Download Topicalization and Stress Clash Avoidance in the History of English 1st Edition Augustin Speyer ebook All Chapters PDF

Download the full version of the ebook at ebookfinal.
com
Topicalization and Stress Clash Avoidance in the

History of English 1st Edition Augustin Speyer
https://ebookfinal.com/download/topicalization-and-stress-
clash-avoidance-in-the-history-of-english-1st-edition-
augustin-speyer-2/
OR CLICK BUTTON
DOWNLOAD EBOOK
Download more ebook instantly today at https://ebookfinal.com

Instant digital products (PDF, ePub, MOBI) available
Download now and explore formats that suit you...
Topicalization and Stress Clash Avoidance in the History

of English 1st Edition Augustin Speyer
https://ebookfinal.com/download/topicalization-and-stress-clash-
avoidance-in-the-history-of-english-1st-edition-augustin-speyer-2/
ebookfinal.com
The Cambridge History of English Romantic Literature The

New Cambridge History of English Literature 1st Edition
James Chandler
https://ebookfinal.com/download/the-cambridge-history-of-english-
romantic-literature-the-new-cambridge-history-of-english-
literature-1st-edition-james-chandler/
ebookfinal.com
Studies in the History of the English Language V Variation

and Change in English Grammar and Lexicon 1st Edition
Robert A. Cloutier (Ed.)
https://ebookfinal.com/download/studies-in-the-history-of-the-english-
language-v-variation-and-change-in-english-grammar-and-lexicon-1st-
edition-robert-a-cloutier-ed/
ebookfinal.com
Studies in the History of the English Language V Variation

and Change in English Grammar and Lexicon Contemporary
Approaches 1st Edition Robert A. Cloutier (Ed.)
https://ebookfinal.com/download/studies-in-the-history-of-the-english-
language-v-variation-and-change-in-english-grammar-and-lexicon-
contemporary-approaches-1st-edition-robert-a-cloutier-ed/
ebookfinal.com
The Routledge History of Literature in English Britain and
Ireland 2nd Edition Ronald Carter
https://ebookfinal.com/download/the-routledge-history-of-literature-
in-english-britain-and-ireland-2nd-edition-ronald-carter/
ebookfinal.com
Yellowstone National Park 1st Edition Byron Augustin
https://ebookfinal.com/download/yellowstone-national-park-1st-edition-
byron-augustin/
ebookfinal.com
Complications in Cardiothoracic Surgery Avoidance and

Treatment 1st Edition Alex G. Little
https://ebookfinal.com/download/complications-in-cardiothoracic-
surgery-avoidance-and-treatment-1st-edition-alex-g-little/
ebookfinal.com
Ethnocentrism and the English Dictionary Routledge Studies

in the History of Linguistics 3 1st Edition Phil Benson
https://ebookfinal.com/download/ethnocentrism-and-the-english-
dictionary-routledge-studies-in-the-history-of-linguistics-3-1st-
edition-phil-benson/
ebookfinal.com
The Handbook of the History of English Ans Van Kemenade
https://ebookfinal.com/download/the-handbook-of-the-history-of-
english-ans-van-kemenade/
ebookfinal.com
Topicalization and Stress Clash Avoidance in the History of English
Topics in English Linguistics
69
Editors
Elizabeth Closs Traugott
Bernd Kortmann
De Gruyter Mouton
Topicalization
and Stress Clash Avoidance
in the History of English
by
Augustin Speyer
De Gruyter Mouton
ISBN 978-3-11-022023-0
e-ISBN 978-3-11-022024-7
ISSN 1434-3452
Library of Congress Cataloging-in-Publication Data
Speyer, Augustin.
Topicalization and stress clash avoidance in the history of Eng-
lish / by Augustin Speyer.
p. cm. ⫺ (Topics in English linguistics; 69)
Includes bibliographical references and index.
ISBN 978-3-11-022023-0 (alk. paper)
1. English language ⫺ Grammar, Historical. 2. English lan-
guage ⫺ Syntax. 3. English language ⫺ Word order. 4. English
language ⫺ History. I. Title.
PE1075.S64 2010
4201.9⫺dc22
2010002363
Bibliographic information published by the Deutsche Nationalbibliothek

The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie;
detailed bibliographic data are available in the Internet at http://dnb.d-nb.de.
” 2010 Walter de Gruyter GmbH & Co. KG, 10785 Berlin/New York
Cover image: Brian Stablyk/Photographer’s Choice RF/Getty Images
Printing: Hubert & Co. GmbH & Co. KG, Göttingen
⬁ Printed on acid-free paper
Printed in Germany
www.degruyter.com
Preface
This book is a modified and slightly extended version of my PhD thesis

which I submitted at the University of Pennsylvania in 2008. The basic
ideas however even go back to my time as Visiting Scholar at the Linguis-
tics Department of the University of Pennsylvania in Fall Term 2002. An-
thony Kroch taught a seminar on historical syntax at that time in which I
discovered the basic impetus for this work – the decline of topicalization.
The other participants were Silvia Cavalcante, Daniel Ezra Johnson, Bea-
trice Santorini and Laura Whitton. Their comments helped me immensely
in the early steps of this work.
The biggest thanks go, of course, to the supervisor of the thesis, An-
thony Kroch, for his commitment in supervising this thesis. Many of the
ideas expressed in this study go originally back to him. But the contribution
of the other committee members, Eugene Buckley, Rolf Noyer, Donald
Ringe and Jiahong Yuan cannot be valued too highly either, and I wish to
thank them for their commitment. I am also grateful that I could discuss
parts of the work at various stages with Werner Abraham, Brian
McHughes, Ellen Prince, Marga Reis, Arnim von Stechow, Hubert Truck-
enbrodt and William Barry. Their comments were extremely helpful. All
remaining errors are my own, of course.
As the project went on, it became necessary to conduct experiments in
countries with a strong supply of German and English native speakers. The
preliminary German experiment was conducted in the Phonetics Lab of the
Universität des Saarlandes at Saarbrücken in summer 2004. My thanks go
to William Barry who made it possible for me to use the equipment and
who also lent a willing ear to discussions, further Uta Panten and Dominik
Bauer, who assisted me in doing the recordings. The preliminary English
experiment was conducted in the Phonetics Lab of UPenn in spring 2005;
many thanks to Maciej Baranowsky for technical support. The main ex-
periments, whose results are recorded in this thesis, have been made
throughout the year of 2006 and in the beginning of 2007 at the Phonetics
Labs of UPenn, of Tübingen University and ‘on the street’. Many thanks go
to Somdev Kar, Marjorie Pak and Jonathan D. Wright for their technical
support. In this context I also wish to thank my mother, Dietlinde Speyer,
among other things because she supplied me with a large pool of linguisti-
cally naïve German native speakers among her colleagues.
vi Preface
Preliminary versions of parts of this study were read at various confer-

ences, PLC 27 (February 2003, Philadelphia), NWAV 32 (October 2003,
Philadelphia), International Conference on Linguistic Evidence (January
2004, Tübingen), TaCoS 2005 (June 2005, Stuttgart) and Interspeech 2006
(September 2006, Pittsburgh). I want to tank the audiences of these confer-
ences for innumerable useful hints and comments in the respective discus-
sion periods.
Also I wish to thank my fellow-knights of the ‘Dr.-Cardona-Happy-
Hour’, Jonathan Gress-Wright, Neville Ryant, Joel Wallenberg, and espe-
cially Jean-Francois Mondon for their friendship and for proofreading a
draft of this opus and correcting my English, if it proved to be too baroque
to be of any practical use to the reader.
The final thank goes to the editor of TiEL, Elizabeth Traugott, who
made a tremendous impact by her comments throughout the time in which I
was preparing the manuscript for publication. It was her suggestions about
material that goes beyond the original thesis that allowed the book to be a
comprehensive study.
Frankfurt / Main, January 2010 Augustin Speyer

Contents
Preface v
1. Introduction 1
1.1 Overview 1
1.2 Some background 3
1.2.1 Pragmatic dimensions 3
1.2.2 Modularity of Grammar 8
1.2.3 Prominence 9
1.2.4 Grid construction 12
1.2.5 The syntactic field model 17
1.3 Further concepts 19
1.3.1 Verb second 19
1.3.2 The reconstruction of sentence prosody 21
2. Topicalization in Middle and Modern English – 24

A prosodically induced change in syntactic usage
2.1 The decline of topicalization 24
2.2 The pragmatic properties of topicalization 29
2.2.1 The discourse-pragmatic functions of topicalization in 30
Modern English
2.2.2 The discourse-pragmatic functions of topicalization in Old 37
and Middle English
2.3 A possible explanation: rigidity of word order 42
2.3.1 The rigidification of English word order 43
2.3.2 Rigidification as an explanation for the decline of 48
topicalization?
2.4 A third explanation: the Clash Avoidance Requirement 50
2.4.1 Type of subject 51
2.4.2 Scene-setting elements 56
2.4.3 Formulation of the Clash Avoidance Requirement 61
2.4.4 The loss of the V2 word order option 62
2.5 Prosody beats Pragmatics 73
2.6 Summary 77
viii Contents
3. The Clash Avoidance Requirement in Modern English 80

and German
3.1 Double foci 80
3.1.1 The acceptability of topicalization in Modern English 81
3.1.2 Experimental data on double foci 87
3.2 Conceptual aspects of the Clash Avoidance Requirement 113
3.2.1 The Clash Avoidance Requirement, the Rhythm Rule and the 113
OCP
3.2.2 The domain of the Clash Avoidance Requirement 117
3.2.3 Modularity and the Clash Avoidance Requirement 119
3.3 The Clash Avoidance Requirement in German 121
3.4 Summary 131
4. Phonological Aspects of the Clash Avoidance Require- 132

ment
4.1 The relationship of metrical prominence and focus 132
4.1.1 Comparison of metrical prominence and focal emphasis 133
4.1.2 The metrical calculus and the primacy of the focus indicator 139
4.1.3 The nucleus as a continuation of the metrical prominence 150
system
4.1.4 Metrical prominence, focus and the intonational system 162
4.2 Clash and pause 164
4.2.1 Mechanisms for clash resolution 165
4.2.2 Why a pause? 168
4.2.3 Morphemes as domain of focus 173
4.3 Summary 176
5. Topicalization and the Clash Avoidance Requirement in 177

Old English
5.1 V2 and V3 in Old English 178
5.2 Is V3 really V3? 187
5.2.1 Verb-last sentences 188
5.2.2 Verbal particles 198
5.2.3 Modelling 201
5.2.4 The consequence of Infl-medial V3 209
5.3 V3 in (Old High) German 210
5.4 The nature of the two subject positions 217
5.5 A short outlook on poetry 227
5.6 Summary 232
Contents ix
6. Concluding remarks 234
Appendix: All Old English OSV-sentences with full noun 237

phrase accusative object and subject
Notes 255
References 266
Primary sources 266
Research literature 267
Index of names 283
Index of subjects 285
2
Chapter 1
Introduction
1.1. Overview
The main concern of this study is to demonstrate how a general phonologi-

cal, or more specifically, a prosodic requirement – the Clash Avoidance
Requirement (= CAR) – can influence the syntactic usage of a given lan-
guage, English. So it is, on a more abstract level, about the interaction of
seemingly disparate aspects of the language, namely phonology and syntax.
The way they interact is highly dependent on principles of information
structuring, the effects of the interaction are observable over a given time
span, further insights come from a comparison with German, the close rela-
tive. It is consequently fair to say that this study touches on four linguistic
disciplines, Syntax, Phonology, Pragmatics and Historical Linguistics.
Topicalization is an exemplary case for demonstrating this interaction
and the power of the Clash Avoidance Requirement, and therefore much of
this text will be devoted to a discussion of topicalization in the history of
English. In the second part of the study we will see that the Clash Avoid-
ance Requirement is responsible for a gradual decrease in the rate of topi-
calization in Middle and Early Modern English to a stable, yet low, fre-
quency. This decrease in topicalization is observable only in cases in which
the loss of the verb second word order option (= V2), which happened in
the same time span, leads to potential violations of the Clash Avoidance
Requirement. They can occur when two full noun phrases come to stand
adjacent to each other, because then both noun phrases have a certain like-
lihood of bearing focal emphasis. In this case, that is, when there are two
phrases with focal emphasis in a sentence, the Clash Avoidance Require-
ment requires that they must be separated by at least one element of minor
prominence. In this study the decline of topicalization will be attributed to
the danger of CAR-violations in the wake of the loss of the V2 word order
option. Alternative explanations, such as the idea that the decline in topical-
ization has to do with the growing rigidity of word order in English, or that
the decline in topicalization is due to the gradual loss of pragmatic contexts
in which topicalization was used, will be argued against.
The study begins with some definitions and an overview over concepts
mentioned throughout the study in chapter 1. After having shown in the
2 Introduction
second chapter how the Clash Avoidance Requirement influenced syntactic

usage in earlier periods of English, the third part of this investigation will
be devoted to the Clash Avoidance Requirement in present day English and
German and its technical description. I will present experimental data
which shows that speakers of English and German prefer to avoid uttering
two foci adjacent to each other, but if they are forced to do so, they rescue
the Clash Avoidance Requirement by inserting a pause.
In a fourth more theoretically oriented part, I will discuss the reasons
why speakers typically choose pause insertion and not other clash resolu-
tion mechanisms in situations of focus clash. The properties of rule-
governed metrical prominence and semantic focal prominence are so dif-
ferent on a descriptive level that focus cannot simply be reduced to being a
continuation of the metrical prominence system. Moreover, different rules
are used to generate them which interact in a typical way, but remain quite
distinct. A focus indicator is only assigned if there is a narrow focus on a
word; otherwise, rule-governed metrical prominence takes care of the as-
signment of prominence up to the topmost level. The Clash Avoidance
Requirement holds on this topmost level, the clause level, both in the pres-
ence and absence of focus, and can be easily formalized in the framework
of Metrical Stress Theory, following Hayes (1995), as a ban on non-
branching feet.
In the fifth part, I will turn to Old English and show that here also the
Clash Avoidance Requirement plays a central role in the interaction be-
tween syntactic usage and phonology. This is especially obvious in a hall-
mark problem of English syntax, the alternation of surface V2 and V3 word
order. This alternation will be shown to be governed by the CAR: As we
can observe, the alternation appears in such a way that the element with the
least likelihood of bearing focus always immediately follows the topical-
ized phrase, either the subject if it is topical (most often realized as pro-
noun), or the verb if the clause has a full noun phrase, non-topical subject.
The former case yields V3-sentences, the latter V2-sentences. This pattern
corroborates the view (cf. Haeberli 2002) that Old English syntax was not a
strict V2-syntax in the fashion of Modern German, but that Old English had
two subject positions for subjects of a different (information-structural)
shape, and thus resembled much more Modern English syntax than the
classic West-Germanic (= German) type.
Some background 3
1.2. Some background
As it should be useful to give some preliminary definitions of notions and

ideas that this study makes use of, let me briefly introduce some relevant
concepts. I will devote two sections to this end. The first section (1.2)
touches on the theoretical frameworks to be applied. In this section the
pragmatic dimensions are introduced that will be discussed, the model of
grammar and the metrical theory which I assume, and the German field-
model, whose terms we will encounter frequently. The second section (1.3)
discusses more specific concepts, viz. what we mean when we say “verb
second”, and how it is possible at all to determine prosodic properties in
written texts, even written texts of a bygone stage of the language.
1.2.1. Pragmatic dimensions
One does not need to be a functionalist to recognize that in a number of

languages one of the most important factors determining surface word or-
der is discourse and information structure. Latin is certainly among those
languages, but so is German, and, to some extent, even a language like
English (Mathesius [1928] 1964).
But information structure is not a unitary notion that always influences
word orders in the same way. The term information structure is rather a
cover term for several ways in which information can be ordered. In the
1960s and 1970s, in the wake of the teachings of the so-called Prague
school (e.g. Firbas 1974), it was assumed that there is only one information
structural dimension – a ‘communicative dynamism’, which subsumed
theme-rheme, background-focus, given-new, frame-proposition. But at
present many researchers assume that there are indeed several information
structural ordering principles (cf. to a similar multi-layered conception of
information structure Féry and Krifka [2008]). Let us call these principles
‘pragmatic dimensions’. It is important to note here that these dimensions
are not reducible to one another (as the length of a physical object cannot
be traced back to its depth, for instance), but exist independently and try to
order the information in their own way, in consequence sometimes coming
into conflict with other dimensions, of course.
Four dimensions are relevant here. I do not wish to imply that there are
not more dimensions, but these have been selected, partly because they
proved to be of importance, partly because they influence the prosody of a
clause directly. They are the following:
4 Introduction
– newness: old versus new information,

