Download the full version of the ebook at

https://ebookfinal.com

Computability complexity and languages

fundamentals of theoretical computer science
2. ed., transferred to digital print Edition
Davis
https://ebookfinal.com/download/computability-
complexity-and-languages-fundamentals-of-
theoretical-computer-science-2-ed-transferred-to-
digital-print-edition-davis/

Explore and download more ebook at https://ebookfinal.com

 Recommended digital products (PDF, EPUB, MOBI) that
you can download immediately if you are interested.

Physics and Theoretical Computer Science From Numbers and

Languages to Quantum Cryptography J.-P. Gazeau

https://ebookfinal.com/download/physics-and-theoretical-computer-
science-from-numbers-and-languages-to-quantum-cryptography-j-p-gazeau/

ebookfinal.com

Languages and Machines An Introduction to the Theory of

Computer Science 3rd Edition Thomas A. Sudkamp

https://ebookfinal.com/download/languages-and-machines-an-
introduction-to-the-theory-of-computer-science-3rd-edition-thomas-a-
sudkamp/
ebookfinal.com

Fundamentals of Computer Science using Java 1st Edition

David Hughes

https://ebookfinal.com/download/fundamentals-of-computer-science-
using-java-1st-edition-david-hughes/

ebookfinal.com

Wheat flour milling 2. ed., 2. print Edition Hibbs

https://ebookfinal.com/download/wheat-flour-milling-2-ed-2-print-
edition-hibbs/

ebookfinal.com
Universal Algebra and Applications in Theoretical Computer
Science 1st Edition Klaus Denecke (Author)

https://ebookfinal.com/download/universal-algebra-and-applications-in-
theoretical-computer-science-1st-edition-klaus-denecke-author/

ebookfinal.com

Digital evidence and computer crime forensic science

computers and the Internet 2nd ed Edition Eoghan Casey Bs
Ma
https://ebookfinal.com/download/digital-evidence-and-computer-crime-
forensic-science-computers-and-the-internet-2nd-ed-edition-eoghan-
casey-bs-ma/
ebookfinal.com

The Photoshop Darkroom 2 Creative Digital Transformations

1st Edition Harold Davis

https://ebookfinal.com/download/the-photoshop-darkroom-2-creative-
digital-transformations-1st-edition-harold-davis/

ebookfinal.com

Computability and logic 4th ed Edition George S. Boolos

https://ebookfinal.com/download/computability-and-logic-4th-ed-
edition-george-s-boolos/

ebookfinal.com

Comprehensive Coordination Chemistry II From Biology to

Nanotechnology 2 ed in 10 Vol Set Volume 02 Fundamentals
Physical Methods Theoretical Analysis and Case Studies 2nd
Edition Mccleverty J.A. (Ed.)
https://ebookfinal.com/download/comprehensive-coordination-chemistry-
ii-from-biology-to-nanotechnology-2-ed-in-10-vol-set-
volume-02-fundamentals-physical-methods-theoretical-analysis-and-case-
studies-2nd-edition-mccleverty-j-a-ed/
ebookfinal.com
Computability complexity and languages fundamentals of
theoretical computer science 2. ed., transferred to digital
print Edition Davis Digital Instant Download
Author(s): Davis, Martin;Sigal, Ron;Weyuker, Elaine J
ISBN(s): 9780122063824, 0122063821
Edition: 2. ed., transferred to digital print
File Details: PDF, 48.81 MB
Year: 2003
Language: english
Second Edition
Computability,
Complexity, and
Languages
Fundamentals of
Theoretical Computer Science
This is a volume in
COMPUTER SCIENCE AND SCIENTIFIC COMPUTING
A Series of Monographs and Textbooks
Editor: Werner Rheinboldt
A complete list of titles in this series is available from the publisher upon request.
Second Edition
Computability,
Complexity, and
Languages
Fundamentals of
Theoretical Computer Science
Martin D. Davis
Department of Computer Science
Courant Institute of Mathematical Sciences
New York University
New York, New York

Ron Sigal
Departments of Mathematics and Computer Science
Yale University
New Haven, Connecticut

Elaine J. Weyuker
Department of Computer Science
Courant Institute of Mathematical Sciences
New York University
New York, New York

Morgan Kaufmann is an imprint of Academic Press

A Harcourt Science and Technology Company

San Diego San Francisco New York Boston

London Sydney Tokyo
This book is printed on acid free paper

Copyright © 1994, 1983 Elsevier Science (USA)

All Rights Reserved.

