English As A Germanic Language
English As A Germanic Language
English As A Germanic Language
University of Antwerp
Inhoudsopgave
CHAPTER 0: PREFACE.....................................................................................................6 1. a. b. 2. a. b. c. d. e. Contrastive Linguistics..........................................................................................6 A more theoretical approach.............................................................................6 Hawkins (1986) and transformational grammar................................................6 Grammatical Morphology......................................................................................7 Examples?..........................................................................................................7 What is counted?...............................................................................................7 Nominal Morphology..........................................................................................7 Word Order freedom..........................................................................................8 Case Syncretism................................................................................................8
CHAPTER I: Introduction: the theoretical interest of English/German contrasts...........10 1. English/German contrasts in relation to Universal Grammar and Language Typology....................................................................................................................10 a. b. c. 2. a. 3. Universal Grammar (Chomsky)........................................................................10 Typological Universal Grammar.......................................................................10 Hawkins...........................................................................................................10 The unity of English/German contrasts...............................................................11 Keenan.............................................................................................................11 Methodology.......................................................................................................12
CHAPTER II: Grammatical Morphology..........................................................................13 1. 2. a. b. 3. Inflectional morphology: a contrastive overview.................................................13 Case marking in the noun phrase.......................................................................13 Morphology......................................................................................................13 A contrastive summary...................................................................................15 The use of the cases...........................................................................................16
2
a. b. 4. 5.
Case assignment in phrasal categories............................................................16 The predictability of case assignment.............................................................16 The semantic consequences of the loss of inflectional morphology...................17 Lexical morphemes and derivational morphology..............................................18
CHAPTER III: Word order freedom.................................................................................20 1. 2. 3. Clause-internal word order freedom: a contrastive illustration...........................20 Causes and Consequences of the Word Order Freedom Contrasts.....................21 Case syncretism and verb position.....................................................................23
CHAPTER IV: Basic grammatical relations and their semantic diversity.......................24 1. 2. 3. 4. 5. Some basic data.................................................................................................24 More on subjects.................................................................................................25 Theoretical preliminaries....................................................................................25 The grammatical relations generalisation...........................................................25 Causes and consequences of the contrasts in semantic diversity......................26
CHAPTER V: Raising structures.....................................................................................29 1. a. 2. 3. 4. Contrastive summary..........................................................................................29 What is raising?...............................................................................................29 The semantic interpretation of Raising structures..............................................30 Causes and consequences of the contrasts in Raising structures.......................31 A note on the Subject-to-Object Raising Dispute................................................31
CHAPTER VI: Extractions...............................................................................................32 1. a. b. 2. 3. 4. Contrastive summary..........................................................................................32 General regularities.........................................................................................32 Problem............................................................................................................33 The Argument Trespassing Generalisation.........................................................34 Pied piping..........................................................................................................34 Causes and Consequences of the Extraction and Pied Piping Contrasts.............35
CHAPTER VIII: The unity of English/German contrasts: a realignment in the mapping between form and meaning.........................................................................................38 Contrastive typology: summary................................................................................38 CHAPTER IX: Verb-final order in German......................................................................39 1. a. b. 2. a. b. Verb-final structures in German..........................................................................39 German............................................................................................................39 English and Dutch in comparison to German...................................................39 Contrastive summary and some consequences..................................................39 Case 1: Complementation and Periphrasis......................................................40 Case 2: Echo Sentences...................................................................................40
CHAPTER X: Leaking behind the verb in German.........................................................41 a. b. c. 1. 2. 3. 4. 5. Leaking.........................................................................................................41 Types of clauses..............................................................................................41 Which constituents are easier to move to the right?.......................................41 Extraposition.......................................................................................................42 Postverbal movement of constituent other than the subject..............................43 Extraposition from NP.........................................................................................44 Rightward movement in English.........................................................................44 Contrastive summary..........................................................................................45
CHAPTER XI: Verb-second structures in German and English.......................................46 1. a. 2. a. b. c. d. e. 3. Verb-second in German......................................................................................46 Conclusion.......................................................................................................46 Verb-second in English........................................................................................47 Subtype 1: subject-auxiliary inversion.............................................................47 Subtype 2: Subject- simple verb inversion.......................................................47 Subtype 3: Inversion after Locative Phrase Preposing.....................................47 Subtype 4: Comparative Substitution..............................................................47 Subtype 5: Participle Preposing.......................................................................48 Contrastive summary..........................................................................................48
4
4. 5.
