Jóhanna Barðdal
The Impersonal Construction in Icelandic/Scandinavian/Germanic and its Development
Abstract
The Impersonal construction is an Indo-European inheritance, which in Germanic has survived only in three languages, i.e. Icelandic, Faroese and German. New research has revealed that impersonal predicates are much more frequent in Modern Icelandic than, for instance, in Old Swedish, Old English and Modern High German. This raises the question whether the construction has become productive in the history of Icelandic or whether the construction had already fallen into disuse in the other Germanic languages during the Middle Ages. The goal of this project is to study which semantic fields and which predicate structures the impersonal predicates in Modern Icelandic belong to, to investigate the status of the construction in the history of Icelandic, and to survey its development in Germanic in general. Because of how original Icelandic is, basic knowledge of the construction is a necessary prerequisite for knowing which semantic fields should be investigated in particular in the other/older Germanic languages, and which predicate structures we may expect impersonal predicates to have. At that point answers can be generated on questions such as what was the productivity of the Impersonal construction, and what was the syntactic function of the oblique argument in the history of Germanic, i.e. did it behave like a syntactic subject, as in Modern Icelandic, or like an object. In addition, the project will cast light on productivity in general, but syntactic productivity is an emerging research area and very much unexplored.
Introduction
The main goal of this project is to carefully study the Impersonal construction in Icelandic, i.e. both the semantics and the predicate structure of the construction. This entails an investigation of verbs and predicates which select for oblique subjects, i.e. predicates which typically occur with accusative (mig dreymir 'I dream'), dative (mér leiðist 'I'm bored') or genitive subjects (mín missti við 'I was missing'). These kinds of predicates have received considerable attention, and much time and effort has been invested in studying them (see section on earlier research below). However, these investigations are more or less confined to the syntactic status of the preverbal oblique argument and/or the syntax of the construction, and have completely neglected the semantics of the construction. My project, on the other hand, will concentrate on the semantics of these predicates, it will investigate and map the various semantic fields the predicates belong to, and study the relation between the various groups. Also, I plan to study in particular the form, or the predicate structure, of the various impersonal predicates. I will investigate whether there is a difference in the semantics of the accusative, dative and genitive predicates. That is, I will study thoroughly each sub-construction in regards to whether there has been a semantic specialization within the sub-construction or whether the predicates all belong to the same semantic groups.Next I plan to carry out a comparative study between Old Norse-Icelandic and Modern Icelandic in order to investigate whether the construction has changed its status in the history of Icelandic.
The third goal of the project is to investigate the status of the construction in the Scandinavian/Germanic languages in general. Because of how well the construction has been preserved in Icelandic compared to the other Germanic languages, it gives us a unique opportunity to increase our knowledge of the structure of the other early Germanic languages. In that context I am particularly interested in studying the syntactic status of the preverbal oblique in early Germanic, since there are strong indications that the traditional analysis of the oblique as an object cannot be maintained. If it turns out that the oblique behaved syntactically as a subject that knowledge will demolish the traditional view of the research community on the structure of the earlier Germanic languages and demand a complete revision of our ideas on the structure of Old Germanic.
Finally, the last goal of the project is to increase our knowledge of the productivity of syntactic constructions to improve our possibilities of putting forth well-founded theories of productivity, which is an emerging research area.