No part of this publication may be reproduced or transmitted in any form or by any
means, electronic or mechanical, including photocopy, recording, or any information
storage and retrieval system, without permission in writing from the publisher.

Requests for permission to make copies of any part of the work should be mailed to:
Permissions Department, Harcourt Brace & Company, 6277 Sea Harbor Drive,
Orlando, Florida 32887-6777

Morgan Kaufmann Publishers

340 Pine Street, Sixth Floor, San Francisco, California 94104-3205
http://www.mkp.com
Transferred to Digital Printing 2003
Academic Press
An Elsevier Science Imprint
525 B Street, Suite 1900, San Diego, California 92101-4495, U.S.A.
http://www.academicpress.com

Academic Press
84 Theobalds Road, London VVC1X 8RR, UK
http://www.academicpress.com

Library of Congress Cataloging-in-Publication Data

Davis, Martin 1928
Computability, complexity, and languages: fundamentals of
Theoretical computer science / Martin D. Davis, Ron Sigal,
Elaine J. Weyuker. -2nded.
p. cm. -(Computer science and applied mathematics)
Includes bibliographical references and index.
ISBN 0-12-206382-1
1. Machine theory. 2. Computational complexity. 3. Formal
Languages. I. Sigal, Ron. II. Weyuker, Elaine J. III. Title.
IV. Series.
QA267.D38 1994
511.3-dc20 93-26807
CIP
PRINTED IN THE UNITED STATES OF AMERICA
02 03 04 05 06 SB 9 8 7 6
 To the memory of Helen and Harry Davis
and to
Hannah and Herman Sigal
Sylvia and Marx Weyuker

Virginia Davis, Dana Latch, Thomas Ostrand

and to
Rachel Weyuker Ostrand
This page intentionally left blank
 Contents

Preface xiii
Acknowledgments xvii
Dependency Graph xix

1 Preliminaries 1
1. Sets and ^-tuples 1
2. Functions 3
3. Alphabets and Strings 4
4. Predicates 5
5. Quantifiers 6
6. Proof by Contradiction 8
7. Mathematical Induction 9

Part 1 Computability 15
2 Programs and Computable Functions 17
1. A Programming Language 17
2. Some Examples of Programs 18
3. Syntax 25
4. Computable Functions 28
5. More about Macros 32
VII
viii

3 Primitive Recursive Functions 39

1. Composition 39
2. Recursion 40
3. PRC Classes 42
4. Some Primitive Recursive Functions 44
5. Primitive Recursive Predicates 49
6. Iterated Operations and Bounded Quantifiers 52
7. Minimalization 55
8. Pairing Functions and Gödel Numbers 59
4 A Universal Program 65
1. Coding Programs by Numbers 65
2. The Halting Problem 68
3. Universality 70
4. Recursively Enumerable Sets 78
5. The Parameter Theorem 85
6. Diagonalization and Reducibility 88
7. Rice's Theorem 95
*8. The Recursion Theorem 97
*9. A Computable Function That Is Not Primitive Recursive 105
5 Calculations on Strings 113
1. Numerical Representation of Strings 113
2. A Programming Language for String Computations 121
3. The Languages «y and <9Jn 126
4. Post-Turing Programs 129
5. Simulation of ò^n in t? 135
6. Simulation of F in J/y 140
6 Turing Machines 145
1. Internal States 145
2. A Universal Turing Machine 152
3. The Languages Accepted by Turing Machines 153
4. The Halting Problem for Turing Machines 157
5. Nondeterministic Turing Machines 159
6. Variations on the Turing Machine Theme 162
7 Processes and Grammars 169
1. Semi-Thue Processes 169
2. Simulation of Nondeterministic Turing Machines by
Semi-Thue Processes 171
Contents ix