CHAPTER XII: Verb-first structures in English and German...........................................50 1. a. 2. 3. 4. Verb-first in English.............................................................................................50 Comparison of verb-first and verb-second in English.......................................50 Verb-first in German...........................................................................................50 Contrastive summary..........................................................................................50 Causes and consequences of the Verb-First contrasts........................................51
CHAPTER 0: PREFACE
1. Contrastive Linguistics
1940s in the context of second language teaching Interference Transfer: - false friends
E.g. If I asked, he would E.g. *If I would ask, he would Devoicing of word-final stops in Dutch vs. English + aspiration; in English vs. Dutch a. A more theoretical approach. Eastern Europe; Poland, Romania, English is a popular language Hawkins (1986); BrE linguist link with typology few features, many languages Hawkins many features, few languages Holistic approach: unifying the contrasts, 1 answer. Corpus Linguistics. E.g. translations of modals in books of Harry Potter; all different types of grammatical translation. The more to the West, the more auxiliaries that are being used. Global English, staying and growing bigger as the worlds dominating language
a. Hawkins (1986) and transformational grammar Surface: (phonology, morphology, syntax) Semantic representation Standard Theory Chomsky 1965
2. Grammatical Morphology
In general, all the grammatical distinctions that are drawn within English inflectional morphology are drawn in German as well, though not vice versa.
a. What is counted? Grammatical distinctions: slots in (nominal, verbal, ) paradigms Not the constructions in any one slot German vs. English plural Not the number of lexical items to which the distinctions apply E.g. German Onkel vs. English uncle(s)
a. Nominal Morphology The case morphology of German properly includes that of English Form is one thing. Function is another So, what is it that determines the use of the cases? Predicates Prepositions E.g. Frans Plank Ich winkte ihm. Ich rief ihm zu. +Dat. +Dat.
I waved him hither. I called him. English not quite unproblematic either [+Acc.] pronoun where one would expect a [+Nom.] one. Drift towards the invariable word (Sapir 1921)
7
English morphology is more vague Pretty close to the bedrock of morphological distinctions that all languages need to draw, and beyond which syncretism cannot penetrate.
Plank has three points Systematic relation between degree of specificness in the selectional restrictions of predicates and specificity of basic grammatical relations <Irrelevant> German with prefixes vs. English with stems
a. Word Order freedom Clause-internal order only German is freer Freedom in German, not in English Freedom rule in German and English but more productive in English
Why? Principle of covariation of functional equivalents. The more we assign a language overt case marking the freer can be its basic word order and conversely. - Keenan 1978 In German word order variation is related to pragmatics German unmarked IO + DO DO + IO only when DO is definite/theme Unmarked DO + PP DO + IO only when DO is definite/theme And in English word order?
a. Case Syncretism Related to word order fixation, also to SVO order? Greenberg 1996 - Universal 41: If in a language the verb follows both the nominal subject and the nominal object as the dominant order, the language almost always has a case system. OV OV VO VO + Case - Case + Case - Case English German, Old English <rare, but still Ijo>
Word order change, then loss of case Word order change simultaneous with loss of
8
Need a language for contrastive point of view ~ Sapir-Worf theory E.Sapir wrote Drift This isnt random at all; subtitle of the book = Unifying the contrast. (=Typology) Relationship: Many contrast are related; Form German: Direct complex form, simple meaning English: Indirect simple form, complex meaning
Meaning
the
theoretical
interest
of
Absolute universals: define grammatical properties that are exemplified in every language. Generative grammarians: since universals are properties that all languages share, they can be derived from the study of any one language.
a. Typological Universal Grammar = universals are characterizations of the regularities in the ways languages may differ from one another/structures which differ from one another are among the primary objects of study. Classifications of languages made in terms of the properties or property clusters mentioned in the variation-defining universals are then referred to as typologies. Cross-linguistic.
b. Hawkins Motivation: search for principles that underlie cross-language variation Approach: look at a large number of variant linguistic properties in a small number of languages.
It will be argued that the phonological changes that destroyed the English case system were the ultimate trigger that set in motion the syntactic changes leading to the present contrasts. 3 reasons for comparing two languages Property A Property B
10
Property C English and German are both Germanic languages; the contrasts between them do not involve small changes in limited grammatical areas, but profound readjustments across all the major areas of the grammar. Property D Hawkins shall present evidence that there is a descriptive generalisation which unites these major areas of contrast.
Consequences: It suggests that the historical changes that gave rise to current contrasts followed a single typological direction. If these changes had been more random, with no common directionality over time, we would have expected a much less systematic relationship between the two languages than we actually get. Sapir: notion of drift
11
Phonetic attrition (e.g.: Den Man liebt die Frau (woman does the loving) Case syncretism (*The man loves the woman (Accusative falls away, hard to tell whos the subject)) Syntactic changes (The woman loves the man) Synchronic: both sensible compromises between productive and receptive ease Diachronic: morphosyntactic changes
1. Methodology
Primarily inductive in this book, data-driven. Although its important, when comparing two languages, to employ common theoretical assumptions in the descriptions of both. Describe the differences Arrive at generalisations Explain everything Synchronic Diachronic
12
Old English has a morphology very similar to that of German today. And whereas German has preserved its inflectional system largely intact, English has reduced it. Movement of English towards case syncretism. Grammatical distinctions are the slots in the paradigm. E.g. singular plural in Dutch, English and German: German Sing/Plur al Plural Forms 2 (Sing & Pl) English 2 Dutch 2
1 more form
Old English had a system of case inflections that was similar in major respects to what we find in Modern German. Genitive in Modern English: no changes are made. Theres also the possibility of the double genitive a.k.a. the group genitive The expressive power of definite article case is identical to that of all other determiner + noun sequences, and also to that of all determiner + adjective + noun and zero-morpheme + adjective + noun sequences. Weakest distinction: nominative and accusative; feminine and neuter.