– topicality: topic versus comment (roughly = theme versus rheme),
– focus: focus versus background,
– scene-setting: scene-setting versus proposition-internal.
In the following the definitions are given for each dimension. The defini-
tions depend basically on Féry and Krifka (2008).
Newness is a rather self-explanatory concept, although one has to ask,
what the scope of ‘new’ or ‘old’ is – new/old for the hearer, new/old for the
speaker or new/old in the discourse. In this study I use the old-new-
distinction exclusively in the sense related to the discourse: Information
that has been previously mentioned in the discourse counts as old (or given,
or, as Prince (1981a) calls it, evoked), whereas information that is men-
tioned for the first time counts as new. Examples for discourse-givenness
and newness are given under (1). There are practical reasons for that
choice, in that in dealing with written texts we may on the one hand assume
that the writer only uses entities which are old to him, on the other hand we
can trace only newness or evokedness within the discourse – we have no
idea what would be old or new for the typical recipient of such literature in
the time in which it was composed.
(1) Rudolf Bupfinger, inspector of the state’s criminal investigation unit,

was sitting in his office. All of a sudden the door was flung open and
a young man stumbled into the room. He held a hatchet in his hand.
What the inspector found even more remarkable was the knife which
was protruding from the back of his visitor, who fell down, pale-
faced.
– Discourse-new information: underlined
– Discourse-old information: in italics.
There are several intermediate stages to the old-new-distinction, either to be

conceived of as different points on a scale, as in Gundel, Hedberg, and
Zacharski (1993),1 or as different entities altogether, as in Prince (1981a).
One of the intermediate stages is the status that Prince (1981a) calls ‘infer-
able’, which means that a given entity has not been mentioned in itself be-
fore, but other entities which are typically associated with this entity are
present in the discourse universe, so that the hearer can infer it via logical
or plausible reasoning. An example is given under (2). Here we know from
world knowledge that rescue squads typically contain at least one para-
Some background 5
medic, so the mention of a paramedic is in some ways premediated by the

mention of rescue squad. Inferable information normally patterns with old
information.
(2) The first thing the inspector did was calling the rescue squad. On
arriving, the paramedic felt for the pulse.
Old and new information are often encoded differently; old information
tends to be realized by pronouns (if felicituous reference is guaranteed or at
least likely), whereas new information is realized by phrases containing
‘real’ lexical material. Example (1) follows this pattern to some extent; it is
obvious that the referent of he must be a person that is salient in the dis-
course. The fact that old information patterns with pronouns in general will
prove to be relevant in the further course of this study.
Let us turn to topicality. What counts as a topic has been a matter of
debate, partly because there is a great deal of terminological insecurity
connected with this concept. Some studies define ‘topic’ as the element
which is at the leftmost position of the sentence (hence the term ‘topicaliza-
tion’ for movement of elements to the left periphery).2 This is not the sense
in which the term ‘topic’ is used here. Other studies (e.g. Chafe 1976)
equate topic with old information. As I have introduced old information as
an independent notion, I obviously do not follow this usage either. In this
study, topic is understood in a non-structural, pragmatic sense as the entity
that the sentence is ‘about’ (following Reinhart’s [1981] definition, which
is the standard definition of theme in the Prague school tradition and which
in the end goes back to Paul [1875: 125]); the rest of the sentence adds
information to this particular entity. An example is offered in (3), in which
all sentences except the first add information to the ominous young man,
who is referred to by a pronoun, as is typical for topics.
(3) Bupfinger looked sadly at the young man. Obviously he had been in
a hurry to come here, but before he reached his victim, someone
thrusted the knife into his body. He was clad in a blue jeans and a T-
shirt, very unobtrusive.
To determine what the ‘topic’ of the sentence is, therefore, requires a cer-
tain amount of intuition, which most people however possess. An attempt
to cast these intuitions into a more formal framework was made by Center-
ing Theory (Grosz, Joshi, and Weinstein 1995; Walker, Joshi, and Prince
1998), which makes crucial use of the fact that topics are usually old in-
6 Introduction
formation, and that topics tend to be realized by predictable syntactic

means. In English, for instance, topics tend to be realized as pronouns and
frequently function as the subject of the sentence. This latter property
probably is true for all Indo-European languages (cf. Lehmann 1976).
Focus is strictly speaking not a purely information structural term, but
rather a semantic term, because we can identify a semantic operation that is
associated with the presence of focus (Rooth 1985). We can distinguish
several kinds of focus, e.g. presentational focus (4a), contrastive focus (4b;
Rochemont 1986), verum-focus (4c, see e.g. Höhle 1992), and probably
more.
(4) a. One thing Bupfinger found strange: The leather boots which the
young man was wearing.
b. Normally men of his age preferred sneakers. Such leather boots
Bupfinger only knew from Jane-Austen-movies.
c. But this guy WAS wearing them, that was the weird thing.
For English, focus is associated with prominence on the focalized element,

and this prominence is the highest one in the sentence (see Jackendoff
1972). This means that focus is, in contrast to e.g. old/new information or
topics, explicitly marked in the linguistic output. We assume, following
Jackendoff (1972) and subsequent literature, that focus is realized by an
abstract [+ focus]-feature that is associated at PF with an extra layer of
prominence (more detailed see section 4.1.2). Other languages use other
strategies to mark focus, e.g. focus particles (e.g. Japanese), pre-specified
focus positions (e.g. Hungarian), or a combination of prominence and parti-
cle (e.g. German). A presentational focus falls on an element that is new to
the discourse and whose newness should be emphasized at the same time.
Contrastive focus falls on elements that stand in a partially ordered set
(henceforth poset for short) relation to each other as members of a set that
is either evoked previously in the discourse or is evoked by the first men-
tioning of one of its members. Verum-focus is a very specialized type of
focus; it lies on the verb and emphasizes the claim that the proposition is
true. All these different kinds of focus can, in the end, be reduced to con-
trastive focus, as Rooth (1985) showed: in all cases of focus a set, consist-
ing of salient entities, is evoked of which the focused element is a member.
The meaning of focus can be summarized as ‘it is X, and not other mem-
bers of the salient set containing X, although they would have been equally
eligible’. I want to mention here a point that I elaborate on later, viz. that I
reserve the notion of focus to cases in which a salient set is clearly identifi-
Some background 7
able, which as a rule coincides with what is known as instances of ‘narrow

focus’. ‘Wide focus’, where the set would consist on possible propositions
or object-verb pairs does not fall under this strict definition of focus. There-
fore not all English sentences do have a focus, under this view; the highest
prominence in a sentence is not automatically associated with focus.
Scene-setting, finally, is an information-structural dimension, but with a
semantic side to it. As opposed to, say, concepts such as topic-comment or
newness, scene-setting elements have direct implications for the truth value
of a sentence (whereas, e.g., it is irrelevant for truth conditional purposes
whether a given expression is thematic or rhematic, for instance). We can
define scene-setting elements as elements that specify the situation under
which the truth value of the proposition has to be evaluated (definition fol-
lowing Jacobs [2001]). They do not belong to the core proposition. Exam-
ples can be found in (5).
(5) In the year 2008, wearing such shoes was most remarkable. All the
more since it was a hot summer day. Only the day before a heavy
thunderstorm struck the town with unwont violence.
Although these four pragmatic dimensions are independent of each other,

there are certain typical intersections (see also Speyer 2008a). Topics are,
as a rule, also old information.3 Not all old information functions as a topic,
however. New information is often focused, but it need not be. Foci can be
new information or old information (this is often the case with contrastive
foci). A phrase can be topic and focus at the same time under certain cir-
cumstances. We will encounter the intersection between ‘topic’ and ‘focus’
in section 2.2 of this study. Scene-setting elements tend to be old informa-
tion. New scenes can be introduced, though, and in that case these expres-
sions usually receive focus.
The dimensions are often in conflict with each other. This is because
each dimension poses certain requirements on the linguistic output and
speakers tend to follow these requirements: Old information is likely to be
placed before new information, topics are put before their comment, foci
are preferably realized at one of the edges of the utterance, and scene-
setting elements are usually positioned before the proposition. All of these
ordering requirements make sense independently from the point of view of
sentence processing: it eases processing if old and new information are not
jumbled together but are ordered somehow (Musan 2002). Also, it is more
sensible to first evoke the ‘filecard’ (= topic) and only afterwards the mate-
rial that has to be added to this filecard (= the comment), if we want to use
8 Introduction
Heim’s (1982) famous metaphor.4 It is better, if one wants to emphasize

something, to put it in a position where it coincides with one of the clausal
edges and therefore can be treated as separate processing unit. And if a
situation’s truth value is to be evaluated, it is more practical to know the
situation before hearing the material that is to be evaluated. So each dimen-
sion has a certain ‘claim’ on sentence structure and order, so to speak.
Which one of these claims determines the shape of the output varies by
cases, although languages tend to have a ranking of the dimensions (see
Speyer 2008a).
1.2.2. Modularity of Grammar
I assume a modular model of grammar in the tradition of Chomsky (1995,

2001). I assume the modified T-model (or rather: Mercedes-star-model;
[6]) in which there are three components: Narrow Syntax, Logical Form
(LF) and Phonetic Form (PF). In this study we are mostly interested in PF.
In Narrow Syntax we are interested only insofar as it contributes to the PF-
representation.
Narrow Syntax is the module in which material from the lexicon – in
Minimalism referred to as ‘enumeration’, at this stage represented as ab-
stract concepts and feature bundles – is assembled and in which the first
transformations take place, such as movement of the subject to SpecIP, for
instance. At the place at which Surface Structure used to be in the Extended
Standard Model (e.g. Chomsky 1981) is now a bifurcation that does not
count as an independent level of representation. The output which narrow
syntax has produced feeds into two modules, LF and PF. LF is the module
where movement operations take place that are not represented in the form
of the sentence that is uttered (since the branch leading to the actual utter-
ance is PF, and we have left this track at the bifurcation) and that concern
mostly the correct semantic representation of the utterance, e.g. scopal
properties. PF, on the other hand, is the module in which the syntactic
structure is eventually flattened out, transformed into a linear string. Lexi-
cal Insertion takes place (see Halle and Marantz 1993) and purely phono-
logical operations are performed, such as the assignment of prosody and the
adjustment of the rhythmic structure. These are in principle not relevant for
the semantic interpretation (with the apparent exception of focal emphasis,
of course),5 but they make possible the vocal production and give cues to
the syntactic structure which, after reduction of the two-dimensional struc-
ture into a one-dimensional string, is no longer directly observable.
Some background 9
(6)
Lexicon
Narrow Syntax
LF (module) PF (module)
LF(representation) PF (representation)
In contrast to Chomsky (1995), but in accordance with many other genera-

tive grammarians (see Fanselow 1991; Haider 1997; Rizzi 1997; Haider
and Rosengren 2003: 206; Erteschik-Shir 2005), I assume that there are
also movement operations that are not governed by strictly syntactic fea-
tures, but that are discourse-structurally motivated. This implies that there
are also functional projections that can host phrases with a certain discourse
structural status, such as the ones identified by Rizzi (1997). Movement to
these projections is not warranted by Narrow Syntax, if information struc-
ture is not considered as part of the semantic representation (but cf. Asher
and Lascarides [2003] for a ‘semantic’ view of information structure). For
this reason it should be considered whether the place where such movement
operations take place is perhaps PF rather than narrow syntax (even more
radically Erteschik-Shir 2005). We could view PF procedurally as consist-
ing of several sub-modules, one in which additional, non-syntactically mo-
tivated and non-semantically interpretable movement operations take place,
one in which the structure is reduced to a string, one in which Lexical In-
sertion takes place, one in which the rhythmical structure is assigned and
one in which the well-known phonological rules of sandhi, assimilation etc.
take place. But this question is beyond the scope of this study.
1.2.3. Prominence
Prominence is used here as a cover term for the property a linguistic entity
has (usually a syllable) to be perceived as ‘stronger’ than other linguistic
entities of the same sort. I will make a distinction between the phonological
and the acoustic aspects of this concept. Acoustically, a syllable A is more
prominent than a syllable B if A has higher values than B on certain meas-
10 Introduction
urements – pitch especially, but also volume and duration. In other words:
A syllable A is more prominent than a syllable B if it is higher-pitched,
louder, and possibly takes more time to articulate, such as REE in refeREE,
or CAT in a tortoise-shell CAT. One can say that syllable A is also more
prominent, that is, higher, louder and longer, than a non-prominent instance
A’ of the same syllable. CAT in a tortoise-shell CAT is more prominent
than cat in the cat with the HAT.
Phonologically speaking, prominence can be represented by construct-
ing a metrical tree and/or building a grid in which strong and weak marks
are assigned; the more strong marks are assigned to a syllable, the more
prominent this syllable is. The grid reflects the grouping of syllables and
larger units into feet; the prominence that is assigned is dependent on the
headedness of the feet. Further below a distinction will be made between
prominence that is assigned by rules and prominence that is the outcome of
focus. I will distinguish these types of prominence terminologically in the
following way.
On the phonological level, prominence assigned by the metrical calculus
(the system that is described by rules of prosody and grid production) will
be referred to as metrical prominence (or simply prominence). The rule-
governed construction of metrical prominence can be disturbed by a focus
indicator, which is prominence (or, as I will often call it in order to distin-
guish it from metrical prominence, emphasis) associated with a focus fea-
ture.
The highest prominence assigned by the metrical calculus of a given
unit will be called its prominence peak. The highest clausal prominence
will be called the clausal prominence peak.
On the level of phonetic representation, the term stress will be used for
the acoustic correlate of metrical prominence, and the term focal emphasis
or simply focus for the acoustic correlate of the focus indicator (for the
usage of focus in this sense see e.g. Wells [2006]). By use of these terms I
do not wish to imply that one of these phonetic entities has fundamentally
different properties from the other (e.g. that stress is louder than the rest,
and focus is higher pitched than the rest, or the like); ‘stress’ in my usage
can include pitch movement, longer duration etc. The phonetic correlate of
the clausal prominence peak is called sentence stress or nucleus.
In making this distinction I follow Ladd (1996: 160), who seems to be
quite close to the consensus of the last few years. Ladd makes a distinction
between ‘normal stress’ and ‘focus-to-accent’. ‘Normal stress’ is rule-
governed and thus prominence that can be calculated. Normal stress applies
to all domains, including the clause. The highest stress of the clause is re-
Some background 11
ferred to by Ladd as sentence stress; Newman (1946: 176) calls it nucleus,

