Africa and the Oceania in the late 19thand early 20th century. The failure of most of these colonial place names to survive into theinterwar period notwithstanding, the inventory of the erstwhile German-based place names is instructive linguistically because the items under inspection share a number of properties which make these place names stand out not only in comparison to the average European-based place names in territories under the control of other colonizers but also in comparison to those place names that were in use in Germany proper around the year 1910, which—in contrast to constructed colonial place names— generally have a long history and had therefore already been subjected to prior language change.
We will focus on two especially striking characteristics of colonial place names, namely their polysyllabic structure and their status as compounds. A case in point is the name of the former capital of Deutsch-Neuguinea, Herbertshöhe, which consists of four syllables, which is representative of a binary compositional structure, and which shows a typical transfer from a single personal name to a determinative constituent in the new toponym. Both of these traits are overwhelmingly dominant among the more than 3,000 (partly) German colonial place names identified so far so that one might want to speak of fixed patterns. We provide a survey of the syllabic, morphological, and semantic patterns—also considering modifying collocations in place name use—which are characteristic of German colonial place names. These data will be compared to evidence from other European colonialisms in order to determine what makes the German case special. We argue that German colonial place names yield a picture which differs from that of the place names in metropolitan Germany. These differences are presented and evaluated systematically.
ReferencesNübling, Damaris, Fabian Fahlbusch & Rita Heuser. 2012. Namen. Eine Einführung in die Onomastik. Tübingen: Narr.
In German, proper names deviate from appellatives in many respects. Apart from phonological and graphematic properties, there are also noticeable morphological differences between proper names and appellatives. One major distinction between these two classes is the inflectional poverty of proper names: Unlike many appellatives (cf. 1), they do not take inflectional elements that affect the shape of a word (cf. 2), or even do not take inflectional elements at all (cf. 3). This holds true for different kinds of proper names (toponyms, anthroponyms, ergonyms etc.) and for both case and number inflection.
(1) Koch [kɔχ] (‘cook’) – Köche [koe.çə] (‘cooks’)
(2) Koch [kɔχ] (family name, singular) – Kochs [kɔχs] (family name, plural)
(3) d-es Iran-s vs. d-es Iran-Ø (‘the-GEN Iran-GEN’)
Preserving the word form of proper names seems to be functionally motivated. The literature has emphasized several aspects that lead to the need of word form preservation: Proper names often exhibit non-native (phonological and graphematic) properties; they can be unfamiliar, some are used very infrequently and they deviate from appellatives with regard to their semiotic properties (cf. e.g. Eisenberg 2013: 145, Nübling 2012: 241, 244 and Nübling 2014: 119). Furthermore it has been claimed that morphological differences are used (indirectly) for formally marking the onymic status of proper names in German (cf. Nübling 2005: 37–41) and additionally, the animacy hierarchy seems to play an important role (cf. Nübling 2014: 119).
However, most proper names exhibit more than one property that allegedly creates the motivation of word form preservation, which complicates the examination of single factors. As a result, the status of each single motivation remains unclear to some extent so far and statements on the importance of single factors are formulated quite vaguely.
In my talk, I will present my attempt to operationalize and to sort out the alleged motivations for word form preservation using the example of proper names that are prone to synchronic variation in Contemporary German. While personal names are rarely inflected now (cf. Ackermann in prep.), especially toponyms vary significantly with regard to their morphological marking (cf. 3). They therefore seem to be a promising testing ground to examine the relative importance of the motivations that induce the need of word form preservation.
The addressed samples are taken from the web-corpus DECOW2012 (cf. Schäfer & Bildhauer 2012). They contain selected proper names, which differ with regard to the relevant motivations so that single motivations can be studied in more detail (e.g. the mountains Himalaya and Aconcagua, which differ with regard to their token frequency). A further focus will be laid on homonymic pairs of proper names and appellatives (e.g. Spiegel as name of a German news magazine vs. the appellative ‘mirror’).
By presenting a large-scaled statistical analysis and a detailed examination of single words which are particularly instructive, I hope to refine the understanding of why proper names deviate morphologically from appellatives in German.
ReferencesAckermann, Tanja. in prep. The German possessive -s-construction with personal names.