13
Morphology of nouns or determiners is not enough to understand the meaning of the sentence. And its that what German shows the most.
14
Dutch
English
>
2
de, het
<
1
the should be one, only different realisation (allomorphs? )
Indef. Art. 6
>
<
2 or 1
een
a, an
Adj.
6
guter, gutes gut, guten gute, gutem,
2
goed, goede
1
good
Forms for each slot; English has lesser expressive power than German.
15
a. A contrastive summary Distinctions in the personal pronouns: Dutch turns out to be more an extreme in the case of the pronouns. It sometimes has reduced forms, which leads to more pronouns. Case marking on German pronouns is more expressive than it is in noun phrases with nominal heads. The English pronoun system is much richer, however, and makes the following formal distinctions: I/me/mine, you/yours, he/him/his, she/her/hers, it/its, we/us/ours/, they/them/theirs and who/(whom)/whose.
What does it mean to be in the middle?: Different interpretations (see the counting of pronouns, etc.) Sometimes not even in the middle, where German or English are in the middle. e.g.: Dutch doubles its pronouns; mij/me, hij/ie, ik/k (no allomorphs)
16
Acc. Acc.
Dat. Dat.
Acc.
Gen.
17
2-place 3-place
Nom. Nom.
Acc. Acc.
Dat. Dat.
a. The predictability of case assignment In generative grammars of Modern German, the case requirements of verbs and also of prepositions and adjectives are specified lexically, i.e. as part of the non-predictable grammatical information which is stored together with a lexical item in the lexicon
18
German has not remained unchanged either, of course: there has been some morphological syncretism; there has been a decline in the use of certain form. Certain well-established forms, like the plural form, have resisted the drift to invariable words, possibly because they symbolize certain stronger form cravings we do not yet fully understand. Invariable words and the absorption of semantic nuances, result in greater ambiguity (or vagueness) for the morphemes in question. I.e. the range of meanings covered by English grammatical form classes is larger than it was, and larger than in German. We need a linguistic context E.g. where do you live? where do you go? On the other hand, its interesting that certain morphological distinctions have not been lost in English. These semantic distinctions are clearly felt to be more than a nuancing and so syncretism has been avoided. Pretty close here to the bedrock of morphological distinctions that all languages need to draw, and beyond which syncretism cannot penetrate. (Not true: Chinese can function without plural forms!)
What Sapir appears to be drawing attention to is a general realignment in the assignment of meanings to linguistic forms. Languages can vary in the extent to which numerous semantic distinctions are explicitly drawn in surface morphology. Not to draw these distinctions in surface doesnt render them inexpressible and doesnt make the language in any sense impoverished it simply means that their expression is carries by fewer linguistic forms, each with broader semantic coverage (i.e. more ambiguity or vagueness). And the precise interpretation that is explicitly marked in the one language will typically be pragmatically derived in the other, or may on occasion not be drawn at all. One more comment: one more distinction that English kept is the case in (some) pronouns E.g. Who did you see? Whom did you see? I saw him. He saw me But! John and I/me want to like to add that Who wants to leave now? *I/Me!
19
We consider the difference between affected and effected objects Effected objects: objects that are brought into existence by the action described in the predicate Affected objects: antecedently existing objects that are involved in the action described by the predicate.
No distinction in English, whereas German makes one. What about Dutch? Sometimes, it distinguishes affected and effected objects, like German, but sometimes it doesnt, like English. Plank (1980) has three very general and interesting points to make: He claims that languages like German whose predicates impose tighter selectional restrictions on their arguments, will also be those languages whose basic grammatical relations are well differentiated semantically, i.e. which associate relatively specific semantic roles with these grammatical relations. The degree of specificity in the selectional restrictions of the predicates of a language is systematically correlated with the degree of semantic specificity of its basic grammatical relations. The tighter its selectional restriction, the narrower the range of semantic interpretations for its basic grammatical relations; the more neutral or looser in its selectional restrictions, the more semantically diverse its basic grammatical relations.
20
He argues that only those noun classes that are identified in classifier languages will figure in the selectional restrictions of non-classifier languages such as English and German. He observes that the different selectional restrictions for otherwise synonymous verbs are often signalled in German by derivational prefixes on otherwise identical stems, rather than by varying the lexical stem itself. In English, by contrast, such differences in selectional restrictions are more often expressed by changing the lexical stem itself, and as we have seen English will regularly not impose the relevant selectional restriction where German must draw the distinction in question.
21
word
order
freedom:
contrastive
Starting point: it is often observed that word order in German is freer than in English.
Syntactic rules: Clause-internal bounded movement Dative accusative shift (Eng.) Adverb movements Subject hopping (Ger.) The relatively greater freedom of these movements in German does NOT extend to rules crossing clause boundaries, and hence we need to restrict our claim about the greater word order freedom of German to clauseinternal movements only.