and this term has often been used to denote this concept (e.g. Chomsky and
Halle [1968] in their Nuclear Stress Rule). Ladd (1996: 293 n.2) points out
that often the term default accent is used. This usage is, however, due to
misunderstanding of the term as he himself coined it in Ladd (1980), where
it denotes a completely different concept: it is used only in words that are
deaccented to refer to the position on which the accent would fall if the
word under discussion were not deaccented.
The prominence associated with focus does not have an accepted desig-
nation; Ladd (1996: 161) refers to it as accent, focus, or emphasis. This
kind of prominence obviously has a semantic side to it, which metrical
prominence does not have. Connected with this usage is the idea that every
utterance has a focus somewhere, either a ‘wide / broad focus’, meaning
focus on the clause as a whole, the verb phrase or some other relatively
large unit, or a ‘narrow focus’, meaning focus on just a word or an even
smaller unit. The unit for semantic focus-assignment is variable; most often
it is a whole word, although the focal emphasis is of course realized only on
one syllable of this word, usually the syllable which would be the most
prominent one anyway. A consequence of this perspective is that sentence
stress always coincides with focal emphasis, as this is where the highest
prominence of the sentence is, if the sentence or the biggest part of it is in
wide focus. Ladd (1996: 161) describes the matter in this way:
‘Given the idea of broad focus, ‘normal stress’ rules can be seen as a description
of where accent is placed when focus is broad.’
If we have narrow focus, the rules for sentence stress are blocked from
applying in a regular fashion, as here the “accent goes on the focused
word” (Ladd 1996: 161).
There are other definitions of ‘stress’ and ‘accent’. Ladd’s definition
depends on Bolinger’s (1961, 1972) distinction and is more or less identical
to the distinction used by Sluijter (1995). For Bolinger, and the tradition of
phonologists before him, accent is the term used for the highest prominence
in a given unit, whereas stresses are the prominences on lower levels (the
word, the phrase). He was perhaps the first to draw attention to the fact that
it is exactly the highest prominence peak that often is not predictable by
rules, but reflects semantic and pragmatic notions such as emphasis, new-
ness, contrast, etc., what was termed focus soon thereafter (Jackendoff
1972). This development, of course, caused a certain terminological insecu-
rity, as there were now two competing meanings of the term ‘accent’:
12 Introduction
1. highest prominence in the clause, or

2. prominence associated with focus.
These meanings coincide exactly then when we assume that each sentence
has a focus, and this is the line taken by e.g. Schmerling (1976); Ladd
(1980); Selkirk (1984). Without the idea of broad focus, these meanings
coincide only then when there is a narrow focus on some word. In other
words: Only when a word is focused in a clause, this clause will have focal
emphasis. Otherwise it may have an accent in the sense of (1.) in the quote
above, but if we assume that both definitions must hold for focal emphasis,
sentences without narrow focus do not have focal emphasis at all, but sim-
ply sentence stress. This is the line I will take in later sections.
One consequence of the terminological complexities sketched here is
that there are many special uses of the terms stress and accent. Schane
(1979: 485), for instance, defines stress as the phonetic manifestation of
prominence and accent as the underlying representation of it. In other stud-
ies, accent is the term used on the production side. Wells (2006) for in-
stance uses accent only as the phonetic realisation of prominence associ-
ated with a pitch gesture, whereas the underlying prominence associated
with focus is simply called focus. Sentence stress is called nucleus, which
has the advantage that one does not have to commit oneself to the question
whether the nucleus is a kind of metrical prominence (rule-generated, no
focus) or a kind of focal emphasis (broad focus).
1.2.4. Grid construction
The theory of grid construction used in this study is based on Metrical

Stress Theory (Hayes 1995) with elements of Idsardi (1992); cf. also Halle
and Idsardi (1995). The grid is constructed in the following way: each rele-
vant element (in this study, the lowest relevant level is the word level, but
the theory works the same way below the word level) is assigned a strong
grid mark. In this study, asterisks are used for strong grid marks, and dots
for weak grid marks. The next higher line adds alternating strong and weak
marks following certain rules. This process is equivalent to the bracketing
in Idsardi (1992). The lines are not simply a continuum, but (at least) three
disctinct levels can be identified which serve as the domains for promi-
nence assignment and for metrical rules. These levels are the word, the
phrase, and the clause level (corresponding in conception, but not necessar-
Some background 13
ily in detail, to the levels of word, phonological phrase and intonational

phrase of the Prosodic Hierarchy, cf. Truckenbrodt 2007: 436). In this in-
troduction I use a simplified version with a continuous grid, for ease of
explanation. The rules for the assignment of strong and weak marks are
parametrized, that is, different in a limited way for different languages. In
English and German, the two languages that are the focus of this study, the
rules are different for the domains below the word and the domains higher
than the words. The basic rule for grid construction on levels higher than
the word in English is as follows, cast in terms of Metrical Stress Theory:
Iamb Construction Rule:

Assign iambs from right to left.
It is easy to see that this is an iterative version of the Nuclear Stress Rule,
as we know it from e.g. Newman (1946: 176) and Chomsky and Halle
(1968: 90). It means that the assignment process starts at the rightmost
word of the clause, assigning a strong mark to it, assigning a weak mark to
the penultimate word, assigning a strong mark to the third-last word and so
on, until the clause has been scanned completely. The next higher line uses
the same assignment rule, and puts alternating strong and weak marks on
the grid. It is not simply a copy of the line below, as the only positions that
are available for assignment are the ones with strong marks on the lower
line. The assignment process for this level goes on, until the clause has
been parsed completely. In this fashion, line after line is added until further
assignment would be vacuous, i.e. until a line is reached where only one
iamb can be assigned. We will say that the parse is exhausted on this level.
The relative prominence of the elements in the clause is a result of the rela-
tive number of strong grid marks each element has received. Schematically,
the assignment process is shown in (7).
(7) . *
* . *
* . * . *
*. *. *. *. *
*********
There are two factors that can interfere with this strict assignment. One is
phrasing, the other eurhythmy. By ‘phrasing’ I mean the fact that not only
the word and the clause are relevant domains for prominence assignment,
but also the phrase. We thus need an intermediate level of representation.
14 Introduction
Each phrase must contain at least one strong mark (Truckenbrodt 2006),
with the sole exception of functional elements such as pronouns. Besides
the word, the (phonological) phrase and the clause (= intonational phrase),
probably no other members of the Prosodic Hierarchy (cf. Nespor and Vo-
gel 1986) are relevant for prominence assignment (cf. also Truckenbrodt
2006). And the ‘phrase’ I am talking about here is not necessarily the Pho-
nological Phrase of Selkirk (1984) and Nespor & Vogel (1986), but rather a
phrase that is roughly identical to a syntactic constituent: either an immedi-
ate constituent, that is, a syntactic phrase immediately dominated by VP (in
its base-generated position, i.e. before movement of material to functional
projections such as IP and CP), or the head of a VP, also in its base-
generated position. Precedents for such a ‘direct correspondence approach’
are e.g. Cinque (1993) and Seidl (2001).
As pointed out above, I assume that there are three relevant levels for
asignment of prominence: The word evel (ω), the phrase level (P) and the
clause level (C). Each level consists of one or more lines. On each level, a
different set of rules for grid construction applies. First, the grids for single
words are constructed, by the general rules for grid construction as given in
e.g. Hayes (1984: 35), following Liberman and Prince (1977: 315–316,
322), and by the relevant rules for the word level. The peak mark of each
word is projected on the next higher level, the starting point for phrase grid
production. The relevant rules add lines to the grids of individual phrases,
until the level is exhausted, i.e. until a line is reached on which only one
foot can be assigned. The strong marks of the phrases are projected to the
first line of the next higher level, the clause level, and serve as starting line
for the production of the final grid, following the relevant rules on the
clause level. Again, lines are added, until the level is exhausted. Every
phrase that is dominated by VP and its extended projections IP and CP
projects one strong mark onto the bottom line of the clause level (see
Truckenbrodt 2006; with the exception of phrases that consist only of in-
trinsically weak elements, such as pronominal DPs). In this study, no
higher unit than the clause is taken into account, although the sentence (=
Utterance) constitutes a higher level.
The idea that for each level several lines can be constructed until the
level is exhausted goes back to the notion that phrasal (and clausal) metri-
cal prominence assignment happens cyclically (see e.g. Selkirk 1984). So
the assignment process would proceed as in (8). Since what the metrical
calculus basically does is assign feet, we may as well mark the feet in the
grid.
Some background 15
(8) ( . *)
( . *) ( . *)
* * * * C
(. *)| | | ( . *)
( *) ( . *) | ( *) | ( . *)| ( . *) ( . *)
* * * | * | * * | * * * * P
* | * | * | * | * | * | * | * | * | * ω
[word word word][word] [word word] [word word word word]
This kind of representation takes account of two requirements on grid pro-

duction that seemingly are in conflict with each other: On the one hand,
there is the additive nature of metrical prominence, in the sense that every
level builds on former levels, i.e. that a more prominent metrical promi-
nence is the result of the addition of prominence marks on different levels.
This implies that metrical prominence on the word level and on the phrase
level must be represented in the same grid (cf. Truckenbrodt 2006), as the
final audible gradation in prominence is the addition of metrical promi-
nence marks on the word, phrase and clause level. On the other hand, there
is the fact that the assignment rules are potentially different for each of the
three levels. Take for instance metrical prominence in German: The rule for
the assignment of metrical prominence on the word level is identical to the
Latin rule, namely that the first moraic trochee, counting from the right,
under extrametricality of the final syllable, receives the main prominence
(Speyer 2009b). The rule for metrical prominence on the phrasal level, on
the other hand, also counts from the right, but here it is iambs and not mo-
raic trochees that are assigned – see the version of the Iamb Construction
Rule above. The metrical prominence assignment rule for the clausal level,
in the end, is similar to the rule for the phrasal level, but it treats verbal
material at the edge as extrametrical.
We have to bear in mind furthermore that a metrical grid can be subject
to another process, namely eurhythmy (cf. Hayes 1984). Eurhythmy is ba-
sically a well-formedness condition on grids; the basic rules are, freely after
Hayes (1984), that the highest prominence marks should be kept as far
apart as possible (‘Phrase Rule’), and that in-between a strict alternation of
strong and weak marks should be strived for. The grid in (7) would be per-
fectly eurhythmic. A grid like (8), on the other hand, would not be eu-
rhythmic. The processes trying to obtain eurhythmy would first push the
second highest mark to the first constituent (9a), thereby making the grid
conforming to the Continuous Column Constraint (Hayes 1995: 34–37),
16 Introduction
then repair the equal heights of the intervening material by destressing the
column which is closer to the next highest prominence peak – in that case
the left of the two constituents (9b). Then the grid will be eurhythmic and
an adequate metrical representation of an English sentence with the con-
stituent structure given in (8). Note in this connection that certain function
words such as the article or personal pronouns are not counted into the
computation normally because they do not have word stress and therefore
do not receive a strong mark even on the word level. They are only in-
cluded into the computation when they happen to bear focal emphasis. In
this case they of course receive a grid mark motivated by the focus feature,
the ‘credit grid mark’ which I will elaborate on in section 4.1.
(9) a. ( . *)
( * ) ( . . *)
* * * * C
(. *)| | | ( . *)
( *) ( . *) | (*) | ( . *)| ( . *) ( . *)
* * * | * | * * | * * * * P
* | * | * | * | * | * | * | * | * | * ω
b. ( . *)
( * ) ( . *)
* . * * C
(. *)| | | ( . *)
( *) ( . *) | (*) | ( . *)| ( . *) ( . *)
* * * | * | * * | * * * * P
* | * | * | * | * | * | * | * | * | * ω
b. ( . *)
( * ) ( . *)
* . * * C
(. *)| | | ( . *)
( *) ( . *) | (*) | ( . *)| ( . *) ( . *)
* * * | * | * * | * * * * P
* | * | * | * | * | * | * | * | * | * ω
some dark stranger gave blue flowers to-the mildly surprised girl
Some background 17
Eurhythmy, however, is a special effect of a much more basic requirement

of language, the Principle of Rhythmic Alternation (on its importance for
grammar see e.g. Schlüter 2005). I will postpone a discussion of this prin-
ciple to later sections and chapters (especially 2.4, 3, and 4). The theory as
it is sketched out here is probably too simple and would need additional
features if applied to other problems. But it seems to be accurate for the
domain which we are mainly interested in: the domain of phrasal and
clausal metrics, which I may call summarizingly supraverbal metrics.
It would have been possible to use Optimality Theory with the appropri-
ate metrical formulations (for an overview over such systems in OT see e.g.
Truckenbrodt 2007), but I decided to stay within the frameworks of classi-
cal autosegmental and suprasegmental theories, as a reformulation in OT
terms would have no effect on the results to be described or on the explana-
tions I will be proposing. A short sketch of an OT variant is outlined at the
end of section 3.2. The grid serves as input for the assignment of intonation
contours, indicating the positions of the different low and high pitch ac-
cents and boundary tones (on intonational contours see e.g. Pierrehumbert
[1980] for English, Féry [1993] for German). I assume that grid production
and the assignment of intonational contours are two distinct processes (cf.
also Truckenbrodt 2006). Therefore I will not treat questions of intonation
proper (i.e. contour formation, pitch accent realization) here, but confine
myself to the construction of the grid, as this is sufficient for the purposes
of this study.
1.2.5. The syntactic field model
The Feldermodell (‘field model’) dates from the early years of German
linguistics as a mode of representation for the sentence patterns of Modern
German. It was introduced in the 1820s by Simon Herling (Herling 1821;
see Abraham and Molnarfí 2001), and gained momentum especially under
the influence of Drach (1937). According to the most common versions of
the field model (cf. e.g. Höhle 1986; Grewendorf, Hamm, and Sternefeld
1987; Reis 1987: 147–148; Abraham & Molnárfi 2001), a sentence can be
divided into the following parts which stand in the order given here:
Vorfeld – linke Satzklammer – Mittelfeld – rechte Satzkl. – Nachfeld