Earlier research
Research of verbs/predicates, which select for oblique subjects in Modern Icelandic are more and less confined to the syntactic status of the oblique argument (see Zaenen, Maling och Thráinsson 1985, Sigurðsson 1989, Jónsson 1996, Barðdal 2001a, 2002). Regarding the semantics of the construction the prevalent view within the research community has been to regard these as "experiencer" and "benefactive" predicates (see for instance Rögnvaldsson 1983), i.e. predicates, which either express emotions or the interests of the referent denoted by the subject. Jónsson has observed in a recent paper (1997-98) that this is more complicated and that we have to assume that the subjects of these predicates can have the thematic roles "theme" and "patient", i.e. that the subject does not necessarily have to be an experiencer or a benefactive subject but that it can also refer to the entity/person which is affected by the event denoted by the predicate. He gives examples like "bátnum hvolfdi" 'the boat capsized' as a theme and "Sveini batnaði" 'Sveinn got better (from the disease)' as a patient. I have pointed out (Barðdal 2001b, Ch. 2) that this is an oversimplification of facts, and that in a rigorous investigation of the semantics of the Impersonal construction at least the following groups need to be discerned (see also Barðdal 2004):
Cognition verbs (vera kunnugt), Perception verbs (smakkast), Verbs of bodily states (vera flökurt), Verbs expressing changes in bodily states (batna), Emotion verbs (leiðast), Intention verbs (ganga e-ð til), Permission verbs (leyfast), Verbs of obligation (bera), Verbs of inclination (hætta til), Property verbs (liggja hátt rómur), Verbs of gain (áskotnast), Verbs of completion (takast), Verbs of hindrance (seinka), Verbs of advancing (fara fram), Verbs of diminishing (fækka), Verbs of saying (hrjóta af vörum).
This semantic classification is, however, not exhaustive since the classification of these predicates into semantic groups was not a part of my doctoral project. At the moment there exists a near complete list of impersonal predicates in Icelandic (Jónsson 1998). I intend to use that list as a base for my classification of predicates into semantic groups. In addition, there also exists a Cand. Mag. thesis by Bernódusson (1982), which gives a thorough description of the various predicate structures impersonal predicates have in Icelandic. Unfortunately, this thesis is both unpublished and written in Icelandic. I will, on the other hand, publish my research in English.Regarding the status of the construction in Germanic, recent research of the Impersonal construction in earlier English is confined to approximately 50 verbs (Allen 1995), and in earlier Sweden to approximately 40 predicates (Falk 1997). In Modern High German it is also assumed that the Impersonal construction is instantiated by approximately 100-120 predicates (Barðdal 2004). These figures are undeniably very different from the figures for Icelandic, where at least 1000 impersonal predicates exist. The question arises what the original situation was. If it is realistic to assume that the Scandinavian/German languages have a common ancestor, have the impersonal predicates then fallen into disuse in the other Germanic languages already during the middle ages, or has the construction become productive in Icelandic? A small pilot study in Chapter 7 in (Barðdal 2001b) shows that in the same amount of texts, within the same genre, impersonal predicates which select for preverbal datives occur more often in Old Norse-Icelandic than in Modern Icelandic: both the type frequency (72 vs. 48 predicates) and the token frequency (5.8 vs. 4.2%) is higher. This may suggest that the construction was more productive in Old Norse-Icelandic than in Modern Icelandic. It is also possible that predicates, which could occur in the Impersonal construction have simply not been properly identified in the earlier Germanic languages. In order to be able to answer these questions, primary work on the semantics and the predicate structure of the Impersonal construction in Modern Icelandic needs to be carried out. First at that point do we possess the knowledge essential for locating and identifying the potential impersonal predicates in the other earlier Germanic languages, and the knowledge essential for answering the questions that I have put forward here.
Subparts
The project can be divided into the following parts:I To categorize impersonal predicates in Modern Icelandic and divide them into semantic groups, and to account for the relations between the groups. I will here use the tools of lexical semantics and cognitive linguistics. Within that school it is assumed that knowledge of language is of the same nature as all other knowledge, and thus that linguistic categories exhibit the same properties as all other cognitive categories, namely that certain instances are central/prototypical, and that others are related to the central instances via links and networks.
II To account for the semantics of the oblique subject, and the semantic relations between the various instances of oblique subjects, i.e. both for each sub-construction and also between the accusative, dative and genitive sub-constructions. Here I will first and foremost make use of theories within cognitive linguistics, namely "grammatical semantics".