German has movement rules effecting word order rearrangements which are without parallel in English E.g. Subject Hopping: subject hops to the end of the sentence Zu Weihnachten gab Peter dem Bruder das Buch Zu Weihnachten gab dem Bruder Peter das Buch. Zu Weihnachten gab dem Bruder das Buch PETER
24 possible relative orderings of the four constituents, i.e. Peter, zu Weihnachten, dem Bruder and das Buch Results: With the appropriate contrastive stress on the German sentences, all 24 were judged grammatical Without the assistance of contrastive stress, some eight of the 24 were judged ungrammatical, with 16 exhibiting varying degrees of grammaticality. English translations: only 9 were considered as grammatically correct. Dutch translations: only 12 were considered as grammatically correct. Dutch is more like English than like German.
English SVX
Dutch SVX
German SVX
22
XSV
XVS
X V (X) S
Subject Hopping
23
Where there are similar movement processes, they apply more productively in German than in English E.g. Verb-Second Only if you are rich will I If you are rich, I will Nur wenn du reich bist, werde ich Wenn du reich bist, werde ich
24
S VP
25
The relative rigidity of word order in Modern English has an interesting consequence for the relationship between form and meaning. Languages whose morphology makes possible the kinds of word order permutations that we have seen in German can use these options for pragmatic purposes. Now, assuming that such pragmatic differences exist, it follows that the more fixed word orders of English are correspondingly more ambiguous with respect to these pragmatic functions. I.e. many pragmatic distinctions which receive their own syntactic encoding in German and Russian etc., dont receive distinctive encoding in English and one and the same syntactic form of English ranges over pragmatic differences in meaning which can be disambiguated in these languages.
Critical comment: English does have syntactic strategies for these kinds of pragmatic effects. Cleft construction: E.g. It is Maxim who defends Viktor The one that defends Viktor is Maxim. Pseudo-cleft construction: E.g. Who defends Viktor is Maxim
But, these also exist in German so the general point about the vagueness of English still stands. The pragmatic word order principle is discussed in terms of the distinctions between Theme and Rheme. Given = Theme New = Rheme. The more fixed word orders of English frequently lead to a reversal of this pragmatic principle, and hence to the kind of pragmatic ambiguity in syntactic structure that we saw in the English S V O structures. Critical comments: The terms Theme and Rheme are not well-defined. There appears to be other pragmatic functions performed by word order variants which have been lumped together under a simplifying ThemeRheme rubric. The more fixed word orders of a language like English are correspondingly more ambiguous (or vague) in surface structure syntax with respect to these functions, compared to German and the Slavic languages.
26
Critical note: There is a dependence: claim 1 follows from claim 2. Claim 2 needs additional argumentation (Sapir (1921)): I.e. the verb acts as an anchor, with the two most frequent and important grammatical relations fixed in their position relative to it.
Object +Acc.
left of V
right of V
The situation is complicated. On the one hand, we have Greenbergs statistical universal. (If in a language the verb follows both the nominal subject and nominal object as the dominant order, the language almost always has a case system.). This is equivalent to saying that that if a language does not have a surface case system, it will (almost always) not have SOV order and hence the loss of a case system historically can be used to explain The loss of SOV structures in the history of English and; The resulting contrasts with Modern German
27
On the other hand, Greenbergs universal is statistical only: Ijo is an example of an SOV and non-case-marked language. Both the loss of case and the word order change took a long time
No agreement
Comparing English and German, we can conclude: That German dative- marked NPs should not be regarded as direct objects, and that accusatives NPs only should be equated with direct objecthood That the class of direct objects in English is larger than in German, since it comprises both the accusative-marked and dative-marked NPs of German That rules such as passive (NP preposing) which are sensitive to direct objecthood apply in a greater number of environments in English than in German to create derived subjects (which can then undergo rules sensitive to subjecthood
The experiencer in these 1- or 2-place constructions is represented by a +Dat. or by an +Acc. in German, whereas English regularly employs a (nominative) subject.
28
Curme: the accusative denotes the direct object or thing affected or produced, and the dative the indirect object, the person or thing to whose advantage or disadvantage the action accrues.
It is clear that semantic distinctions are operative here and that German is systematically excluding certain semantic roles from being mapped onto the (accusative) direct object relation with the predicate in question. By contrast, English regularly collapses the semantic roles of both accusative- and dativetaking verbs into a common grammatical entity, direct object and into a common surface form.
1. More on subjects
The subject-forming possibilities of German are in general a proper subset of those in English, since non-agentive semantic roles in German frequently resist being mapped onto subjects where this is possible in English. Both in terms of grammaticality and in terms of frequency, therefore, the subject-forming possibilities of non-agentive arguments in English are greater than those of German.
2. Theoretical preliminaries
Not to be studied
29
No longer connected to case syncretism = DRIFT. But why? Kirkwood (1978) offers a plausible speculation for why the set of English subjects should have increased beyond the set predicted by case syncretism alone. His argument rests on the following two simple and indisputable facts: Modern English has more fixed S V O word order than it used to; and yet there are certain very general pragmatic principles of word order rearrangement (e.g.: Theme before Rheme) which seem to be present to some extent at least in all languages.