prefield left sentence bracket middle field right sent. br. back field
18 Introduction
Before the vorfeld, another – marked and very restricted – position (vorvor-
feld, ‘pre-prefield’) can be introduced.
Each of these ‘fields’ has special properties:
– The verbal elements all stand in the satzklammern. In main clauses the finite
part of the verb is in the left satzklammer, the remainder of the verbal mate-
rial in the right one. In subordinate clauses all verbal material is in the right
satzklammer, the complementizer is in the left one.
– The nachfeld is usually filled with subordinate clauses or otherwise ‘heavy’
elements.
– Most of the non-verbal sentence material stands in the mittelfeld. There are no
constraints whatsoever on what can stand in the mittelfeld, as long as it is not
verbal. There are certain constraints on the order of elements, however (see
e.g. Hoberg [1997], as summarizing representative of an abundant research
literature).
– The vorvorfeld can only contain main clause connectives and material which
can be shown to be left dislocated.
We are mostly interested in the vorfeld. The vorfeld in Modern German can
contain exactly one constituent. There are some exceptions to that, and the
further back in history we go the more frequent these exceptions become,
so that we are forced to assume that the one-constituent-only constraint of
Modern German is a recent development, and that originally more than one
constituent could stand before the left sentence bracket. This is going to be
of immediate importance for Early German and English. We will return to
this question in section 5.3.
It can easily be seen that the Feldermodell translates directly into mod-
ern generative terms (cf. den Besten 1981; Vikner 1995; slightly differently
Sabel 2000): the vorfeld corresponds to SpecCP, the left satzklammer to the
C-head, the mittelfeld to everything under C’ save for the – in German
right-peripheral – V-head(s) and the I-head, which form the right
satzklammer. The nachfeld contains IP adjuncts to the right.
For Modern English, using the field model does not make much sense
and does not offer great insights, although it could be done (the left sen-
tence bracket contains all verbal material, the default filler of the vorfeld is
the subject, although more than one phrase can stand in the vorfeld, and the
distinction between mittelfeld and nachfeld is hard to draw as there is never
overt material in the right sentence bracket). The positions of the field
model would not correspond however to generative entities in Modern Eng-
lish. This is different for earlier stages of English in which the sentential
structure shared some properties with Modern German. Therefore, terms of
Further concepts 19
the field model will occasionally be used for Old and Middle English in the
course of this study.
1.3. Further concepts
1.3.1. Verb second
In this study I will frequently make use of the term verb second (V2). The
usage of the label V2 tended to be rather imprecise in the past, and there-
fore it is perhaps useful to dwell a bit on this subject. V2 can be used in a
more typological manner to express the property a language can have of
putting the verb in the second position in the sentence, that is, the position
after the first constituent. Note that whoever uses V2 in this sense does not
have to commit oneself to a specific analysis: he or she simply states that at
the surface we have the verb in second position, no matter what the under-
lying analysis is that takes care of having the verb at exactly that spot.
A related notion is that of the verb second constraint which on a de-
scriptive level says not much more than the following: some languages
(among which are the Germanic languages) show a tendency to build their
sentences in such a way that the verb is in second position. The reasons for
this tendency are unknown. Brandt et al. (1992) assume the presence of
sentence type features that have to be saturated by movement of the verb to
C and in some cases (with wh-questions and declarative sentences) also
another phrase to SpecCP. Erteschik-Shir (2005) sees it as a phonological
process. Lately the hypothesis has been put forward that verb-seconding
(and by that the creation of a ‘vorfeld’) serves to establish a topic-comment
structure. Under this view, the verb serves as marker which divides the
sentence into these two parts (Hinterhölzl 2009). But this is of no concern
for us here. The only thing to mention is that again, if one uses ‘verb sec-
ond constraint’ on this descriptive level, nothing is said about the underly-
ing structure.
There is however a less non-committal usage of the term. At least since
Vikner (1995), ‘V2’ is often used to denote a special syntactic configura-
tion, in which there is one functional projection above IP (which is usually
referred to as CP). The V2-effect is derived by moving the verb into the
head of that projection and some other constituent into the specifier projec-
tion of it (10). This corresponds closely to the analysis of the Modern Ger-
man declarative sentence by den Besten (1981). When the term V2 is used,
20 Introduction
it is often implied that something like the structure in (10) is necessarily the
underlying structure of any V2-sentence.
The problem is now, of course, that a surface V2 order can be the out-
come of a variety of analyses, of which the one outlined under (10) is only
one. For instance, a verb second order can also be the result of a structure
as in (11).
(10) CP
XP C’
some phrase2
C IP
verb1
… t2 … t1 …
(11) CP
XP C’
some phrase2
C IP
e
e I’
I VP
verb1
… t 2 … t1 …
It turns out that in Old English we have both kinds of V2: V2 by movement
of the verb to C and of some phrase to SpecCP (I will hitherto refer to this
kind of V2 as CP-V2) and V2 by movement of some phrase to SpecCP, but
no movement of the verb from I to C and no element in the specifier posi-
tion of the projection in whose head the verb has landed (e.g. Kroch and
Taylor 1997; Haeberli 2002). I denote it here as IP-V2 for the ease of the
exposition. We will get back to that question more precisely in part 5.
When I use V2 in this study I do not mean V2 by movement of the verb
to C. For this special usage I use the term CP-V2. The structure of V2 I am
mostly concerned with is the version of V2 outlined in (11). It is important
to note that this sentence structure is optional throughout the history of
English (quite in contrast to CP-V2 in languages which have this structure,
where it tends to be compulsory), and therefore it makes sense to speak of
Further concepts 21
the ‘V2 word order option’ when talking about English. This implies, of
course, that all cases of CP-V2, which was used in very limited contexts
throughout the history of English (namely wh-questions, negative inversion
and the like) are not covered by that term. The changes I describe do not
affect CP-V2. English has (CP-)V2 in wh-questions today just the same
way as it had 1200 years ago. The changes affect only V2 without move-
ment of the verb to C. All instances of modern (and thereby also
Old/Middle) English CP-V2 are not subject of this study.
1.3.2. The reconstruction of sentence prosody
Much of the discussion to follow hinges on the assignment of focus (and in

the end focal emphasis) to several elements in the Middle English and
Early Modern English texts that constitute the corpora which are used for
this study (Kroch, Santorini, and Delfs 2004; Kroch and Taylor 2000; Tay-
lor et al. 2003). A problem with this investigation which will come imme-
diately to mind is the fact that prominence is ordinarily not encoded in writ-
ten texts. So how can we base any argumentation about written texts on
focal emphasis? Another question that arises is: can we learn anything
about a prototypically oral phenomenon like prosody of emphasis from
written texts at all?
Let me begin with the latter, more fundamental question. There is clear
evidence from psycholinguistic experimental research that reading and
writing are interdependent (e.g. Patterson and Colehart 1987; Fodor 2002;
cf. also Schlüter 2005: 51ff.). Most importantly in this context is perhaps
the study of Bader (1998), in which scope-ambiguous sentences involving
focus particles (e.g. nur ‘only’ as in [12]) were read silently and afterwards
interpreted by the participants as if the Clash Avoidance Requirement was
at work – i.e. with wide scope instead of narrow scope on ihr, which would
have involved additional emphasis on ihr and thus stress clash.
(12) Zu mir hat Maria gesagt, dass man nur ihr Geld
To me has Mary said that one only her money
beschlagnahmt hat
confiscated has
Reading 1: ‘Mary said to me that the only thing that happened was
that her money was confiscated’ (stress in the German sentence:
…núr ihr…)
22 Introduction
Reading 2: ‘Mary said to me that only the money belonging to her

was confiscated, nobody else’s.’ (stress in the German sentence: nùr
íhr)
There is nothing in the way the sentence is written to promote this interpre-
tation. So the interpretation hinges on a hypothetical oralized version of the
sentence and we can say rather confidently that written texts in general can
serve as objects of research involving prosodic phonology.
Let me now come to the point outlined first. While it is true that promi-
nence – rhythmical and focal – is not encoded, it can be reconstructed nev-
ertheless. There are two facts that make this reconstruction possible: we can
identify the pragmatically based focus structure of a written text, and in the
case of older stages of English and German we may infer what the interac-
tion of focus and prosody looked like from the pragmatic analysis.
It is true in general that we can analyse the information structure of any
piece of text, written or spoken (cf. also Doherty 2006). This is obvious
from the fact that we can read a book or a newspaper and follow the infor-
mation structure without any problems, although there is no direct prosodic
information available. To understand a text always means to be able to
follow its information structure. Now, the assignment of non-rhythmical
(focal) prominence is always governed by information structure, mainly the
parameters of newness and contrast. This means that we can make in-
formed guesses as to which elements of a sentence would receive focal
emphasis if spoken out loud (on this problem see e.g. Petrova and Solf
2009), just in case we can make informed guesses as to which elements are
informationally focused. To identify informational foci in a written text,
however, is not that difficult. If, for instance, the focus theory of Rooth
(1985) is used, all one has to do is to hunt for elements that stand in con-
trast to other elements in the local discourse. So it is possible to identify, at
least approximately, the focus structure of any extended written discourse.
How focus interacts with prosody, on the other hand, is a different mat-
ter. In living languages we can study the interaction directly. In ‘dead’ lan-
guages, we cannot do so, at least in principle, which is a possible objection
to the method used in this study. And, one may object further, it is quite
pointless anyway to map the focus structure of written texts to a hallmark
property of spoken language, viz. prosody.
Both objections can be refuted. Let me begin with the second one. It is
true that prosody is not written down, but it is the same language faculty
that generates spoken utterances and written texts. It is fair to assume that
patterns of syntactic usage that manifest themselves in the spoken language
Further concepts 23
can also be found in written texts of a low to middle stylisitic level (exclud-
ing highly-stylized prose and poetry) since in such texts no other rules –
rhetoric, stylistic, etc. – interfere. From this it follows that, if the normal
usage of syntax in spoken language is prosody-sensitive – and this is espe-
cially true if there is optionality in the syntactic output – we may try assum-
ing that texts of a low to middle stylistic level will show the same prosodic
sensitivity. If this assumption leads to interpretable results in line with other
aspects of our scientific understanding of language structure and history,
we can take the assumption to be justified. Just this outcome is what I hope
to present in the body of this work.
The objection that we do not know how focus interacts with prosody in
a dead language is to some extent well-founded, but here we have to distin-
guish between languages that are really ‘dead’ – such as Sumerian, Egyp-
tian or Hittite – and languages that may not exist in the form in which the
records we are interested in are written, but for which close successor lan-
guages exist which we can study directly. Latin, for instance, is not as
‘dead’ as Sumerian, as there are several daughter and granddaughter lan-
guages of Latin in everyday use by almost a billion people.
The case is even stronger for Old English and Middle English, which
are the languages on whose prosody some parts of the argumentation de-
pend. The reason is that Modern German and Modern English are ex-
tremely similar with respect to the focus-prominence mapping. In both
languages, focus is associated with a pitch accent on the focalized element
itself. The realization of the pitch accent might differ in detail phonetically,
but the basic system is the same (as one sees in comparing e.g. Pierrehum-
bert [1980] and Féry [1993]). From this fact we can infer that focus was
associated with a pitch accent also in the common ancestor of these lan-
guages, which is Proto-West-Germanic. If this is so, however, we can also
conclude that all stages between Proto-West-Germanic and Modern English
and German respectively had the same association. Pretty much the same
goes for phrasal and clausal rhythmical prominence, by the way – the rules
for nuclear stress or phrasal stress assignment are not identical in Modern
German and English, but are so similar that they can be reduced to one
another – hence it is fair to assume that Proto-West-Germanic followed
similar rules and constraints, too.
If these two conditions hold it should be possible to reconstruct both
metrical prominence structure and focus indicator assignment in any text in
English or German with a fair degree of confidence, from their respective
earliest attestation on. I base my discussion in what follows (section 3.1;
chapter 4) on this hypothesis.
Chapter 2
Topicalization in Middle and Modern English –
A prosodically induced change in syntactic usage
The second part of this study is devoted to the exploration of an empirical

fact, the historical decline of topicalization in English. This phenomenon
became evident only with the recent availability of parsed corpora of his-
torical stages of English (York-Toronto-Helsinki Parsed Corpus of Old
English Prose [YCOE]: Taylor et al. 2003; Penn-Helsinki Parsed Corpus of
Middle English [PPCME]: Kroch and Taylor 2000; Penn-Helsinki Parsed
Corpus of Early Modern English [PPCEME]: Kroch, Santorini, and Delfs
2005). The rate of topicalization decreases in the course of the Middle Eng-
lish and Early Modern English periods. Section 2.1 is devoted simply to the
demonstration of this decline. The following sections will discuss three
alternative explanations for this decline, an obvious hypothesis (change in
the pragmatic environments compatible with topicalization) which will be
shown to be insufficient as explanation for the Middle and Early Modern
English data (2.2), another simple hypothesis (decline of topicalization
caused by rigidification of word order), which does not work at all (2.3),
and a more complex one (complex in the sense that more parts of the
grammar are involved), namely that a special kind of topicalization – dou-
ble-focus-topicalization with focused full noun phrase subjects – was, for
prosodic reasons, possible only as long as the V2 word order option was
available. I will argue for this alternative (2.4). This explanation entails that
prosody is more important than unambiguous pragmatic encoding. That this
is true is shown by a comparison of German texts (with unconstrained topi-
calization) with their English translations (with prosodically constrained
topicalization; section 2.5).
2.1. The decline of topicalization
Topicalization in the non-technical sense of moving a constituent other than

the subject to the left edge of the clause is one of the not-too-numerous
examples of a construction that involves non-canonical word order in Mod-
ern English. Non-canonical word order is a cover term for all word orders
The decline of topicalization 25
different from S – V – O. The surface string scheme of topicalization in

Modern English can be given as
X–S–V…,
where S = subject, V = finite verb and X = any constituent. In Old and

Middle English, topicalization could also be of the form
X–V–S…,
because in Old and Middle English the verb-second (= V2) option still
played a role in that V2 was a common word order in declarative matrix
clauses. The verb-second constraint in its weakest form has been a common
property of all Germanic languages, although some of the languages lost or
at least modified this constraint. In its most general form it says that the
verb should occupy the place after the first constituent. It is easy to see that
a sentence of the form X – V – S conforms neatly to this constraint. The
structure of such a sentence would be as given in (11) of the first chapter,
repeated below under (1), that is: with the verb moved not to C°, but to the
highest possible projection of the I-architecture, let us say, T°. C° would be
covertly filled (perhaps by a sentence type operator, as assumed by Brandt
et al. [1992]), so that movement of the verb to C° is impossible. Movement
of the subject to SpecTP would be blocked because an empty expletive
occupies that position (Haeberli 2002).
(1) CP
XP C’
some phrase2
C TP
e
e T’
T MP
verb1
NP M’
subject3
M VP
t1
t3… t2 … t1 …
26 Topicalization in Middle and Modern English
A topicalized sentence with V3 word order, the type familiar from Modern
English, would have a similar structure, with the difference that movement
of the subject into SpecTP would not be blocked (2). What I say here about
the structure of topicalized sentences most certainly goes for Middle Eng-
lish. We will see (ch. 5) that Old English made use of identical structures.
(2) CP
XP C’
some phrase2
C TP
e
NP T’
subject3
T MP
verb1
NP M’
t3
M VP
t1
t3… t2 … t1 …
Let us go back to topicalization. Typical Modern English topicalization

cases are given in (3).
(3) a. In the afternoon, I usually go for a walk.