III To account for how the various predicates are formally composed, i.e. which predicate structures are involved. These can be simple verbs (leiðast 'be bored), particle verbs (reiða af 'do in life'), verbs selecting for prepositional phrases (óa við e-u 'be afraid of sth'), predicates where the main verb seems to be the linking verb vera 'be' as in vera kalt 'be cold'. As I have pointed out in an earlier paper (2001a) predicates with 'be' as a main verb seem to be of many types, for instance with an adjective (vera kalt 'be cold'), with a prepositional phrase (vera til lífs 'save sby’s life'), or with a noun phrase (vera vorkunn 'be pityful'). They can occur with only one argument, for instance a subject in dative (vera viðbjargandi 'be hopeless'), or with two core arguments, i.e. both a nominative and a dative (falla 'fall'). Sometimes the nominative is the pronoun það 'it' which anticipates a clause (svíða það sárt að… ‘find it difficult that…’). Sometimes it anticipates a full noun phrase, which is often an idiomatic part of the predicate (hverfa veröldin (nom) ‘lose sight of the world’). This part of the project will build upon the unpublished Cand. Mag thesis of Bernódusson (1982), which gives a thorough description of the various predicate structures available to impersonal predicates in Icelandic. As I said earlier, unfortunately Bernódusson's work is both unpublished and written in Icelandic, but I will publish my work in English, and thus make it available to a broader research community.
IV To investigate which impersonal predicates in Icelandic behave in such a way that either the nominative stimulus or the dative human argument can function as syntactic subjects (falla 'fall'). Not all dat-nom verbs in Icelandic behave in such a way. This can be illustrated with the fact that it is possible to say both Þetta hefði fallið mér vel and Mér hefði fallið þetta vel, whereas the former construction is not possible with líka 'like': *Þetta hefði líkað mér vel. It is possible that this group of alternation was larger in Old Icelandic than in Modern Icelandic, and that the size of the group, and the exact inventory of the group, can vary for different speakers of Icelandic. For this purpose I will scrutinize Old Icelandic texts and do an experimental study with modern Icelandic speakers.
V To investigate the status of the construction in the Icelandic language history. A small pilot study in my dissertation (Ch. 7) has shown that in the same amount of texts, within the same genre, predicates selecting for preverbal datives occur more often than in corresponding modern Icelandic texts: their type frequency is higher (72 vs. 48 predicates) and their overall token frequency is also higher (5.8 vs. 4.2%). This may indicate that the construction was more productive in Old Icelandic than in Modern Icelandic. More research is needed on this.
VI To do a careful investigation of the development of the Impersonal construction in a wider context than only in Icelandic. That is, to study the development of the construction in Scandinavian/Germanic. This will entail a scrutiny of Older Germanic texts, including Gothic, and an investigation of the status of the construction in Modern German (Barðdal 2002, Eythórsson and Barðdal 2003). In this context the syntactic status of the preverbal oblique in Old Norse-Icelandic is very interesting. It has been argued that the oblique is a syntactic subject, like in Modern Icelandic (Rögnvaldsson 1996, Barðdal 2000, Barðdal and Eythórsson 2003). The question then arises what the syntactic status of the oblique was in the other Scandinavian/Germanic languages. Since the construction still exists in Modern German it is possible to do more research on it (see Barðdal 2002). This part of the project will be carried out in collaboration with Thórhallur Eythórsson, a historical linguist at the University of Manchester.
Gain
This project will generate primary knowledge of the Impersonal construction in Icelandic, both its predicate structure and semantics. That is a necessary prerequisite for continuing research on the Impersonal construction in Scandinavian/Germanic. Such knowledge will both be beneficial to me in my research and also to all other non-Icelandic researchers interested in the topic, which have not access to the Impersonal construction in their native languages and not necessarily any access to the Impersonal construction in the original form of the construction in Old Norse-Icelandic and/or Modern Icelandic.It is also important to investigate the syntactic status of the preverbal oblique in the earlier Germanic languages, because if it turns out that the oblique behaved syntactically like a subject, as it does in Modern Icelandic and Modern Faroese, that will radically alter our view of the structure of the older Germanic languages, and completely alter the premises for historical research.