Result: form of conflict: the syntactic rules force fixed word order, pragmatic principles favour word order rearrangements in accordance with the information structure of a whole discourse. Conflict resolved by permitting more types of NPs to actually be subjects and hence to occupy immediately preverbial position
Syntax Pragmatics
Semantics
Semantic of S
diversity
31
English Case Syncretism Word order fixation S V O fixation Semantic expansion of S Makes historical sense too! Distinction between CODING and BEHAVIOUR: YES YES YES YES
German NO NO NO NO
32
German and English are at the opposite poles of a historical drift for which other Germanic languages (including earlier English itself) provide intermediate stages. The morphological coding of subjecthood in German is in one-to-one correspondence with the syntactic notion of subject defined behaviourally. In earlier English the class of such syntactic subjects expands and there is then a temporary mismatch with surface coding, until surface cases are brought back into realignment with the syntactic notion. There are also different types of behaviour. Or the implication is just premature or wrong. First behaviour, then coding or at the same time is possible too. So 2) is ruled out, not first coding and then behaviour. First it behaves as a subject, then its dressed up as a subject. Moreover, notice that the period of mismatch between morphology and syntax in the history of English (i.e. the dative subject period) is not very long-lasting. Throughout the Germanic languages as a whole, the periods of correspondence between morphology and syntax are much more protracted than those of noncorrespondence. Theres a pattern emerging: English morphemes were argued to be systematically broader in semantic range; basic sentence patterns of English were argued to be ambiguous with respect to pragmatic distinctions that can explicitly drawn in German. And now to basic grammatical relations of English have been shown to have a systematically greater semantic range than their German counterparts. In all three cases, English is packing more meanings into common surface forms in both morphology and syntax, with corresponding greater ambiguity. Recapitulation
Nominal Inflection Pronouns Verbal Semantics WO rigidity SVO Der Subj2-place Der Subj3-place Basic Subject
Dutch
German German Dutch German German Dutch German German Dutch Dutch German German Dutch
Dutch
Dutch
33
Object to subject raising: the object of the subclause becomes the subject of the main clause. Critical comment: Maybe S-O Raising is something else! Theres a semantic relation between V and S too. Related to passivization. English and Dutch are similar concerning 2-place derived subjects too. Also called Tough movement E.g. It is easy to convince John. John is easy to convince.
German 5 DO Bounded
34
S to S S to O O to S
Dutch Dutch
Bounded or unbounded? E.g. English: This book is easy for me to force Charles to read. Dutch: *Dit boek is gemakkelijk voor mij om Karel te dwingen lezen. German: *Dieses Buch ist leicht fr mich Karl zu lesen zu zwingen. D like E D like G Conclusion prepositional objects boundedness
35
English is more vague than German. In German, theres a less structural ambiguity.
1. NP-V to VP
S to S Raising EQUI
John ceased to be ill. John wanted to be ill. I believe John to be ill. I persuaded John to be ill.
2. NP-V-NP to VP
S to O Raising EQUI
36
3. NP-be-Adj. to VP
This is reminiscent of the greater ambiguity of SU-V and SU-V-O structures in English, which also involve greater semantic diversity for basic grammatical relations. What is meant by ambiguity? When a NP-V-NP-to-VP can be mapped onto more than one predicate type in semantic representation.
If John is just a normal subject, it remains unclear why German should not allow it. Also not clear why English does allow it.
37
Unbounded movement rules such as Relative Clause Formation etc apply in more environments in English than in German. In fact, it will be argued that the contrastive situation is as precise as in the raising examples: wherever German can extract, so can English, but not vice versa. Why unbounded? Because things can move over intermediate clauses. The English and German extraction data between them enable us to set up a hierarchy of increasing extraction difficulty. a. General regularities Extraction out of an infinitival object complement of a 2-place predicate is always straightforward in both languages. E.g. The man who I have tried to kill was your friend. Der Mann, den ich zu tten versucht habe, war dein Freund. Extraction out of an infinitival object complement of a 3-place predicate is always possible in English. The corresponding German extractions are typically good. But when the sentence is slightly more complex, German native speaker judgments begin to be uncertain. E.g. Who has he requested you to marry? Wen hat er dich gebeten zu heiraten? Extractions out of finite object complements are possible in English, but typically impossible in German. E.g. I dont know who the police thought that the guilty man was. * Ich wei nicht, wer die Polizei glaubte, da der Schuldige sei. Extractions out of non-strictly subcategorised infinitival phrases are completely impossible in German, but possible in English, though English native speakers begin to experience unease at this point. E.g. What did he come to order to pick up? * Was ist er gekommen, um abzuholen? Extractions out of non-strictly subcategorised finite clauses (clauses who can only be surrounded by certain words) are impossible in both languages. E.g. * Which film did you go to the movies, even though you didnt want to see? * Welchen Film bist du ins Kino gegangen, obwohl du nicht sehen wolltest?
38
a. Problem Two dimensions Internal morphology: infinite vs. Finite External morphology: role of clause in higher clause
German Infinit > Finite e Obj 2-place Obj 3-place Non-obj subj Subj + + non- ?/-
English Infinit > e Obj 2-place Obj 3-place Non-obj subj Subj + + non- ? / Finite + + -
non- -
40
1. Pied piping
VP Pied Piping E.g. Der Mann, den ich zu tten versucht habe. Der Mann, den zu tten ich versucht habe. *The man whom to kill I have tried. *De man wie te vermoorden ik heb geprobeerd. Possible in German, not in English. Notice that the effect of VP Pied Piping is to keep the VP den zu tten intact as a constituent in surface structure, by fronting the whole thing. As a result, the immediate predicate of den remains tten, the verb with which it contracts a semantic relation. PP Pied Piping E.g. * Die Frau, der ich ins Kino mit ging. Die Frau, mit der ich ins Kino ging. The woman, who I went to the movies with. The woman, with whom I went to the movies. De vrouw, waar ik naar de bioscoop mee ging. De vrouw, met wie ik naar de bioscoop ging.