b. Beans he likes, but peas he hates.
c. This proposal, we discussed at length.
d. Pterodactylus, it is called.
There are several types of topicalization that can be distinguished by the

pragmatic-information structural and intonational properties of these re-
spective types. Examples (3a–d) exemplify some cases, namely preposing
of a scene-setting element (3a), topicalization in the stricter sense in double
focus constructions (3b), preposing of non-contrastive, discourse anaphoric
element (3c) and focus movement (3d). As the respective properties of
these types will become relevant only in section 2.2, I will postpone a more
detailed discussion. The only distinction which I wish to make here is the
one between object topicalization (i.e. preposing of an argument of the
verb, illustrated by (3b, c, d) and topicalization of a scene-setting element
The decline of topicalization 27
(in the form of a prepositional phrase or an adverbial phrase, see 3a). I will
concentrate for the most part of the following on object topicalization.
If we look in the Penn-Helsinki parsed corpora of Middle and Early
Modern English (Kroch and Taylor 2000; Kroch, Santorini, and Delfs
2005) and the York-Toronto-Helsinki Parsed Corpus of Old English Prose
(Taylor et al. 2003), we notice that there is a continuous decline in object
topicalization from earliest Middle English into early Modern English. The
rate of topicalization of direct objects in earliest Middle English is just over
11% – which means that in 11% of all main clauses that have a direct ob-
ject this direct object is preposed – and declines to a rate of about 3.5% by
the late 17th century, a rate comparable to, though perhaps slightly higher,
than the rate in Modern English (Table, Figure 1).
Table 1. Rate of direct object topicalization6
oe1–2 oe3–4 me1 me2 me3 me4 eme1 eme2 eme3
sent. 6184 10002 5329 3642 9608 5583 7719 10103 7057
with DO
whereof 736 1080 570 228 558 257 376 428 247
topical-
ized
% 11.9 10.8 10.7 6.3 5.8 4.6 4.9 4.2 3.6
14
12
10
8
%
0
oe1/2 oe3/4 me1 me2 me3 me4 eme1 eme2 eme3
period
%topic
Figure 1. Rate of direct object topicalization

The decline in topicalization is a feature of virtually all dialects. The texts

of the North are deviant for reasons that will become evident,7 but other-
wise we see a clear tendency in all dialects towards a decline in topicaliza-
tion (Table and Figure 2). It is not completely clear what caused the unex-
pected dip in me2 in the East Midlands, but otherwise the dialect areas save
for the North show the same development.
So we can say that the topicalization of accusative noun phrases, most
of which are functioning as direct object, has declined over the course of
English language history. In quantitative terms, it dropped from around
12% to around 3% between 1100 and 1700 AD.
Table 2. Decline of topicalization in the different dialect areas
North
all sent. 652 759
with DO
whereof 83 120
topicalized
% 12.7 15.8
East Mid-
land
all sent. 2392 1821 4718 3219 7719 10103 7057
with DO
whereof 315 29 267 158 376 428 247
topicalized
% 13.2 1.6 5.7 4.9 4.9 4.2 3.5
West
Midland
all sent. 6184 10002 2827 2109 1815
with DO
whereof 738 1080 246 133 93
topicalized
% 11.9 10.8 8.7 6.3 5.1
South
all sent. 110 1170 2022 549
with DO
whereof 9 116 38 6
topicalized
% 8.2 9.9 1.9 1.1
The pragmatic properties of topicalization 29
18
16
14
12
10
%
8
6
4
2
0
period
North East Midland West Midland South
Figure 2. Decline of topicalization in the different dialect areas
2.2. The pragmatic properties of topicalization
The obvious question is: what can be the cause for this decline? I try to
answer it in the remainder of this chapter. There are three candidate an-
swers, two obvious ones (which prove to be wrong) and one less obvious
one (which I will argue for). The second of the obvious answers will be the
subject of section 2.3, so I will not discuss it here. The first of the obvious
ones, however, I do discuss in this section. It would go like this: we know
that topicalization is sensitive to information structuring processes: In
Modern English, topicalization is only possible if certain discourse struc-
tural requirements are met that will be summarized in this section. A hy-
pothesis which immediately comes to mind is that perhaps in earlier periods
of English there were more discourse structural contexts under which topi-
calization was possible. The decline in topicalization would thus really be a
gradual loss of contexts in which topicalization was felicitous.
In order to test this, the first thing that needs to be done is to examine
the discourse structural contexts in which topicalization is felicitous in
Modern English (2.2.1). In a second step the discourse structural properties
of topicalization in Middle and Old English have to be examined and com-
pared to those in Modern English. It will turn out that in Old English there
was indeed one more discourse structural configuration – viz. topic–
comment in the strict sense – compatible with topicalization. The possibilty

to topicalize in such cases was however lost by earliest Middle English.
However, as the decline did not stop in Middle English but continued, this
cannot be the explanation for the decline in Middle and Early Modern Eng-
lish (2.2.2).
2.2.1. The discourse-pragmatic functions of topicalization in Modern

English
First we have to stop to think under what circumstances Modern English

allows topicalization. It turns out that topicalization can occur in a variety
of rather disjunct contexts in Modern English. The most important ones
have been mentioned under (3) in section 2.1, repeated under (4).8 First I
will elaborate shortly on the distinguishing properties of these types.
(4) a. In the afternoon, I usually go for a walk.

b. Beans he likes, but peas he hates.
c. This proposal, we discussed at length.
d. Pterodactylus, it is called.
Scene-setting preposing is the most common type. Scene-setting elements

have been defined in section 1.2.1 as elements that describe or limit the
situation, in which the proposition with which they are connected is evalu-
ated. As they have scope over the proposition as a whole in some sense, at
least in the semantic sense, it is conceivable that they are frequently moved
to a position where they overtly take the remainder of the utterance into
(structural) scope. Scene-setting elements are usually not in focus. The
prosody of the clause, induced especially by eurhythmy, is such that the
scene-setting element bears metrical prominence of similar height as the
element at the right edge of the clause that is realized as the nuclear stress.
Scene-setting preposing does nothing to establish a theme–rheme (= topic–
comment) or a focus–background structure, it is on a completely different
plane, as one will recall from the introduction in ch.1.
Topicalization in the stricter sense (= double-focus-topicalization) is
exemplified in (4b). Topicalization in the stricter sense, as it is used here, is
confined to preposing in sentences with two foci, one focus on the topical-
ized element, one on some other element in the remainder of the sentence.
The topicalized element in such constructions is often called contrastive
topic (e.g. Lee 2006), a term which I will not use here as I limit the use of
‘topic’ to ‘aboutness-topics’. It is important to point out that the topicalized

constituent is always in focus in this type of topicalization (Barry 1975: 1;
van Hoof 2003; Dryer 2005; contra Birner and Ward 1998: e.g. 83). Focus
is understood in the sense of Rooth (1985): by using focus on an entity, one
evokes a set of alternatives to this entity. In the semantic representation of
the utterance we have thus a variable in the place of the focus and an addi-
tional statement that identifies the variable. Informally in (5):9
(5) a. [[he likes beans]]s,g = LIKE (he, beans)

b. [[BEANS he LIKES]]s,g = ∃x [LIKE (he, x) | (x ∈ M =
{…, beans,…}) ∧ x = beans]
As the reader will remember from the introductory chapter, the type of
focus that most obviously conforms to this definition is contrastive focus,
but other types of focus, like presentational focus (see Rochemont 1986) or
verum-focus (see Höhle 1992) can be derived from that (see e.g. Rooth
[1985: 10–12], who explicitly relates new information together with the
contrastive quality of his set-based proposal).
This notion of focus is very general so far. In fact, it is not enough for
topicalization to have two focal emphases in general. The two foci have
quite distinct semantico-pragmatic properties. For double-focus topicaliza-
tion the generalization holds that the entity referred to by a topicalized con-
stituent stands in a partially ordered set (short: poset) relation to a set
evoked earlier in the discourse, but recently enough that it is still salient
(Hirschberg 1986: 122; Prince 1986: 208–210, 1999: 7–10).
This is, however, not yet a sufficient condition for topicalization. Often
the poset-definition works for both focused phrases in a clause with two
foci. How does the speaker decide which one to topicalize?
Kuno (1982), working on multiple wh-questions, developed the idea
that the wh-phrase selected to be fronted is the one that provides a ‘sorting-
key’ “for sorting relevant pieces of information in the answer” (Kuno 1982:
141; cf. to the following also van Hoof [2003]). Let me illustrate this with
an example, adapted from Kuno (1982: 140–141). Let us assume a multiple
question such as (6).
(6) Which students did they give A’s to in which subjects?
An appropriate answer to that would be (7a), whereas (7b) sounds deviant.

Sentence (7b), however, would be a perfectly felicitous answer to a ques-
tion formulated as in (8), to which in turn (7a) would be an infelicitous

answer.
(7) a. They gave an A to John Doe in geometry, biology and English, to

Richard Roe in history and music,….
b. In geometry, they gave an A to John Doe and Mary
Higginbotham in history to Richard Roe and Jane Merri-
weather,…
(8) In which subjects did they give A’s to which students?
It is easy to see that the surface order of the wh-phrases (and thereby their
scopal relationship, see Kuno [1982: 144]) corresponds to the surface order
of their respective answering phrases. The answering expression in the
higher position is topical in the sense that the following expression is
‘about’ the actual value of the fronted wh-expression and not about the
value of any other wh-expression. This does not mean that it is an arche-
typical topic. It has, however, the characteristic of ‘aboutness’ in common
with archetypical topics (cf. Lee [2006] for discussion on this point). I wish
to point out once more that focus-background and topic-comment are two
entirely different pragmatic dimensions. It is not the case that topics are
automatically background or foci are automatically part of the comment.
But with sorting-key elements, focus and topic intersect, such that a phrase
can be focus and topic at the same time.
We can visualize this in the following way: Questions like (6) and (8)
evoke two lists (therefore Kuno [1982: 137] refers to such cases as ‘multi-
ple-choice questions’). The items on these lists are in a relationship to each
other (9): the relationship, expressed by the question, is ‘have grade A in
(x,y)’, where x is taken from list 2, and y from list 1.
The relationship is, however, not a trivial 1:1 relationship. In order to
organize the information about the real relations holding, one has to decide
first, which list organizes the information in the ongoing discourse. This is
the sorting-key. Let’s say, I am more interested in the subjects list, so I
order the information according to subjects (10). In some ways, this is like
schönfinkelization of a function. In double focus constructions a non-
permutable function between two sets is established, and the sorting keys
are the elements in the set to which the function assigns members of the
second set. The equivalent to this in natural language would be a sentence
in which the sorting-key list has scope over the elements of the other list,
that is, in which the information is ordered after the sorting-key.
(9) List 1: subjects List 2: students
geometry John Doe

arithmetic Richard Roe
biology Jane Merriweather
physics Mary Higginbotham
English Becky Sharp
Spanish James Steward
history
music
art
(10)
geometry John Doe
Mary Higginbotham
arithmetic Mary Higginbotham
biology John Doe

Becky Sharp
… …
What Kuno (1982) does not say explicitly, but clearly assumes, is that the
answer sentences to such a multiple wh-question tends to have a form in
which the scopal relationship is overtly expressed. In other words: The
sorting key elements tend to be placed before the elements of the other list.
This surface ordering has the effect of enforcing the distributivity of the
sorting key list over the second list.
At this point topicalization comes into play. The topicalized element in a
double-focus topicalization sentence corresponds to the sorting key, and it
is only the sorting key that can be topicalized. Let me illustrate this with an
example. A question such as (11a) clearly has as its sorting-key ‘kinds of
vegetables’. The answer (11b) is at least as felicitous as the answer (11c). If
the question gave the persons as sorting-key (as question [11d] does) the
sentence (11b) would no longer be a felicitous answer, only (11c). Note
that both foci fulfil the condition of poset as stated by Prince (1999). Poset
is thus a necessary, but not a sufficient condition on topicalization.
(11) a. Which kind of vegetable does which person like?

b. Beans are the right stuff for John, and peas the right stuff for
Mary.
c. John likes beans, and Mary likes peas.
d. Which person does which kind of vegetable like?
It is this type of topicalization which much of the discussion in this and the
next part of my study is about, since this is the type that potentially pro-
duces prosodic violations as will be discussed in section 2.4.
All kinds of constituents can be topicalized (see Birner and Ward 1998:
45–46), as long as they conform to Prince’s (1999) and Kuno’s (1982) con-
ditions on topicalization. The second focal emphasis can also be on any
element. Birner and Ward (1998) distinguish between ‘normal’ topicaliza-
tion involving two focused referential phrases (12a), and proposition as-
sessment cases, such as proposition affirmation (12b), proposition suspen-
sion (12c) and proposition denial (12d), all of which have in common that
the second focus is some kind of a verum focus. In addition to that I want to
add split topicalization. This construction is special because both foci hap-
pen to be in the same (quantified) phrase; the quantifier is stranded whereas
the content part of the phrase is topicalized (12e).10 Otherwise, conditions
similar to normal topicalization hold. In Modern English the use of split
topicalization is strongly restricted, and example (12e) is just barely accept-
able. I want to introduce it here nevertheless as it will be relevant for the
treatment of German in later sections.
(12) a. Baseball I like a lot better.

b. (It was necessary to pass,) and pass I did.
c. (Mark submitted his report) if submit it he did.
d. John Madden he’s not.
e. He knows tons of endocrinologists, surgeons and laryngologists.
But ophthalmologists he knows none.
(examples a–d from Birner and Ward 1998)
Anaphoric preposing is the preposing of a noun phrase whose referent is

typically not an entity but a set of propositions, namely a relevant portion
of the previous discourse. This noun phrase is most often a demonstrative
pronoun (13a), but full NPs are possible, too, with (13b) or without (13c) a
demonstrative pronoun.
(13) a. This we all know.

b. This discussion I find important.

c. The argument so far everybody could follow.
This construction is highly specific in function. The anaphoric noun phrase

object serves to refer to one or more propositions of the previous discourse
in a summarizing fashion. Preposing the anaphoric object promotes it to a
sentence topic. It fulfils both conditions on typical topichood: its referent is
discourse-old, and it is what the sentence is about. Let us refer to this kind
of topic as a φ-topic (for propositional topic). It differs from an archetypical
topic (let us call it an e-topic, for entity topic) in that the referent of the
anaphoric noun phrase is not an entity but a proposition. In a way an e-topic
is then a topic in first-order logic – an entity that can be an argument to a
predicate; remember that topichood is nothing more than a property of an
argument in a given proposition – whereas a φ-topic is a priori a topic in
second-order logic; the proposition containing it must take a second-order-
object as one of its arguments. The anaphoric process here is not much
more than an act of reference to this second-order-object.
Typically, sentences with a (preposed) anaphoric noun phrase are
‘comments’ (in the sense of rhetorical relations; ⇓ in Asher and Lascarides
2003) on previous discourse segments. These sentences as a whole, as said
above, are the subject matter of the comment made by the speaker/writer,
and therefore can be regarded as topic of this ‘comment’. By virtue of be-
ing topic, they do not bear focal emphasis. Normally such comments have
wide focus, and therefore no other focal emphasis appears in them. As the
highest clausal metrical prominence peak falls on an element in the verb
phrase, that is, close to the right periphery, we get the same prosodic con-
tour as the ones which we observed with preposed scene-setting elements,
namely the highest clausal metrical prominence peak on the right edge of
the clause, another metrical prominence peak on the preposed anaphoric
noun phrase and lower metrical prominence on the subject (14; 15).
(14) scene-setting: (*) (. *)