Increased knowledge of the Impersonal construction can be encouraging for new/continued research on the Impersonal construction in the other Germanic languages, and thus, make data available for comparative research on the development of the construction in Germanic. It will also generate knowledge of the productivity of the construction and changes in the productivity of the construction in the history of the Icelandic/Scandinavian/ Germanic languages. That knowledge may, again, be used to contribute to the development of theories of productivity, and language change in general.
I will publish my research in English and thus participate in the international research on historical linguistics, Scandinavian linguistics, Germanic studies, Construction Grammar and cognitive linguistics. I intend to write two books within this project. The first book will treat the Impersonal construction in Icelandic, its semantics, predicate structure, and the development from Old Icelandic. The second book, written in collaboration with Thórhallur Eythórsson, will investigate the development of the construction in Scandinavian/ Germanic.
References:
Allen, Cynthia. 1995. Case Marking and Renalysis: Grammatical Relations from Old to Modern English. Clarendon Press, Oxford.Barðdal, Jóhanna. 2000. Oblique Subjects in Old Scandinavian. NOWELE 37:25-51.
Barðdal, Jóhanna. 2001a. The Perplexity of Dat-Nom Verbs in Icelandic. Nordic Journal of Linguistics 24:47-70.
Barðdal, Jóhanna. 2001b. Case in Icelandic – A Synchronic, Diachronic and Comparative Approach. Lundastudier i Nordisk språkvetenskap A57. Department of Scandinavian Languages, Lund.
Barðdal, Jóhanna. 2002. "Oblique Subjects" in Icelandic and German. Working Papers in Scandinavian Syntax 70:61-99.
Barðdal, Jóhanna. 2004. The Semantics of the Impersonal Construction in Icelandic, German and Faroese: Beyond Thematic Roles. To appear in W. Abraham (ed.). Sprachtypology und Universalienforschung [Language Typology and Universals] 57-2.
Barðdal, Jóhanna and Thórhallur Eythórsson. (2003). The Change that never Happened: The Story of Oblique Subjects. To appear in Journal of Linguistics 39.
Bernódusson, Helgi. 1982. Ópersónulegar setningar [Impersonal Sentences]. Meistara- ritgerð, Háskóla Íslands.
Eythórsson, Thórhallur and Jóhanna Barðdal. 2003: Oblique Subjects: A Germanic Inheritance. Working Papers in Scandinavian Syntax 71:145-202.
Falk, Cecilia. 1997. Upplevarverb i fornsvenska [Experiencer Predicates in Old Swedish]. Lund University Press, Lund.
Jónsson, Jóhannes G. 1996. Clausal Architecture and Case in Icelandic. Doctoral Dissertation, University of Massachusetts, Amherst.
Jónsson, Jóhannes G. 1997-98. Sagnir með aukafallsfrumlagi [Predicates selecting for Oblique Subjects]. Íslenskt mál og almenn málfræði 19-20:11-43.
Jónsson, Jóhannes G. 1998. A List of Predicates that taka a Quirky Subjects in Icelandic. Ms, University of Iceland.
Rögnvaldsson, Eiríkur. 1983. Þágufallssýkin og fallakerfi í íslensku [Dative Sickness and Case in Icelandic]. Skíma 16:3-6.
Rögnvaldsson, Eiríkur. 1996. Frumlag og fall að fornu [Subject and Case in Old Icelandic]. Íslenskt mál og almenn málfræði 18:37-69.
Sigurðsson, Halldór Á. 1989. Verbal Syntax and Case in Icelandic. Doctoral Dissertation, Lund University.
Zaenen, Annie, Joan Maling and Höskuldur Thráinsson. 1985. Case and Grammatical Functions: The Icelandic Passive. Natural Language and Linguistic Theory 3:441-483.
| Oppdatert 16.03.2006 av JB |
|