41
German must use a pied piped structure, whereas English may or may not
42
NP Pied Piping E.g. * die Frau, deren ich Tochter kenne. Die Frau, deren Tochter ich kenne. *The woman, whose I know daughter. The woman, whose daughter I know. *De vrouw , van wie ik ken de dochter. De vrouw, wiens dochter ik ken. Pied Piping is obligatory in both languages. The grammar of German, which is in general resistant to argument trespassing, chooses to make the application of VP Pied Piping obligatory precisely in more complex structures where argument-predicate matching is that much more of a problem.
German Dutch English VP Pied Piping +/PP Pied Piping + NP Pied Piping + Dutch is the same as English! +/+/+/+/-
43
rather than anaphoric pro E.g. I know that but he doesnt. Ich wei, dass aber er wei es nicht Ik weet the, maar hij weet het niet. Possible in English, but not in German and Dutch. Absence of agent E.g. This tent sleeps four. In diesem Zelt schlft man zu vier. In deze tent kunnen vier mensen slapen. Possible in English, but not in German and Dutch instead of reflexive E.g. The door opens. Die Tr ffnet sich. De deur opent. Possible in English and Dutch, but not in German
44
relative E.g. The woman I love is coming tonight. Die Frau, die ich liebe, kommt heute abend. E vrouw die ik liefheb, komt vanavond. Possible in English, but not in German and Dutch. instead of Prep Pron E.g. I am convinced that Ich bin davon berzeugt, dass Ik ben ervan overtuigd, dat
English 1. Conjunction reduction - NP + 2. rather than anaphoric + pro 3. Absence of Agent 4. instead of reflexive 5. relative + + +
Dutch + + -
German +/-
6. instead of Prep Pron + Dutch is more like German than like English.
45
CHAPTER VIII: The unity of English/German contrasts: a realignment in the mapping between form and meaning
Contrastive typology: summary
German More grammatical morphology More specific selectional restrictions More word order freedom Less semantic diversity of GRs Less raising More pied piping Less deletion (of NPs) Self-study
English Less grammatical morphology Less specific selectional restrictions Less word order freedom More semantic diversity of GRs More raising Less pied piping More deletion (of NPs)
46
Main clauses: verb-second and verb-first constructions Subclauses: verb-final but semantic exploitation a. Case 1: Complementation and Periphrasis
b. Case 2: Echo Sentences Dutch is like German Whether this is the right explanation or not, the important point is again that it is German rather than English which is maintaining a clean division between main and subordinate clause types, and which is exploiting this division for the expression of pragmatic differences in meaning which are not carried in English. E.g. Wie heit den dein Bruder? Wie dein Bruder heit?
If the question is not understood by the hearer, the speaker will normally repeat it using subordinate clause word order. Thereby saying in effect this was the question I asked you just before. Verb opposition is the only surface marker of main verses subordinate clause status: there is no higher verb separated from the embedded proposition by comma intonation; and the complementiser wie is identical in both cases. English translations are only grammatically correct with verb-second and verbfirst orders.
48
a. Which constituents are easier to move to the right? It is easier to extract out of the complement of a two-place predicate than a three-place predicate, whereas a three-place predicate complement is more readily extraposable into rightmost position. (10.1a) Similarly infinitival complements, which are less heavy than finite complements, are easier to extract out of, but more difficult to extrapose(10.1b) And a strictly subcategorized clause is easier to extract out of than a nonstrictly subcategorized clause, whereas this latter is more readily extraposable. (10.1d) The one case of disagreement between the two hierarchies is (10.1c): a nonsubject complement is easier to extract from than a subject complement, and is also more readily extraposable into rightmost position. (OBL is any object except DO)
49
50
1. Extraposition
Extraposition of a subject clause in one-, two- and three-place predicate constructions is regularly optional in German, both when the clause is a tensed subordinate clause containing a verb and with zu + infinitive clauses. E.g. Dass er seinen Hund erschossen hat hat uns alle berrascht Es hat uns alle berrascht, dass er seinen Hund erschossen hat. Abends in der Wirtschaft Bier zu trinken hat uns gefallen. Es hat uns gefallen, abends in der Wirtschaft Bier zu trinken.
For extraposition of DO complements we need to distinguish between DO within a two- and three-place predicate construction, and between different degrees of heaviness of the complement. With a two-place predicate, extraposition is obligatory for tensed clauses. E.g. *Er hatte dass er nicht lange leben wrde gewusst. Er hasse gewusst, dass er nicht lange leben wrde. But optional for infinitival clauses with zu E.g. Er hatte die Frau zu gewinnen gehofft. Er hatte gehofft, die Frau zu gewinnen.