* * *
Yesterday John arrived
(15) anaphoric: (*) (. *)

* * *
This discussion I find important.
Focus Movement, finally, is the preposing of a single focalized element in

an utterance. For Focus Movement the condition holds that the preposed
element stands in a relation to an entity or set already evoked in the dis-
course (cf. Birner and Ward 1998: 84ff.). This is only a necessary condi-
tion. More specifically, the focalized element in the preposed phrase often
represents a value of an attribute (not in the syntactic sense) that applies to
some entity already evoked in the discourse (Prince 1981b: 259). The value
itself is new information, the attribute (expressed by the predicate of the
sentence in which focus-movement takes place) is explicitly stated in the
previous discourse or is at least inferable (Prince 1981b: 259). A typical
example would be (16a; from Prince 1981b: 259), where the value is three
and the attribute is cooking n meals a day.
Furthermore, Focus Movement very often serves to specify the relevant
member(s) of a more general set evoked earlier, such as in Birner and
Ward’s example (95b) in (1998: 84), quoted below as (16b). It is important
to note that the preposed element – the specification – is a referential ex-
pression that specifies the reference more precisely than anything else that
was said earlier in the discourse about the relevant member of the set: often
it is exact dates (such as [16c] = Birner and Ward’s [95c]), unique names
(16b) or the like that are subject to Focus Movement. Thus it is not surpris-
ing that examples like (16d) in which a new term is introduced for a salient
entity are very common, perhaps the most common type of examples. An-
other field of use for Focus Movement is in corrections (16e). We can view
corrections merely as a special case of specification: the set and even the
relevant member of the set have been evoked earlier, but the member has
been referred to with the wrong term in the previous mention. By the cor-
rection the correct relationship between entity and referring expression is
established. In that way it refers more precisely than the previous (wrong)
referring expression (see Birner and Ward 1998: 86 for more examples).
Focus Movement is also used in confirmation questions such as (16f;
adapted from Birner and Ward 1998: 88) in which the person asking wants
to make sure the precise identity of the relevant member of the set.11
(16) a. You […] had to cook for ten childrens on Sunday. […] Three
meals a day I cooked on Sunday
b. A: Are there many black kids in that school now?
B: Not many. I had two really good friends. Damon and Jimmy
their names were.
c. I promised my father – on Christmas Eve it was – to kill a
Frenchman at the first opportunity I had.
d. A: Oh yeah, and here we see a fossil of this weird flying reptile…

Pterodactylus it’s called …
e. A: Now we’ve got this flying lizard… the peri… plero… plesio-
tantalus…
B: Pterodactylus it’s called
f. A (= customer at fast food joint): Gimme a cheeseburger, large
fries and a large Coke.
[five minutes elapse]
B (= employee at fast food joint): Large Coke you ordered?
Later, when the Clash Avoidance Requirement is introduced, it will be easy

to see that Focus Movement never produces an utterance at odds with the
Clash Avoidance Requirement. This is because such sentences contain only
one focus. Consequently there are no other elements that can clash with it.
We will see that a clash occurs if two elements with the same level of
prominence are adjacent to each other. In a clause with focus movement
this can never happen.
At least these four types of topicalization can be distinguished in Mod-
ern English. The list is not necessarily comprehensive, but it is fair to say
that these types are the most common ones. Because in the later argumenta-
tion the prosodic properties of these respective types of topicalization play
a role, it is necessary to point out that double-focus-topicalization is the
only one of these types that can produce prosodic clashes. In the other cases
there is either only one element with maximal focal prominence (focus
movement) or no focal emphasis is involved (scene-setting preposing; ana-
phoric preposing). In these cases the rules will automatically generate a
grid in which the metrical prominence peak on the preposed element and
the nucleus will be separated by at least one phrase (viz. the subject) that
bears lower metrical prominence.
2.2.2. The discourse-pragmatic functions of topicalization in Old and

Middle English
From the preceding section we know when topicalization can occur in

Modern English, and we were able to distinguish four types. If we now turn
to Old English it looks at first glance as if the hypothesis that the loss of
topicalization is a loss of environments in which topicalization is pragmati-
cally felicitous hits home: apart from the four pragmatically determined
enviroments in which topicalization may occur in Modern English we find
a fifth, namely topicalization of an aboutness e-topic. Two examples are

given in (17). The topic of the discourse portion from which sentence (17a)
is taken is the Holy Spirit, which is referred to by means of an anaphoric
pronoun, as is quite usual in Old English. One could argue that the usage of
a demonstrative pronoun could always have some deictic force and there-
fore this is perhaps not a simple e-topic. But we can easily find examples in
which the topicalized element is a personal pronoun (17b), and here we can
be sure that they are not deictic at all. Example (17b) has an accusative
experiencer in the preverbal position. Some of the examples with topical-
ized personal pronoun can be related to accusative experiencer verbs and
their tendency to put the experiencer first (like in German, where Mir ge-
fällt das Haus is less marked then das Haus gefällt mir ‘I like the house’,
although vorfeld-movement of dative objects is otherwise marked in Ger-
man), but by far not all. (17c) shows an example of a sentence where it is a
non-deictic personal pronoun which doubtlessly represents an e-topic that is
topicalized.12
(17) a. Þone asende se Sunu,

this sent the Son
‘The son sent this one’
(coaelhom,+AHom_9:114.1350)
b. ne hine ne drehð nan ðing,
and-not him not troubled no thing
‘and nothing troubled him’
(coaelhom,+AHom_11:558.1780)
c. & hit Englisce men swyðe amyrdon.
and it English men fiercely prevented
‘and the Englishmen prevented it fiercely’
(cochronE,ChronE_[Plummer]:1073.2.2681)
For reasons into which I will go in section 5.1, I view personal pronouns as
ordinary noun phrases and not as syntactic clitics. The fact that e-topics are
highlighted by topicalization is a property Old English inherited from
Proto-Germanic; the fact that two not directly related languages such as e.g.
German (Jacobs 2001; Speyer 2008a) and Swedish (Rahkonen 2006) use
the sentential-initial position as a topic position indicates that this is a heri-
tage from a stage of the language before North and West Germanic split.
The presence of topicalized personal pronoun objects can be used as in-
dicator for the topicalization of e-topics. We know that e-topics tend to be
realized pronominally, as e-topics are maximally salient. For that reason,
reference by a pronoun is quite unambiguous and therefore unproblematic

(see e.g. Walker, Joshi, and Prince 1998). Thus, we may say that pronouns
are the archetypical mode of realization of an e-topic. A consequence of
this fact is that, if we can find topicalized pronouns in a fairly large number
in a given period, we can assume that e-topics could be topicalized in this
period. As pronominal objects can be found easily in syntactically parsed
corpora we can measure the frequency of these likely cases. In Table 3 and
Figure 3 we see what we get.
Table 3. Number and rate of topicalized personal pronoun objects
oe oe me1 me2 me3 me4 eme1 eme2 eme3

1/2 3/4
all pr. 200 603 285 213 454 316 107 155 96
obj.
whereof 22 40 11 2 8 0 1 0 1
tpczd.
% 11.0 6.6 3.9 0.9 1.8 0 0.9 0 1.0
30
25
20
15
%
10
0
oe 1/2 oe 3/4 me1 me2 me3 me4 eme1 eme2 eme3
period
% topic. O-pron.
Figure 3. Number and rate of topicalized personal pronoun objects
The percentage is the rate of topicalized pronominal objects among sen-

tences with pronominal objects. We see that it begins to trail down already
in the Old English period, while it is practically gone after the first period
of Middle English. The solitary examples after that are either consciously
archaic (and this archaic use can be found even much later, see [18]) or
contrastive, hence not e-topics.
Exploring the Variety of Random
Documents with Different Content
The Project Gutenberg eBook of Jack the Englishman
This ebook is for the use of anyone anywhere in the United States
and most other parts of the world at no cost and with almost no
restrictions whatsoever. You may copy it, give it away or re-use it
under the terms of the Project Gutenberg License included with this
ebook or online at www.gutenberg.org. If you are not located in the
United States, you will have to check the laws of the country where
you are located before using this eBook.
Title: Jack the Englishman
Author: H. Louisa Bedford
Illustrator: Walter Paget
Release date: November 12, 2019 [eBook #60676]
Language: English
Credits: Produced by Al Haines
*** START OF THE PROJECT GUTENBERG EBOOK JACK THE

ENGLISHMAN ***
Cover art
THE BAG BROKE WITH THE FORCE OF THE BLOW. p. 35.
JACK,
THE ENGLISHMAN
BY
H. LOUISA BEDFORD
AUTHOR OF
"HER ONLY SON, ISAAC" "MRS. MERRIMAN'S GODCHILD," ETC.
ILLUSTRATED BY WAL PAGET
LONDON
SOCIETY FOR PROMOTING
CHRISTIAN KNOWLEDGE
NEW YORK AND TORONTO: THE MACMILLAN CO.
Printed in Great Britain by Wyman & Sons Ltd.,
London, Reading and Fakenham.
CONTENTS
CHAPTER
I. HIS TITLE
II. A CHUM
III. NEW NEIGHBOURS
IV. A BUSH BROTHER
V. A CHURCH OFFICIAL
VI. MINISTERING CHILDREN
VII. A BISHOP'S VISIT
VIII. TWO LEAVE-TAKINGS
IX. A SURPRISE VISIT
X. A BUSH TOUR
XI. A NARROW ESCAPE
XII. GOING HOME
XIII. TWO VENTURES OF HOPE
JACK, THE ENGLISHMAN
CHAPTER I
HIS TITLE.
It was a beautiful spring afternoon in the northern hill districts of

Tasmania. The sky was of a bird's egg blue, which even Italy cannot rival,
and the bold outline of hills which bounded the horizon, bush clad to the
top, showed a still deeper azure blue in an atmosphere which, clear as the
heaven above, had never a suggestion of hardness. Removed some half-
mile from the little township of Wallaroo lay a farm homestead nestling
against the side of the hill, protected behind by a belt of trees from the keen,
strong mountain winds, and surrounded by a rough wood paling; but the
broad verandah in the front lay open to the sunshine, and even in winter
could often be used as the family dining-room. The garden below it was a
mass of flowers for at least six months in the year, and there was scarcely a
month when there was a total absence of them.
The house, one-storied and built of wood like all the houses in the
country districts, was in the middle of the home paddock; the drive up to it
little more than a cart track across the field, which was divided from the
farm road which skirted it by a fence of tree trunks, rough hewn and laid
one on the top of the other. A strong gate guarded the entrance, and on it sat
Jack, the Englishman, his bare, brown feet clinging to one of the lower bars,
his firmly set head thrown back a little on his broad shoulders as he rolled
out "Rule Britannia" from his lusty lungs. Many and various were the
games he had played in the paddock this afternoon, but pretending things by
yourself palls after a time, and Jack had sought his favourite perch upon the
gate and employed the spare interval in practising the song which father had
taught him on the occasion of his last visit. He must have it quite perfect by
the time father came again. It was that father, an English naval captain,
from whom Jack claimed his title of "Jack, the Englishman," by which he
was universally known in the little township, and yet the little boy, in his
seven years of life, had known no other home than his grandfather's pretty
homestead.
"But o' course, if father's English, I must be English too. You can't be
different from your father," Jack had said so often that the neighbours first
laughed, and then accepted him at his own valuation, and gave him the
nickname of which he was so proud.
About the mother who had died when he was born, Jack never troubled
his little head; two figures loomed large upon his childish horizon, Aunt
Betty and father. Aunts and mothers stood about on a level in Jack's mind; it
never suggested itself to him to be envious of the boys who had mothers
instead of aunts, for Aunt Betty wrapped him round with a love so tender
and wholesome, that the want of a mother had never made itself felt, but
father stood first of all in his childish affection.
It was more than eight years since Lieutenant Stephens had come out
from England in the man-o'-war which was to represent the English navy in
Australian waters, and at Adelaide he had met Mary Treherne, a pretty
Tasmanian girl, still in her teens, who was visiting relations there. It was a
case of love at first sight with the young couple, who were married after a
very short engagement. Then, whilst her husband's ship was sent cruising to
northern seas, Mary came back to her parents, and there had given birth to
her little son, dying, poor child, before her devoted husband could get back
to her. Since then Lieutenant Stephens had received his promotion to
Captain, and had occupied some naval post in the Australian
Commonwealth, but his boy, at Betty Treherne's urgent request, had been
left at the farm, where he led as happy and healthful an existence as a child
could have. The eras in his life were his father's visits, which were often
long months apart, and as each arrival was a living joy, so each departure
was grief so sore that it took all little Jack's manhood not to cry his heart
out.
"Some day—some day," he had said wistfully on the last occasion,

"when I'm a big boy you'll take me with you," and his father had nodded
acquiescence.
"It's not quite impossible that when I'm called back to England, I may
take you over with me and put you to school there, but that is in the far
future."
"How far?" Jack asked eagerly.
"That's more than I can tell; years hence very likely."