3. Unsolvable Word Problems 176

4. Post's Correspondence Problem 181
5. Grammars 186
6. Some Unsolvable Problems Concerning Grammars 191
*7. Normal Processes 192

8 Classifying Unsolvable Problems 197

1. Using Oracles 197
2. Relativization of Universality 201
3. Reducibility 207
4. Sets r.e. Relative to an Oracle 211
5. The Arithmetic Hierarchy 215
6. Post's Theorem 217
7. Classifying Some Unsolvable Problems 224
8. Rice's Theorem Revisited 230
9. Recursive Permutations 231

Part 2 Grammars and Automata 235

9 Regular Languages 237

1. Finite Automata 237
2. Nondeterministic Finite Automata 242
3. Additional Examples 247
4. Closure Properties 249
5. Kleene's Theorem 253
6. The Pumping Lemma and Its Applications 260
7. The Myhill-Nerode Theorem 263

10 Context-Free Languages 269

1. Context-Free Grammars and Their Derivation Trees 269
2. Regular Grammars 280
3. Chomsky Normal Form 285
4. Bar-Hillel's Pumping Lemma 287
5. Closure Properties 291
*6. Solvable and Unsolvable Problems 297
7. Bracket Languages 301
8. Pushdown Automata 308
9. Compilers and Formal Languages 323
X Contents

11 Context-Sensitive Languages 327

1. The Chomsky Hierarchy 327
2. Linear Bounded Automata 330
3. Closure Properties 337

Part 3 Logic 345

12 Propositional Calculus 347
1. Formulas and Assignments 347
2. Tautological Inference 352
3. Normal Forms 353
4. The Davis-Putnam Rules 360
5. Minimal Unsatisfiability and Subsumption 366
6. Resolution 367
7. The Compactness Theorem 370

13 Quantification Theory 375

1. The Language of Predicate Logic 375
2. Semantics 377
3. Logical Consequence 382
4. Herbrand's Theorem 388
5. Unification 399
6. Compactness and Countability 404
*7. Gödel's Incompleteness Theorem 407
*8. Unsolvability of the Satisfiability Problem in Predicate Logic 410

Part 4 Complexity AM
14 Abstract Complexity 419
1. The Blum Axioms 419
2. The Gap Theorem 425
3. Preliminary Form of the Speedup Theorem 428
4. The Speedup Theorem Concluded 435

15 Polynomial-Time Computability 439

1. Rates of Growth 439
2. P versus NP 443
3. Cook's Theorem 451
4. Other NP-Complete Problems 457
Contents xi

Part 5 Semantics 465

16 Approximation Orderings 467
1. Programming Language Semantics 467
2. Partial Orders 472
3. Complete Partial Orders 475
4. Continuous Functions 486
5. Fixed Points 494
17 Denotational Semantics of Recursion Equations 505
1. Syntax 505
2. Semantics of Terms 511
3. Solutions to W-Programs 520
4. Denotational Semantics of W-Programs 530
5. Simple Data Structure Systems 539
6. Infinitary Data Structure Systems 544
18 Operational Semantics of Recursion Equations 557
1. Operational Semantics for Simple Data Structure Systems 557
2. Computable Functions 575
3. Operational Semantics for Infinitary Data Structure Systems 584

Suggestions for Further Reading 593

Notation Index 595
Index 599
This page intentionally left blank
 Preface

Theoretical computer science is the mathematical study of models of

computation. As such, it originated in the 1930s, well before the existence
of modern computers, in the work of the logicians Church, Gödel, Kleene,
Post, and Turing. This early work has had a profound influence on the
practical and theoretical development of computer science. Not only has
the Turing machine model proved basic for theory, but the work of these
pioneers presaged many aspects of computational practice that are now
commonplace and whose intellectual antecedents are typically unknown to
users. Included among these are the existence in principle of all-purpose
(or universal) digital computers, the concept of a program as a list of
instructions in a formal language, the possibility of interpretive programs,
the duality between software and hardware, and the representation of
languages by formal structures, based on productions. While the spotlight
in computer science has tended to fall on the truly breathtaking technolog­
ical advances that have been taking place, important work in the founda­
tions of the subject has continued as well. It is our purpose in writing this
book to provide an introduction to the various aspects of theoretical
computer science for undergraduate and graduate students that is suffi­
ciently comprehensive that the professional literature of treatises and
research papers will become accessible to our readers.
We are dealing with a very young field that is still finding itself.
Computer scientists have by no means been unanimous in judging which
XIII
XIV Preface