With a three-place predicate a DO complement must again be extraposed if it is a tensed clause E.g. *Ich werde meinem Sohn dass er morgen den spteren Zug nehmen soll empfehlen. Ich werde meinem Sohn empfehlen, dass er morgen den spteren Zug nehmen soll. but also if it is zu + infinitive clause containing surface arguments of the infinitive E.g. *Ich werde meinem Sohn morgen den spteren Zug zu nehmen empfehlen. Ich werde meinem Sohn empfehlen, morgen den spteren Zug zu nehmen.
If the complement is an oblique constituent (OBL) within a three-place predicate construction, the same situation obtains as in the extraposition of DO complements in a three-place predicate construction. Obligatory for tensed clauses E.g. *Er hat mich dass ich seinen Bruder zum Bahnhof fahre gebeten. Er hat mich (darum) gebeten, dass ich seinen Bruder zum Bahnhof fahre. Obligatory for zu + infinitive clauses with surface argument E.g. *Er hat mich seinen Bruder zum Bahnhof zu fahren gebeten.
51
Er hat mich (darum( gebeten, seinen Bruder zum Bahnhof zu fahren. Optional for zu + infinitive alone E.g. Er hat mich zu arbeiten gebeten. Er hat mich (darum) gebeten, zu arbeiten.
52
Conclusion: this distribution of obligatory and optional applications is completely in accordance with the hierarchies mentioned before. 3-place predicates: more obligatory applications than 2-place predicate constructions 2-place predicates: more obligatory applications than 1-place predicate constructions
PPs are also found to the right of the participle constituents of complex verbs. E.g. Man setzte die Urlauber in einem Boot ber.
53
Man setzte die Urlauber ber in einem Boot. Men zette de vakantiegangers in een boot over. Men zette de vakantiegangers over in een boot. Dutch is quite similar to German.
1. Extraposition from NP
This rule, like its English counterpart, moves the S of a complex NP into rightmost position within its clause and hence to the right of finite and non-finite verb forms in German. E.g. Der Bauer hat die Kuh, die schlechte Milch gab, geslachtet. Self-embedded in the higher relative clause. grammatical in contrast to self-embeddings in English. Der Bauer hat die Kuh geschlachtet die schlachte Milch gab. *The farmer had the cow which bad milk gave, slaughtered. The farmer has slaughtered the cow which gave bad milk. De boer heeft de koe die slechte melk gaf, geslacht. De boer heeft de koe geslacht die slechte melk gaf. English has got only 1 choice; Dutch and German have multiple choices. Perfectly
I recommended to the teacher who was so concerned that Mary see a doctor. So: First Dutch is like English: S V OE but not quite English: S Vf Vnf O Dutch: S Vf O Vnf OE German: S Vf O Vnf OE Now: English is
S Vf Vnf O X
OR
S Vf Vnf X OE
1. Contrastive summary
We have seen than English and German can have very similar clause-final constituents, both as a result of sharing similar rightward-moving rules like Extraposition from NP, and as a result of English base-generating clause-final constituents which would be derived by rightward movement in German. Yet, because of the basic final position of the verb in German, German rightwardmoving rules can apply non-vacuously in numerous environments which are without parallel in English, creating (when application is optional) pre-verbal and post-verbal doublets for which there is only one translation equivalent in English. If the extraposed constituent originates in subject position, both languages will share clearly unextraposed and extraposed structures. But when the extraposed constituent originates in non-subject position, i.e. to the right of the verb in English, English will, in the absence of other constituents to the right in the input structure, show no sight of Extraposition ever having taken place. Hence, corresponding German doublets, English has only one translation equivalent.
2 causal factors VO constructions/ languages: early V facilitates integration of arguments. OV constructions/languages: arguments determine meaning of verb If extraposition, then non-rigid OV
55
56
1. Verb-second in German
It does in general seems to be the case that what is topicalised to the front of the sentence must be a single constituent. And in Bierwischs grammar, the finite verb is then moved from its underlying final position into second position, behind the initial constituent, by a rule which we shall refer to as Verb-Second. PPs and Advs can be topicalised, both those which are non-strictly subcategorized with respect to the main verb and those which are strictly subcategorised. E.g. Mglicherweise hat er uns vergessen. Mogelijk is hij ons vergeten. In Mnchen haben wir viel Bier getrunken. In Mnchen hebben we veel bier gedronken. NPs of all cases can be topicalised, including nominative-marked subjects. E.g. Den Hund sieht die Katze De hond ziet de kat De kat ziet de hond. Not ambiguous in German, but ambiguous in Dutch and English. An embedded VP can be topicalised E.g. Das Auto zu reparieren, hat er uns versucht. *De auto te repareren, heeft hij ons gevraagd. Not possible in Dutch This same process of VP-topicalisation will also front infinitives and participles alone. E.g. Gewinnen mssen wir. Winnen moeten we contrastive stress! More examples in book a. Conclusion Dutch is a little bit less verb-second than German
57
1. Verb-second in English
The Modern English verb-second structures are clear historical relics of an earlier more productive verb-second pattern, similar to that of German. There are in fact several verb-second rules in the modern language. Each incorporates numerous idiosyncrasies and applies in highly specialized environments. The result of this declining productivity in English, coupled with the invariable regularity of verb-second order in German main clauses, is that the set of structures produced by the verb-second rules of English appears to be exactly a proper subset of the verb-second structures of German. For every verb-second structure in English, there is a corresponding verb-second structure in German, but not vice versa.