But even that distant hope relieved the tension of the big knot in Jack's
throat, and made him smile bravely as father climbed to the top of the crazy
coach that was to carry him to the station some eight miles away.
From that time forward, Jack insisted that Aunt Betty should measure
him every month to see if he had grown a little.
"Why are you in such a hurry to grow up?" she asked, smiling at him one
day. "You won't seem like my little boy any more when you get into
trousers."
"But I shall be father's big boy," was the quick rejoinder, "and he'll take
me with him to England when he goes. Did he tell you?"
Aunt Betty drew a hard breath, and paled a little.
"That can't be for years and years," she said decidedly.
"He said when I'm big, so I want to grow big in a hurry," went on Jack,
all unconscious how his frank outspokenness cut his aunt like a knife. Then
he turned and saw tears in her pretty eyes, and flew to kiss them away.
"But why are you crying, Aunt Betty? I've not been a naughty boy," he
said, reminiscent that on the occasion of his one and only lie, the enormity
of his sin had been brought home to him by the fact that Aunt Betty had
cried.
She stooped and kissed him now with a little smile.
"I shan't like the day when you go away with father."
"But o' course you'll come along with us," he said, as a kind of happy
afterthought, and there they both left it.
And now Aunt Betty's clear voice came calling down the paddock.
"Jack, Jack, it's time you came in to get tidy for tea," but Jack's head was
bent a little forward, his eyes were intently fixed upon a man's figure that
came walking swiftly and strongly up the green lane from the township, and
with a shrill whoop of triumph he sprang from his perch, and went
bounding towards the newcomer.
"Aunt Betty, Aunt Betty," he flung back over his shoulder, "it's father,
father come to see me," and the next minute he was folded close to the
captain's breast, and lifted on to his shoulder, a little boy all grubby with his
play, but as happy and joyful as any child in the island.
And across the paddock came Aunt Betty, fresh as the spring day in her
blue print gown, and advancing more slowly behind came Mr. and Mrs.
Treherne.
"A surprise visit, Father Jack, but none the less a welcome one," said Mr.
Treherne. He was a typical Tasmanian farmer with his rough clothes and
slouch hat, but a kindly contentment shone out of his true blue eyes, and he
had an almost patriarchal simplicity of manner. He bore a high name in all
the country-side for uprightness of character, and was any neighbour in
trouble Treherne was the man to turn to for counsel and help. And his wife
was a help-meet indeed, a bustling active little woman, who made light of
reverses and much of every joy. The loss of her eldest daughter had been
the sharpest of her sorrows, and the gradual drifting of her four sons to
different parts of the colony where competition was keener and money
made faster than in "sleepy hollow," as Tasmania is nicknamed by the
bustling Australians. There was only one left now to help father with the
farm, Ted and Betty out of a family of seven!
But still Mrs. Treherne asserted confidently that the joys of life far
outweighed its sorrows. Perfectly happy in her own married life, her heart
had gone out in tenderest pity to the young Lieutenant so early left a
widower, and a deep bond of affection existed between the two. She took
one of his hands between her own, and beamed welcome upon him.
"It's good luck that brings you again so soon."
"It's a matter of business that I've come to talk over with you all, but it
can wait until after supper. I'm as hungry as a hunter. I came straight on
from Burnie without waiting to get a meal."
"If you had wired, you should have had a clean son to welcome you,"
said Betty. "Climb down, Jack, and come with me and be scrubbed. Don't
wait for us, mother. The tea is all ready to come in."
Jack chattered away in wildest excitement whilst Aunt Betty scrubbed

and combed, but Betty's heart was thumping painfully, and she answered
the boy at random, wondering greatly if the business Father Jack talked
about implied a visit to England, and whether he would want to take his
little son with him.
"He has the right! of course he has the right," she thought. "Aunts are
only useful to fill up gaps," and her arms closed round little Jack with a
yearning hug.
"There! now you're a son to be proud of, such a nice clean little boy
smelling of starch and soap," she said merrily, with a final adjustment of the
tie of his white sailor suit, and they went down to tea hand in hand, to tea
laid in the verandah, with a glimpse in the west of the sun sinking towards
its setting in a sky barred with green and purple and gold.
Little Jack sat by his father, listening to every word he said, and directly
tea was ended climbed again on to his knee and imperatively demanded a
story. It was the regular routine when Father Jack paid a visit.
"And what is it to be?" asked the captain
"Why, Jack, the Giant Killer, or Jack and the Beanstalk. I love the stories
about Jacks best of all, because Aunt Betty says the Jacks are the people
who do things, and she says you and all the brave sailors are called Jack
Tars, and that I'm to grow up big and brave like you, father."
The Captain's arm tightened round his son.
"It's very kind of Aunt Betty to say such good things about the Jacks of
the world. We must try and deserve them, you and I. Well, now, I'm going to
tell you a sort of new version of Jack, the Giant Killer."
"What's a new version?" asked Jack, distrustfully.

"The same sort of story told in a different way, and mine is a true story."
"Is it written down in a book? Has it got pictures?"
"Not yet; I expect it will get written down some day when it's finished."
"It isn't finished," cried Jack in real disappointment.
"Wait and listen—There was once a man——"
"Oh, it's all wrong," said Jack impatiently. "It's a boy in the real story."
"Didn't I tell you mine was a new version? Now listen and don't interrupt
——"
Mr. Treherne leant back in his chair, listening with a smile to the
argument between father and son as he smoked his pipe; Mrs. Treherne had
gone off into the house, whilst Betty, after setting the table afresh for Ted
who would be late that evening as he was bringing home a mob of cattle,
seated herself in the shadow, where she could watch the Captain and Jack
without interruption.
"There was once a man," began the Captain over again, "who looked
round the world, and noticed what a lot of giants had been conquered, and
wondered within himself what was left for him to do."
"No giants he could kill?" asked Jack excitedly, "Were those others all
deaded?"
"Not deaded; they were caught and held in bondage, made to serve their
masters, which was ever so much better than killing them."
"What were their names?"
"Water was the name of one of them."
Jack stirred uneasily. "Now you're greening me, father"—the term was
Uncle Ted's.
The Captain laughed. "Didn't I tell you this was a true story? Water was
so big a giant that for years and years men looked at it, and did not try to do
much with it. The great big seas——"
"I know them," cried Jack. "Aunt Betty shows them to me on the map,
and we go long voyages in the puff-puff steamers nearly every day!"
"Ah! I was just coming to that. At first men hollowed out boats out of
tree trunks, and rowed about in them, timidly keeping close to the shore,
and then, as the years rolled on, they grew braver, and said: 'There's another
giant that will help us in our fight with water. Let us try and catch the wind.'
So they built bigger boats, with sails to them which caught the wind and
moved the ships along without any rowing, and for many, many years men
were very proud of their two great captive giants, water and wind, and they
discovered many new countries with their wind-driven ships, and were
happy. But very often the wind failed them, grew sulky, and would not
blow, and then the ship lay quiet in the midst of the ocean; or the wind was
angry, and blew too strong—giants are dangerous when they lose their
temper—and many a stately ship was upset by the fury of the wind, and
sent to the bottom. Then men began to think very seriously what giant they
could conquer that would help them to make the wind more obedient to
their will, so they called in fire to their aid. Fire, properly applied, turned
water into steam, and men found that not only ships, but nearly everything
in the world could be worked through the help of steam."
Jack was getting wildly interested in the new version. "Oh, but I know,"
he said, clapping his hands. "There's trains, and there's steam rollers; I love
it when they come up here, and there's an engine comes along and goes
from farm to farm for the threshing, and that's jolly fun for the threshers all
come to dinner, and——"
"Yes, I see you know a lot about these captive giants after all," said the
Captain, bringing him back to the point.
"Go on, please; it's just like a game," said Jack. "Perhaps I'll find out
some more."
"I can't go on much longer. It would take me all night to tell you of all
the giants we keep hard at work. Three are enough to think of at a time. Tell
me their names again for fear you should forget."
"Water—one. Wind—two, and Fire, that makes steam—three," said

Jack, counting them off, as he rehearsed them on his father's fingers. "Just
one more, daddy dear," a phrase he reserved for very big requests.
"One more then, and away you go to bed, for I see Aunt Betty looking at
her watch. The last giant that the man of the story very much wishes to
conquer, and has not done it yet, is air. He wants to travel in the air faster
than any train or steamship will take you by land or water."
"Like my new toy, the one grandmother sent for on my birthday seven.
She sent for it all the way to Melbourne, an 'airyplane' she calls it, but it
only goes just across the room, and then comes flop."
"That's just it; at present flying in the air too often ends in flop, and this
man I'm talking of wants to help to discover something that will make
flying in the air safer and surer. There are lots of men all over the world
trying to do the same thing. All the giants I have told you of are too big and
strong for one man to grapple with by himself, but many men joined
together will do it, and the man of the story has been working at it by
himself for years, and at last—at last he thinks he has discovered something
that will be of service to airmen and to his country, and he's going over to
England to test it—to see if his discovery is really as good as he believes it
to be."
Little Jack sat grave and very quiet, pondering deeply.
"What's the man's name, father? The man you're telling about."
"Jack, a Jack who will be well content if he can help to do something big
in conquering the giant Air. It's your father who is the man of the story. I
promised it should be a true one."
Jack's answer seemed a little irrelevant. He slipped from his father's knee
and took his hand, trying with all his might to pull him up from his chair.
"Come, father, come quick and see how big I've grown. Aunt Betty
measures me every month, and says I'm quite a big boy for my age."
Wondering at the sudden change of subject, the Captain humoured his

little son, and allowed himself to be dragged to the hall where, against the
doorpost of one of the rooms, Jack's height was duly marked with a red
pencil.
"Aunt Betty's right. You're quite a big boy for only seven years old."
"I knewed it," cried Jack, in rapturous exultation, "so you'll take me
along with you, dear, and we'll hit at that old giant Air together. Oh, I'm so
glad, so glad to be big."
"Not so fast, sonny," said the Captain, gently gathering him again into
his arms. "You're a big boy for seven years old, but you're altogether too
young for me to take you to England yet."
Jack's face went white as the sailor suit he wore, and his great round eyes
filled to the brim with tears, but by vigorous blinking he prevented them
from falling down his cheeks.
"You said—perhaps when I was big you'd take me with you."
"And that will be some years hence when I'll come back to fetch you,
please God."
"Me and Aunt Betty, too," said Jack, with a little catch in his throat,
"'cause she'll cry if I leave her."
"Jack, it's bedtime, and you will never go to sleep if you get so excited,"
said Aunt Betty decidedly, feeling all future plans swamped into
nothingness by the greatness of the news Father Jack had come to tell.
"Look here, I'll carry you pig-a-back," said the Captain, dropping on to
all fours. "Climb up and hold fast, for the pig feels frisky to-night, and I
can't quite tell what may happen." So Jack went off to his cot in Aunt
Betty's room in triumph and screams of laughter, but the laughter gave way
to tears when bathed and night-gowned he knelt by Aunt Betty's side to say
his prayers. The list of people God was asked to bless was quite a long one,
including various friends of Jack's in the township, but last of all to-night
came his father's name.
"God bless Father Jack, and make Little Jack a good boy and very big,
please, dear God, so as he'll soon have father to fetch him home."
And then, with choking sobs, Jack sprang to his feet and into bed.
"Tuck me in tight, Aunt Betty, and don't kiss me, please. I'll tuck my
head under the clothes, and don't tell father I'm crying. It's only little boys
who cry, he says, and I want to be big, ever so big. I'll grow now, shan't I?
Now I've asked God about it."
Aunt Betty's only answer was a reassuring pat on his back as she tucked
the bedclothes round him. Truth to tell she was crying a little too.
CHAPTER II
A CHUM
"You've sprung it upon us rather suddenly, Jack."
Betty and her brother-in-law sat in the verandah in the glory of the
Tasmanian night. The stars shone out like lamps from the dark vault above
with a brilliancy unknown in our cloudier atmosphere; a wonderful silence
rested on the land, except that at long intervals a wind came sighing from
the bush-clad hills, precursor of the strong breeze, sometimes reaching the
force of a gale, that often springs up with the rising of the sun.
Jack removed his pipe and let it die out before he answered Betty.
"To you I expect it may seem a fad, the result of a sudden impulse, but
really I've been working towards this end ever since aviation has been
mooted, spending all my spare time and thought upon the perfecting of a
notion too entirely technical to explain to anyone who does not understand
aeroplanes. Finally I sent over my invention to an expert in the Admiralty,
with the result that I've received my recall, and am to work it out. There is
no question that at this juncture, when all nations are hurrying to get their
air fleet afloat, we are singularly behindhand, and I feel the best service I
can give my country is to help, in however small a degree, to retrieve our
mistake."
"You don't really think England is in peril, do you?"
"The unready man is always in peril, and England is singularly unready