parts of the subject will turn out to have enduring significance. In this
situation, fraught with peril for authors, we have attempted to select topics
that have already achieved a polished classic form, and that we believe will
play an important role in future research.
In this second edition, we have included new material on the subject of
programming language semantics, which we believe to be established as an
important topic in theoretical computer science. Some of the material on
computability theory that had been scattered in the first edition has been
brought together, and a few topics that were deemed to be of only
peripheral interest to our intended audience have been eliminated. Nu­
merous exercises have also been added. We were particularly pleased to be
able to include the answer to a question that had to be listed as open in
the first edition. Namely, we present Neil Immerman's surprisingly
straightforward proof of the fact that the class of languages accepted by
linear bounded automata is closed under complementation.
We have assumed that many of our readers will have had little experi­
ence with mathematical proof, but that almost all of them have had
substantial programming experience. Thus the first chapter contains an
introduction to the use of proofs in mathematics in addition to the usual
explanation of terminology and notation. We then proceed to take advan­
tage of the reader's background by developing computability theory in the
context of an extremely simple abstract programming language. By system­
atic use of a macro expansion technique, the surprising power of the
language is demonstrated. This culminates in a universal program, which is
written in all detail on a single page. By a series of simulations, we then
obtain the equivalence of various different formulations of computability,
including Turing's. Our point of view with respect to these simulations is
that it should not be the reader's responsibility, at this stage, to fill in the
details of vaguely sketched arguments, but rather that it is our responsibil­
ity as authors to arrange matters so that the simulations can be exhibited
simply, clearly, and completely.
This material, in various preliminary forms, has been used with under­
graduate and graduate students at New York University, Brooklyn College,
The Scuola Matematica Interuniversitaria-Perugia, The University of Cal­
ifornia-Berkeley, The University of California-Santa Barbara, Worcester
Polytechnic Institute, and Yale University.
Although it has been our practice to cover the material from the second
part of the book on formal languages after the first part, the chapters on
regular and on context-free languages can be read immediately after
Chapter 1. The Chomsky-Schützenberger representation theorem for con­
text-free languages in used to develop their relation to pushdown au­
tomata in a way that we believe is clarifying. Part 3 is an exposition of the
aspects of logic that we think are important for computer science and can
Preface xv

also be read immediately following Chapter 1. Each of the chapters of Part

4 introduces an important theory of computational complexity, concluding
with the theory of NP-completeness. Part 5, which is new to the second
edition, uses recursion equations to expand upon the notion of computabil-
ity developed in Part 1, with an emphasis on the techniques of formal
semantics, both denotational and operational. Rooted in the early work of
Gödel, Herbrand, Kleene, and others, Part 5 introduces ideas from the
modern fields of functional programming languages, denotational seman­
tics, and term rewriting systems.
Because many of the chapters are independent of one another, this book
can be used in various ways. There is more than enough material for a
full-year course at the graduate level on theory of computation. We have
used the unstarred sections of Chapters 1-6 and Chapter 9 in a successful
one-semester junior-level course, Introduction to Theory of Computation,
at New York University. A course on finite automata and formal languages
could be based on Chapters 1, 9, and 10. A semester or quarter course on
logic for computer scientists could be based on selections from Parts 1 and
3. Part 5 could be used for a third semester on the theory of computation
or an introduction to programming language semantics. Many other ar­
rangements and courses are possible, as should be apparent from the
dependency graph, which follows the Acknowledgments. It is our hope,
however, that this book will help readers to see theoretical computer
science not as a fragmented list of discrete topics, but rather as a unified
subject drawing on powerful mathematical methods and on intuitions
derived from experience with computing technology to give valuable in­
sights into a vital new area of human knowledge.