a. Subtype 1: subject-auxiliary inversion Three environments: WH + Vf + S NEG + Vf + S So + Vf + S However, so preceding an adverbial or adjective is less productive than the other cases.
a. Subtype 2: Subject- simple verb inversion Two environments Directional adverb + Vf + S Quote + Vf + S
a. Subtype 3: Inversion after Locative Phrase Preposing Locative + Vf + S But: actually the whole VP is inversed, so: LOCATIVE + VP + S Inversions apply on root sentences only in English and are triggered by a single preposed constituent without any comma intonation break. Postverbal subject pronouns are ungrammatical. As with subject-simple verb inversion, inversion after locative phrase preposing is impossible if a DO is present.
a. Subtype 4: Comparative Substitution This rule inverts comparative predicate adjective whose heads are compared by means of more, less, least and as, with the subject NP.
58
a. Subtype 5: Participle Preposing E.g. Our top reporter is speaking to the President now. Speaking to the President now is our top reporter. When we say that all English verb-second structures have corresponding verb-second structures in German, but not vice versa, we therefore mean: when there are identical verb forms in the two languages, any postposing of the subject and preposing of an initial constituent that is possible in English will be possible in German, but not vice versa; but where the verbal forms of the two languages differ, no claim is made.
1. Contrastive summary
The rules generating the structures of Modern English are clearly the fragmented remains of the verb-second rule of early Germanic., which Modern German now applies so exceptionlessly in root clauses. But also an innovation! V2 become VP2 in some cases Locative phrase VP inversion Comparative VP inversion Makes sense: English wants the two parts of the verb together.
1. Verb-third in English
The verb-second structures of German that are without verb-second counterparts in English are generally matched by verb-third structures, a structure with no counterpart in German. Sentences with preposed sentence adverbs are verb-third in English and verbsecond in German. However, there are genuine differences between English and German with respect to the preposing possibilities for locative adverbs. The literal translation of the German is ungrammatical in English both as a verb-second and as a verbthird structure. E.g. In Mnchen wohnt der Mann. *In Munich lives the man * In Munich the man lives. English has verb-third structures for frequency adverbs E.g. Zweimal in der Woche ist sie gekommen. Twice a week she came.
59
Manner adverbs are restricted in their preposing possibilities. E.g. Schn singt die Sngerin. *Beautifully singst the singer. *Beautifully the singer sings.
60
Four cases where topicalised constituents in German have either no counterpart or a less productive counterpart in English Fronted locative adverb Fronted manner adverb Fronted so + Adj/Adv Fronted VP
Why should English have expanded verb-third at the expense of verb-second? The reduction in English verb-second structures can in no way be an automatic or necessary consequence of the innovation of base SVO., in the same way that the eliminations of verb-final structures is thereby explained. reduction of V2 does not automatically follow from VO fixation A language can be VO and have serious V2 E.g. Older English, and PD Scandinavian Swedish Jag har lst boken. I have read the book Jag tror att jag har lst boken. I believe that Igr har jag lst boken. Yesterday AND YET! Reduction of V2 does follow (but not automatically) from SVO fixation SVO is the most frequent realization ov XV (V2) topicalization XSVO strengthens SVO English V2 is typically X Aux SVO X in English XV is typically not the object
As a result, the sentence types of Modern English which cannot have the verb in second position cannot be used for the overt expression of the relevant functions; and, assuming that these functions were also characteristic of earlier stages of the language, the progressive elimination of verb-second structures means the elimination of the explicit expression of these pragmatic functions and their collapsing into the common SVO mould.
61
1. Verb-first in German
Polar questions What about Dutch? Same Imperatives What about Dutch? Same Exclamations What about Dutch? Same Counterfactuals and conditionals What about Dutch? Same Stylistic inversion What about Dutch? Same Dutch seems to be like German
1. Contrastive summary
Historically, English verb-first movement has therefore been on the decline. The environments that trigger it, have been less radically reduced than have those for verb-second structures. But whereas the verb-second structures retain some subject inversions with main verbs, verb-first movements are defined exclusively on auxiliaries or modals.
62
Did we ever laugh! Aux S V Should he finish it first, we will Aux S V O It is striking that most of the reductions in verb-first structures involve the classes of verbs moved and that the environments for verb-first movement are still very similar in the two languages. This contrasts with verb-second rules, whose triggering environments en English are considerably reduced compared with their German counterparts. The most productive English verb-first environment is that of yes-no questions. All auxiliaries and modals can front, and because of do-support, English can always produce a verb-first question corresponding to the fronted non-auxiliary verbs of German.
True, but Aux SVO Imperative verb-first movements: <understood S> VO Exclamations and counterfactuals/conditional clauses: productivity between yes-no questions and imperatives. intermediate in
The consequence of all these reductions in productivity is that canonical SVO word order takes over from verb-first. Yes-no questions are still productively verb-first, but all the ungrammatical inversions in the other environments are now rendered as S (Aux) V (O) structures.
63