for any emergency at the present time. I believe with some men the call of
country is the strongest passion in their blood. For a moment the thought of
leaving the little lad staggered me, for, of course, he's altogether too young
to think of taking him with me. Nobody would mother him as you are
doing, Betty. I would like him to be with you for some years longer yet, if
you agree to continue taking charge of him."
"But of course," said Betty, with a little catch in her throat. "He is my
greatest joy in life. I dread the time when I must let him go."
"Thank you; I want to leave him here as long as possible until it becomes
a question of education. Of course I would like if he shows any inclination
that way that he should follow in my footsteps, either serve in the navy or in
the air fleet."
Betty gave a little gasp. "But the peril, Jack! Think of the lives that have
been already sacrificed."
Jack shrugged his shoulders. "By the time the boy is old enough to think
of a profession, I don't suppose aviation will be much more dangerous than
any other calling that is distinctly combative in character, and if it is, I hope
my son will be brave enough to face it. However, what Jack will be or do
when he grows up is too far a cry to discuss seriously."
"And meanwhile what do you want me to do with him?"
"Just what you are doing now. Bring him up to fear God and honour the
King."
"And when education presses? I can teach him to read and write and a
little arithmetic, but when he ought to go further? Am I to send him away to
a boarding school?"
"I think not, Betty. I would almost rather you let him go to the State
school here, and kept him under your own eye. I don't believe association
during school hours with all and sundry will hurt him whilst he has you to
come back to, and the teaching at some of these schools is far more
practical and useful than at many a preparatory school at home. What can
you tell me of the master here?"
"He's rather above the average, and if he finds a boy interested in his
work is often willing to give him a helping hand. For one thing, I don't
believe Jack will ever want to be much off the place out of school hours.
He's a manly little chap, and loves being about with Ted or father on the
farm. I wish sometimes he had some chum of his own, a little brother, or
what would be almost as good—a little sister. His play is too solitary."
"I'm afraid it's out of your power or mine to cure that," said the Captain,
rather sadly, his thoughts going back to the pretty wife who had been his for
so short a time.
When little Jack appeared at breakfast the following morning there was
no sign of the previous night's emotion, but he was quite inseparable from
his father that day, never leaving his side for an instant if he could help it.
He was much graver than usual, intent upon watching the Captain's every
movement, even adjusting his own little shoulders to exactly the same angle
as his father's, and adopting a suspicion of roll in his walk.
The Captain was to leave by the evening coach, and Betty catching the
wistful look in little Jack's eyes suggested that he should be the one to
escort the Captain down the green lane to the hotel in the township from
which the coach started. Jack, holding his father's hand tight gripped in his
own, scarcely uttered a word as they walked off together. He held his head
high and swallowed the uncomfortable knot in his throat. Not again would
he disgrace his manhood by breaking into tears.
"I'll be real big when you come next time," he ventured at last. "Will it
be soon?"
"As soon as I can make it, Jackie. Meanwhile you'll be good and do as
Aunt Betty tells you."
"Yes, sometimes. I can't always," said Jack truthfully.
"Well, as often as you can. And little or big you'll not forget you're Jack,
the Englishman, who'll speak the truth and be brave and ready to fight for
your country if need be."
"Yes," said Jack, squaring his shoulders a little.
"And I'll write to you from every port—Aunt Betty will show you on the
map the ports my ship will touch at—and when I get home I shall write to
you every week."
That promise brought a smile to Jack's twitching lips.
"Oh, but that's splendid! A letter all my own every week," he said,
beginning to jump about with excitement at the prospect.
"Will it have my name written upon the envelope?"
"Why, yes. How else should the postman know whom it's for? You'll
have to write to me, you know."
That proposition did not sound quite so delightful, and Jack's forehead
puckered a little. He remembered the daily tussle over his copy-book.
"I don't write very well just yet," he said.
"That will have to be amended, for a letter I must have every week. Aunt
Betty will guide your hand at first, and very soon I hope you will be able to
write me a sentence or two all your own, without Aunt Betty's help."
"But what'll I say in a letter?" asked Jack, still distrustful of his own
powers.
"Just what you would say to me if you were talking as you're talking
now; how you get on with your lessons. If you're a good boy or a bad one,
who you meet, what picnics you have; anything you like. What interests
you will surely interest me."
The thought that father would still talk to him when he was away kept
Jack steady through the parting, that, and the fact that a young horse only
partially broken in was harnessed to the steady goer who for months past
had been one of the hinder pair of the four-horse coach, played all manner
of pranks at starting; at first declining to budge at all; then, when the
superior force of the three others made movement necessary, setting his
four legs together and letting himself be dragged along for a few paces,
finally breaking into a wild gallop which was checked by his more sober
mates, until at last finding himself over-matched he dropped into the quick
trot of the other three, fretting and foaming at the mouth, nevertheless, at
his enforced obedience. It was a primitive method of horse-breaking, but
effectual. And so Jack's farewells to his father were diversified by watching
the antics of the unbroken colt, and joining a little in the laughter of the ring
of spectators that had gathered round to see the fun. But when the final start
was made Jack was conscious of the smarting of unshed tears, rubbed his
eyes vigorously with his sturdy fists and set off home at a smart trot,
standing still sometimes and curvetting a little in imitation of the colt that
needed breaking in.
Betty, who stood waiting for him at the gate of the paddock, ready to
comfort and console, saw him gambolling along like a frisky horse, and felt
her sympathy a little wasted. Children's sorrows are proverbially
evanescent, but she was hardly prepared for Jack to return in such
apparently rollicking spirits from the parting with his father of indefinite
duration. And when he came up to her it was of the horse and its capering
that he told her, mimicking its action in his own little person: holding back,
pelting forwards, trying to rear, interspersed with vicious side kicks, and
finally a wild gallop which sobered into a trot.
"That's 'zackly how he went," he said, waiting breathless for Aunt Betty
to catch him up.
Betty was extremely astonished that Jack made no mention of his father,
but later she understood. Tea was over, and before Jack went to bed Betty
allowed him a quarter of an hour's play at any game he chose.
"Would you like to be the frisky horse again, and I will drive you," she
asked, willing to humour his latest whim.
"No, I'll get my slate and write, only you must help me."
This was indeed an unexpected development for Jack, and left Betty
speechless. Jack was quick at reading and quite good at counting, but
writing was his particular bug-bear.
She lifted him on to her lap, and he bent eagerly over the slate on his
knees.
"Now, what do you want to write," Betty asked, taking his right hand in
her own firm, strong one.
"A letter—a letter to father. He's going to write to me every week. How
do you begin? He says I must write every week, same as he does."
"All right! 'My dear Father'—That's the way to begin."
By the time the "r" was reached Jack lifted a flushed face.
"It's awful hard work; I'll never do it."
"Oh, yes we will. We'll write it to-morrow in your copybook. Very soon
it will come quite easy."
And the wish to conquer made Jack comparatively patient at his writing
the following morning. Lessons over, he turned out into the paddock as
usual to play, but somehow all zest for play had deserted him. The effort to
prove himself a man the day before had a reaction. Every game, played
alone, lost its flavour. Hitherto Jack had never been conscious of the need of
a playmate. His whole being had been so absorbed in his father that the
looking forward to his visits, the saving up everything to show him and to
tell him, had satisfied him; but to-day, with that father gone, he floated
about like a rudderless boat, fretful and lonely, not able to voice his vague
longing for something to happen! He opened the gate and looked down the
lane. On the opposite side of the lane was a tenantless house; the half-acre
in which it stood had never been brought into proper cultivation as a
garden, but the flowers and shrubs which had been planted haphazard about
it had grown now into tangled confusion, and Jack, when tired of his own
premises, had often run down there, where, crawling on all-fours through
the long grass and shrubs, he had imagined himself lost in the bush, and
great was his joy when Aunt Betty, not finding him in the home paddock,
would come wandering down the lane, saying in a clear, distinct tone:
"Now where can that little boy have gone? I'm afraid, I'm dreadfully
afraid, he's lost in the bush! I wonder if it's possible he can have strayed in
here."
Then her bright head would be thrust over the gate, and each time Jack
was discovered cowering from sight there would be a fresh burst of rapture
on the part of the much-distressed aunt and roars of delighted laughter from
Jack. It was a most favourite game, but he did not wish to play it to-day.
Yet he resented it a little that a bullock-wagon was drawn to one side of

the road, the wagon piled high with furniture, which was being lifted piece
by piece into the house. His happy hunting-ground was to be his no longer,
for evidently the house was to be occupied by a fresh tenant. Dancing to
and fro with the men who were unlading the dray was a little girl, her face
entirely hidden by a large sun-bonnet, and the rest of her little person
enveloped in a blue overall, below which came a pair of sturdy brown legs,
scarcely distinguishable from the tan shoes and socks below.
Jack's resentment at the thought of losing his playground yielded to

excitement at the prospect of a playmate so close at hand, and he crept
cautiously along his side of the lane to obtain a nearer view of the new-
comer, finally taking a seat against the fence just opposite the house. It was
a minute or two before the little girl discovered him. When she did she
crossed the dividing road and stood just far enough from him to make a
quick retreat to her own premises if a nearer inspection was unfavourable. It
was almost a baby face that peered out from the bonnet: round apple
cheeks, big serious eyes, and a halo of dark curls that framed the forehead.
Her eyes met Jack's for a moment, then dropped in a sudden attack of
shyness, and she showed signs of running away without speaking.
"Wait a bit," said Jack. "Can't you tell us your name?"
The child drew a step nearer. "What's yours?" she said, answering Jack's
question by another.
"I'm Jack, father's called Jack, too."
"I'm Eva, but mummy calls me puss. Is that your place?" with a nod
towards Jack's home.
"Yes, you can come and look at it if you like," and Jack held out a
grubby hand.
Eva paused, looked up the lane and down it.
"Mummy only lets me play with nice little boys," she said.
"All right," said Jack, rising and turning back to go home. That he was
rejected on the score of not being nice enough to play with puzzled him
rather than annoyed.
There was a hasty scuttle after him as Eva ran to catch him up.
"Stop, boy! I think you's nice! You's got booful blue eyes!"
Jack turned, laughing merrily. "You're a funny little kiddie. Do you want
to come, then?"
Eva nodded gravely, thrusting a confident hand in his.
"You're old, a lot older than me," she said, admiring the agility with
which Jack climbed the top of the gate and pulled back the iron fastening to
let her through.
"I'm seven, big for my age, Aunt Betty says, but I want to be a lot bigger
before I'm done with."
"I'm six next bufday," Eva announced. "I had a bufday last week."
"Then you're six now."
Eva shook her head vigorously. "Next bufday, mummy says."
"Oh, you're only five," said Jack dejectedly. A baby of five was really too
young to play with.
"Can you play horses?"
"Yus," suddenly smiling into Jack's face.
"And cricket?"
"Kick it, a ball like this," throwing out her little foot. "Yus."
"Let's see how you run. I'll give you quite a long start, and we'll see
which can get to the house first."
Eva's stout legs acquitted themselves so well that Jack's esteem and
respect grew by leaps and bounds.
"You'll do quite well for a chum, after all," he said as he panted up to her.
"Come along and see Aunt Betty."
Aunt Betty's whereabouts were not difficult to discover. Her song rose
clear and full as a magpie as she busied herself in the dairy which adjoined
the house. The sound of Jack's voice made her turn from her milk-pans to
the doorway which framed him and his little companion.
"Why, Jack, who is the little girl?" she asked.
"Her name is Eva, and I've just settled she shall be my chum," was the
decided answer.
But Eva, frightened at finding herself quite away from her own people,
threw herself on the doorstep and hid her face in a fit of sobbing.
"I won't be nobody's chum! Take me home to mummy," she said.
Betty's arms closed round her consolingly.
"So I will directly Jack can tell me where mummy lives," said Betty.
"Come along, Jack, and show me where to take her."
CHAPTER III
NEW NEIGHBOURS
A resolute-looking little woman faced Betty as she crossed the threshold

of the door of the new neighbour. Betty carefully deposited Eva on one of
the boxes which littered the floor and explained her presence.
"It was kind of you to bring her back. Pussie has a sad trick of poking in
her nose where she's not wanted," said Eva's mother; but the child, restored
to confidence, raised indignant protest.
"Boy does want me; he wants me for a chum, mummy, and I think he's
nice! Just look at him."
Betty watched the grave little face soften into a smile as the eyes rested
first on Eva and then on Jack, who stood shyly in the doorway.
"We are neighbours, then," she said, ignoring Eva's words. She was
clearly a woman who would commit herself to no promise that she might
not be able to keep.
"My father, Mr. Treherne, owns the farm close by. Jack is his little
grandson," said Betty simply, "and I'm his only daughter."
"And my name is Kenyon. Come along, Eva; we'll leave all this alone
until after tea, and when you're in bed I must straighten things a bit," said
Mrs. Kenyon as Betty turned to go.
The voice was tired, and an English voice. The speaker, still young, for
she certainly was well under thirty, inspired Betty with the feeling that she
had had a hard fight with the world.
"Won't you come back to supper with us? I know mother will be glad to
see you, and it's hard to get things comfortable on the first night in a new
house."
"Comfortable!" echoed Mrs. Kenyon, with a note of scorn in her voice.

"It will be days before we can be that. The house has been standing empty
for a long time apparently, and needs soap and water in every corner of it. I
should like to send it to the wash, but as that can't be done I must wash it
myself, every inch of it. I took it because it was cheap!"
"Will you come, then," said Betty again.
"I beg your pardon. You'll think English manners defective, but I'm so
tired I can hardly think of what I'm saying. No, there is so much to be done
I think I will stay here, thanking you all the same for asking us." So Betty
said no more, and taking Jack's hand walked quickly down the road. Jack
chattered all the way about Eva.
"D'you think she'll be my chum, Aunt Betty?"
"We'll wait and see, Jackie, and don't be in too great a hurry. She'll want
you all the more if you don't seem too keen to have her," answered Betty,
smiling, giving the little boy his first lesson in worldly wisdom.
But the thought of the tired face haunted kind Betty as she sat down to
supper. She told her mother something of the new neighbour.
"She's such a decided, determined look and manner, mother. She's been
pretty, and she's rather pretty still, only her face has grown hard, as if she'd
had a lot of trouble. She's young to be a widow."
"What makes you think she's a widow? She did not tell you so."
"There's no sign of a man about the place; she clearly has to fend for
herself, and to English people it's hard work. They're not brought up to be
useful!"
Mrs. Treherne laughed. "She's English, then."
"Yes, she said so, and she's proud and independent; but I think when
Jack is in bed I'll risk the chance of a snub, and go and see what I can do for
her."
An hour later Betty stood again before Mrs. Kenyon's door. From the
inner room came a sound of singing, and through the half-opened door
Betty caught a glimpse of a little bed that stood in the corner, over which
Mrs. Kenyon bent tenderly soothing Eva to sleep with her soft lullaby.
"She has one tender spot in her heart, anyway," thought Betty, giving a
little cough to proclaim her presence. Mrs. Kenyon turned and came toward
her on tip-toe, drawing the door of her bedroom gently to behind her.
"Eva was excited and would not go to sleep. I don't generally spoil her
like that, but she's off now as sound as a top."
"I've come to help you for an hour or two if you will have me."
Mrs. Kenyon's bright eyes scanned Betty from head to foot.
"It's not everyone that I could accept help from, but I'll be glad of it from
you."
So the two worked side by side with a will and with scarcely a word
exchanged between them. They shifted boxes, placed furniture in temporary
safety against the walls, but to Betty fell the lion's share of the lifting.

Download Topicalization and Stress Clash Avoidance in the History of English 1st Edition Augustin Speyer ebook All Chapters PDF

Uploaded by

Copyright:

Available Formats

Download Topicalization and Stress Clash Avoidance in the History of English 1st Edition Augustin Speyer ebook All Chapters PDF

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Download Topicalization and Stress Clash Avoidance in the History of English 1st Edition Augustin Speyer ebook All Chapters PDF

Uploaded by

Copyright:

Available Formats

Download the full version of the ebook at ebookfinal.

Topicalization and Stress Clash Avoidance in the

Download more ebook instantly today at https://ebookfinal.com

Topicalization and Stress Clash Avoidance in the History

The Cambridge History of English Romantic Literature The

Studies in the History of the English Language V Variation

Studies in the History of the English Language V Variation

Yellowstone National Park 1st Edition Byron Augustin

Complications in Cardiothoracic Surgery Avoidance and

Ethnocentrism and the English Dictionary Routledge Studies

The Handbook of the History of English Ans Van Kemenade

Library of Congress Cataloging-in-Publication Data

Bibliographic information published by the Deutsche Nationalbibliothek

This book is a modified and slightly extended version of my PhD thesis

Preliminary versions of parts of this study were read at various confer-

Frankfurt / Main, January 2010 Augustin Speyer

2. Topicalization in Middle and Modern English – 24

3. The Clash Avoidance Requirement in Modern English 80

4. Phonological Aspects of the Clash Avoidance Require- 132

5. Topicalization and the Clash Avoidance Requirement in 177

6. Concluding remarks 234

Appendix: All Old English OSV-sentences with full noun 237

The main concern of this study is to demonstrate how a general phonologi-

second chapter how the Clash Avoidance Requirement influenced syntactic

1.2. Some background

As it should be useful to give some preliminary definitions of notions and

1.2.1. Pragmatic dimensions

One does not need to be a functionalist to recognize that in a number of

– newness: old versus new information,

(1) Rudolf Bupfinger, inspector of the state’s criminal investigation unit,

There are several intermediate stages to the old-new-distinction, either to be

medic, so the mention of a paramedic is in some ways premediated by the

formation, and that topics tend to be realized by predictable syntactic

For English, focus is associated with prominence on the focalized element,

able, which as a rule coincides with what is known as instances of ‘narrow

Although these four pragmatic dimensions are independent of each other,

Heim’s (1982) famous metaphor.4 It is better, if one wants to emphasize

1.2.2. Modularity of Grammar

I assume a modular model of grammar in the tradition of Chomsky (1995,

In contrast to Chomsky (1995), but in accordance with many other genera-

ferred to by Ladd as sentence stress; Newman (1946: 176) calls it nucleus,

1. highest prominence in the clause, or

1.2.4. Grid construction

The theory of grid construction used in this study is based on Metrical

ily in detail, to the levels of word, phonological phrase and intonational

Iamb Construction Rule:

This kind of representation takes account of two requirements on grid pro-

Eurhythmy, however, is a special effect of a much more basic requirement

1.2.5. The syntactic field model

Vorfeld – linke Satzklammer – Mittelfeld – rechte Satzkl. – Nachfeld

1.3. Further concepts

1.3.1. Verb second

1.3.2. The reconstruction of sentence prosody

Much of the discussion to follow hinges on the assignment of focus (and in

Reading 2: ‘Mary said to me that only the money belonging to her

The second part of this study is devoted to the exploration of an empirical

2.1. The decline of topicalization

Topicalization in the non-technical sense of moving a constituent other than

different from S – V – O. The surface string scheme of topicalization in

where S = subject, V = finite verb and X = any constituent. In Old and

(14) scene-setting: () (. )

(15) anaphoric: () (. )