Note to the Reader

Many readers will wish to begin with Chapter 2, using the material of
Chapter 1 for reference as required. Readers who enjoy skipping around
will find the dependency graph useful.
Sections marked with an asterisk (*) may be skipped without loss of
continuity. The relationship of these sections to later material is given in
the dependency graph.
Exercises marked with an asterisk either introduce new material, refer
to earlier material in ways not indicated in the dependency graph, or
simply are considered more difficult than unmarked exercises.
A reference to Theorem 8.1 is to Theorem 8.1 of the chapter in which
the reference is made. When a reference is to a theorem in another
chapter, the chapter is specified. The same system is used in referring to
numbered formulas and to exercises.
This page intentionally left blank
 Acknowledgments

It is a pleasure to acknowledge the help we have received. Charlene

Herring, Debbie Herring, Barry Jacobs, and Joseph Miller made their
student classroom notes available to us. James Cox, Keith Harrow, Steve
Henkind, Karen Lemone, Colm O'Dunlaing, and James Robinett provided
helpful comments and corrections. Stewart Weiss was kind enough to
redraw one of the figures. Thomas Ostrand, Norman Shulman, Louis
Salkind, Ron Sigal, Patricia Teller, and Elia Weixelbaum were particularly
generous with their time, and devoted many hours to helping us. We are
especially grateful to them.

Acknowledgments to Corrected Printing

We have taken this opportunity to correct a number of errors. We are
grateful to the readers who have called our attention to errors and who
have suggested corrections. The following have been particularly helpful:
Alissa Bernholc, Domenico Cantone, John R. Cowles, Herbert Enderton,
Phyllis Franki, Fred Green, Warren Hirsch, J. D. Monk, Steve Rozen, and
Stewart Weiss.

xvii
XVIII Acknowledgments

Acknowledgments to Second Edition

Yuri Gurevich, Paliath Narendran, Robert Paige, Carl Smith, and particu­
larly Robert McNaughton made numerous suggestions for improving the
first edition. Kung Chen, William Hurwood, Dana Latch, Sidd Puri,
Benjamin Russell, Jason Smith, Jean Toal, and Niping Wu read a prelimi­
nary version of Part 5.

Acknowledgments to Reprint of Second Edition

We are grateful to the following people for their careful reading of the
Second Edition: John Case, P. Klingsberg, Ken Klein, Eugenio Omodeo,
David Schedler, John David Stone, and Lenore Zuck.
 Dependency Graph
Chapter 1
Preliminaries

Chapter 9 Chapter 2 Chapter 12

Regular Languages Programs and Proportional Calculus
Computable Functions

Chapter 10 Chapter 3
Context-Free Languages Primitive
Recursive Functions

Chapter 16 Chapter 4 Chapter 8

Approximation A Universal Program Classifying Unsorvable
Ordering« Problems

Chapter 5
Chapter 17 Calculations on Chapter 14
Denotations Semantics Strings Abstract Complexity
of Recursion Equations

Chapter 6 Chapter 15
Chapter 18 Turing Machines Polynomial-Time
Operational Semantics Computabilrty
of Recursion Equations

Chapter 7
Processes and Grammars

Chapter 11 Chapter 13
Context-Sensitive Quantification Theory
Languages

A solid line between two chapters indicates the dependence of the un-
starred sections of the higher numbered chapter on the unstarred sections
of the lower numbered chapter. An asterisk next to a solid line indicates
that knowledge of the starred sections of the lower numbered chapter is
also assumed. A dotted line shows that knowledge of the unstarred
sections of the lower numbered chapter is assumed for the starred sections
of the higher numbered chapter.
xix
This page intentionally left blank
 1
Preliminaries

