Extending Hawkins' comparative typology: Case, word order, and verb agreement in the Germanic languages

of 16

Please download to get full document.

View again

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.
PDF
16 pages
0 downs
4 views
Share
Description
Extending Hawkins' comparative typology: Case, word order, and verb agreement in the Germanic languages
Tags
Transcript
   Nordlyd   34: 167-182, © De Vogelaer 2007  Scandinavian Dialect Syntax   2005  Edited by Kristine Bentzen and Øystein Alexander Vangsnes CASTL, Tromsø. http://www.ub.uit.no/munin/nordlyd Extending Hawkins’ comparative typology: Case, word order, and verb agreement in the Germanic languages Gunther De Vogelaer Ghent University Abstract: In a well-known book, Hawkins (1986), expanding on an srcinal idea by Sapir (1921), attributes a number of typological differences between German and English to the fact that German uses morphological means (i.e. case) to distinguish grammatical relations, whereas English makes use of a strict-SVO word order. Dutch seems problematic to Hawkins’ generalisation, in that neither case nor word order can be used consistently to express the basic grammatical relations. Using verb agreement as an extra parameter, Dutch can be integrated in Hawkins’ typology. In addition, data from Scandinavian languages and Afrikaans indicate that Hawkins’ notion of ‘grammatical word order’ can be replaced by a more precise word order feature, viz. the possibility to place verbs in between subjects and objects in all sentence types. 1. Introduction 1   1.1 Comparing German and English: Hawkins (1986) In a well-known book, Hawkins (1986), expanding on an srcinal idea by Sapir (1921), attributes a number of typological differences between German and English to the way in which both languages distinguish grammatical relations. Whereas German uses morphological means (i.e. case) to do so, English makes use of a rather ‘tight’ syntax (strict-SVO word order), as shown in (1). (1) Hawkins’ contrastive typology of German and English (1986:121) G ERMAN  E NGLISH   More grammatical morphology Less grammatical morphology More word order freedom Less word order freedom More specific selectional restrictions Less specific selectional restrictions Less semantic diversity of GRs More semantic diversity of GRs Less raising More raising Less extraction More extraction More Pied Piping Less Pied Piping Less deletion of NPs More deletion of NPs 1  This paper is an extended version of a section from my Ph.D.-dissertation (De Vogelaer 2005). I would like to thank Magda Devos and Johan van der Auwera for their comments to the relevant chapter. I would also like to thank the audience at the Grand Meeting of the ScanDiaSyn network in Leikanger, August 2005, for their interesting questions and relevant remarks.  G UNTHER D E V OGELAER  168 In addition, a large number of syntactic phenomena are found to correlate with this basic distinction, such as the specificity of selectional restrictions posed by predicates, the semantic diversity of the grammatical relations and the degree of raising, extraction, pied piping, and NP-deletion in the language, as is also shown in (1). Since the emphasis in this paper will be on case and word order, it would lead us to far here to discuss all the phenomena in (1) in detail. Some examples may clarify the typology in (1), however. The specificity of the selectional restrictions on predicates in English is illustrated through the contrast between German ‘verbs of placing’ and English ones: whereas German has a whole set of verbs of placing ( liegen, stehen, sitzen, legen, stellen, setzen ), English almost always uses to be  or to put  . Hence the selectional restrictions on to be  and to put   are less strict than on the German verbs of placing. The same holds for at least one category of transitive verbs, viz. verbs which can take both an ‘affected’ and an ‘effected’ object in English (e.g. to dig  in ‘to dig potatoes’ (affected) and ‘to dig a tunnel’ (effected)). Their German counterparts only take effected objects (e.g. graben ; affected objects are combined with umgraben ). A second point concerns the semantic diversity of the grammatical relations, which is larger in English. For instance, the English direct object subsumes both accusative and dative objects from German. In addition, English is much more tolerant towards non-agentive subjects (such as instruments, in ‘5 Euros should buy you a meal,’ or locations, in ‘This tent sleeps four people’). The third phenomenon in (1) is raising, which is found more often in English than in German, cf. utterances such as ‘I believe John to be ill,’ which are ungrammatical in German. Fourth, English has wider extraction possibilities. For instance, it can extract non-subject complements from finite clauses, as in ‘I don’t know who the police thought that the guilty man was.’ In connection with these wider extraction possibilities, it is observed that English has less pied piping, cf. the ungrammaticality of ‘*the man to kill whom I have tried,’ in which the VP is pied piped, whereas the German equivalent is judged grammatical (‘der Mann, den zu töten ich öfters versucht habe’). The last phenomenon under investigation is NP-deletion. Again, the possibilities in English are wider: as the case distinction between accusative and dative objects is no longer present in English, sentences can be formed such as ‘He went through and out of the tunnel,’ which are not grammatical in German, where one of the objects carries accusative case (‘durch den Tunnel’), and the other one dative case (‘aus dem Tunnel’). In an attempt to unify the contrasts, the typological contrasts in (1) are subsumed under one generalisation by Hawkins (1986:121), which is stated as follows:  E XTENDING H AWKINS ’ COMPARATIVE TYPOLOGY  169 Where the grammars of English and German contrast, the surface forms (morphological and syntactic) of German are in a closer correspondence with their associated meaning. Put differently, German surface forms are much less ambiguous than their English counterparts, which have undergone realignment in the mapping between form and meaning. Ultimately, these typological differences are caused by morphological syncretism in the English NP, which puts the parameter ‘case’ in a special position. Among the other parameters, it seems as if word order is more ‘basic’ than the other ones, 2  as, unlike the other phenomena, there is diachronic evidence suggesting that some word order changes function as a direct ‘compensation’ for ambiguities caused by the loss of case. In English, the oldest attestations of SVO word order are typically found in clauses in which SVO may indeed serve a disambiguating function, viz. in clauses with (non-case-marked) nouns as subjects and objects, rather than (case-marked) pronouns (Bean 1983:139). A correlation between the presence or absence of case on the one hand and word order on the other is also typologically well-attested: SOV-languages show a strong tendency to have case, both in absolute terms and relative to SVO-languages (Greenberg 1966; Siewierska and Bakker 1996:137). Hence a shift from SOV word order to SVO in a certain language might give rise to the loss of case in that language, or, alternatively, a loss of case may cause a shift from SOV to SVO. In Hawkins’ account, SVO word order and morphological case are seen as two so-called ‘functional equivalents’ (Keenan 1978), i.e. two properties in a language that perform the same linguistic function. The relevant function is the marking of grammatical relations in a clause (henceforth ‘GR-marking’): both case and word order can be used to signal which element is the subject of the clause and which one is the object. This is quite obvious for case. SVO word order can replace case, cf. Hawkins (1986:48-49): Subjects and objects have to maintain their fixed position in order to be clearly recognisable as such. And verb position is the 2   To be accurate, Hawkins (1986:125) himself is quite hesitant to consider word order as more basic than the other parameters in the typology: “The word order parameter is thus a secondary consequence of morphological richness in particular (which is just one component of our semantic transparency parameter) and constitutes just a small part of the overall typology for which I am arguing.” Shannon (1990:52), however, in a paper about the place of Dutch in Hawkins’ typology, clearly assumes the presence of fixed word order to be the cause of some of the other phenomena: “due to the fixed word order of English, the only way to get certain NPs in the at times pragmatically preferred initial position is to make them subject.” See section 3.2 for a more detailed discussion.  G UNTHER D E V OGELAER  170 particular vehicle which most conveniently enables these basic grammatical relations to be expressed by means of word order: the subject occurs to the immediate left, and the object to the immediate right of the verb. I.e. the verb acts as an anchor. Hence, it is expected that not only English and German, but also the other Germanic languages should have at their disposal at least one of these two basic parameters to distinguish grammatical relations, either case or SVO word order. After having addressed some methodological issues (section 1.2), it will be shown that Dutch provides a problematic case for Hawkins’ typology. Using data from Dutch, a new typology will be proposed (section 2), which takes into account not only the distribution of case and different word order types in German, English, and Dutch, but in all the Germanic languages (section 3.1). In the rest of section 3, some problematic data from Continental Scandinavian and Afrikaans will be discussed. The main findings are summarised in section 4. 1.2 Methodological preliminaries This paper deals, essentially, with the distribution of a number of syntactic and morphological features, such as different word order types and case, in the Germanic languages. In the discussion of these phenomena, a Usage-Based approach will be taken, in which “substantial importance is given to the actual use of the linguistic system and a speaker’s knowledge of the full range of linguistic conventions, regardless of whether these conventions can be subsumed under more general statements” (Langacker 1987:494; cf. also the papers in Barlow and Kemmer 2000). For instance, for the parameter word order this means that it will not be attempted to determine one specific order that is more ‘basic’ than other possible orderings. Rather, the variability of word order will be taken into account. In doing so, the notion of a language having one basic word order feature will be abandoned. The two alternative parameters that will be used are the unmarked word order in main clauses and in embedded clauses, because, among the Germanic languages, both clause types are known to give rise to different orderings of the S, V, and O. Ideally, a Usage-Based approach to word order or case would include quantitative data on the frequency of the linguistic variants under investigation, which is, however, outside the scope of this paper. But one does not need quantitative data to be able to make a comparison between several genetically closely related languages. In many cases, the structural possibilities in one language form a “proper subset” (cf. Hawkins 1986:4) of the possibilities in one or more of the related languages. Hence it is clear that one of the languages must have expanded or limited the possibilities  E XTENDING H AWKINS ’ COMPARATIVE TYPOLOGY  171 which were common to the ancestral language. As the grammatical properties of some older stages of several Germanic languages are well-described, it can often be established quite easily how the variation within the Germanic languages has come about, and, consequently, in what respect some linguistic changes may have caused other ones. 2. Dutch: a counterexample? Dutch is an interesting test case for Hawkins’ typology, as presented above, in (1). Since Dutch is known to be intermediate between German and English in many ways (cf. Van Haeringen 1956), it is expected that, with respect to the grammatical features in (1), Dutch will sometimes pattern like English, sometimes like German, and sometimes show a ‘mixed’ behaviour. This is indeed the case for most parameters. To give only a few examples: with regard to grammatical morphology (other than case), Dutch has less genders than German and more than English, and the same holds for its plural markers and verb agreement markers. In addition, Dutch has a greater semantic diversity in its grammatical relations than German, but less than English. For instance, it allows (some) recipient subjects in passives. For a detailed comparison of German, Dutch, and English, with extensive references to Hawkins’ comparative typology, see Shannon (1990). As for the basic parameters in (1), word order and case, Dutch occupies an intermediate position as well. This is shown in (2), where the realisation is shown of Hawkins’ basic parameters, case and SVO word order. The Dutch case system is, like the English one, almost completely lost: NPs do not show case, and, in addition, in the pronominal system syncretism is rife as well. The parameter SVO word order is split into two different subparameters in (2), both of which distinguish Dutch (and Frisian and German) from English. 3  On the one hand, Dutch and Frisian main clauses have not abandoned the Verb Second-constraint, like German but unlike English, which has a strict SVO word order in main clauses (in contemporary English, V2 is kept only in a limited number of syntactic environments, such as questions and constructions with topicalised negative elements, whereas it used to be found more frequently in older 3  There are, of course, more differences. For example, as in German, in Dutch main clauses the infinitives and participles are placed at the end, forming a so-called ‘brace’ with the inflected verb in second position. Word order in German and Dutch differs as well, for instance with respect to the types of elements that may appear to the right of these infinitives and participles, i.e. out of the brace. These differences are not really important for the present purpose, though. The main issue here is whether subjects and objects can occur on different sides of the (inflected) verb.
Related Search
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