1. Sets and n-tuples

We shall often be dealing with sets of objects of some definite kind.
Thinking of a collection of entities as a set simply amounts to a decision to
regard the whole collection as a single object. We shall use the word class
as synonymous with set. In particular we write N for the set of natural
numbers 0,1,2,3, In this book the word number will always mean
natural number except in contexts where the contrary is explicitly stated.
We write
a <= S
to mean that a belongs to S or, equivalently, is a member of the set S, and

a £ S
to mean that a does not belong to S. It is useful to speak of the empty set,
written 0 , which has no members. The equation R = 5, where R and S
are sets, means that R and S are identical as sets, that is, that they have
exactly the same members. We write R ç S and speak of R as a subset of
S to mean that every element of R is also an element of S. Thus, R = S if
and only if R ç S and S ç R. Note also that for any set /?, 0 ç R and
R ç R. We write R (Z S to indicate that R ç S but /? * S. In this case fl
1
2 Chapter 1 Preliminaries

is called a proper subset of S. If R and S are sets, we write R U S for the

wwwi of R and 5, which is the collection of all objects which are members
of either R or S or both. R C\ S, the intersection of R and 5, is the set of
all objects that belong to both R and S. R - S, the set of all objects that
belong to R and do not belong to S, is the difference between R and S. S
may contain objects not in R. Thus R — S = R — (R n S). Often we will
be working in contexts where all sets being considered are subsets of some
fixed set D (sometimes called a domain or a universe). In such a case we
write S for D — 5, and call S the complement of 5. Most frequently we
shall be writing S for TV - S. The De Morgan identities

RU S = RnS,

~R~n~S = RuS
are very useful; they are easy to check and any reader not already familiar
with them should do so. We write
{ax ,<3 2 ,...,#„}
for the set consisting of the n objects ax, a 2 , . . . , an. Sets that can be
written in this form as well as the empty set are called finite. Sets that are
not finite, e.g., N, are called infinite. It should be carefully noted that a
and {a} are not the same thing. In particular, a <E S is true if and only if
{a} c S. Since two sets are equal if and only if they have the same
members, it follows that, for example, {a, ò, c] = {a, c, b) = {b, a, c}. That
is, the order in which we may choose to write the members of a set is
irrelevant. Where order is important, we speak instead of an rc-tuple or a
list. We write n-tuples using parentheses rather than curly braces:
(ax ,...,a„).

Naturally, the elements making up an /7-tuple need not be distinct. Thus

(4,1,4,2) is a 4-tuple. A 2-tuple is called an ordered pair, and a 3-tuple is
called an ordered triple. Unlike the case for sets of one object, we do not
distinguish between the object a and the l-tuple (a). The crucial property of
Ai-tuples is
(a{ ,a29...,an) = (b{ ,b2,...9bn)

if and only if

ax = bx , a2 = b2, -.., and an = bn .

If 5 , , S2,..., Sn are given sets, then we write 5, X S2 X ••• X Sn for the

set of all Ai-tuples (ax, a2,.. -, an) such that ax G Sx,a2 G S2,..., an ΠSn .
2. Functions 3

Sx X S2 X ••• X Sn is sometimes called the Cartesian product of

S{, S 2 , . . . , Sn . In case 5j = S2 = ••• = 5„ = 5 we write 5" for the Carte­
sian product ^ x S 2 X - x 5 r

2. Functions
Functions play an important role in virtually every branch of pure and
applied mathematics. We may define a function simply as a set / , all of
whose members are ordered pairs and that has the special property
(a, b) e / and (a,c) e/ implies b = c.
However, intuitively it is more helpful to think of the pairs listed as the
rows of a table. For / a function, one writes f{a) = b to mean that
(a, b) e / ; the definition of function ensures that for each a there can be
at most one such b. The set of all a such that (a, b) e / for some ò is
called the domain of / . The set of all f(a) for a in the domain of / is
called the range of / .
As an example, let / be the set of ordered pairs (n,n2) for n e N.
Then, for each n ^ N, f(n) = n2. The domain of / is N. The range of / is
the set of perfect squares.
Functions / are often specified by algorithms that provide procedures
for obtaining f(a) from a. This method of specifying functions is particu­
larly important in computer science. However, as we shall see in Chapter
4, it is quite possible to possess an algorithm that specifies a function
without being able to tell which elements belong to its domain. This makes
the notion of a so-called partial function play a central role in computabil-
ity theory. A partial function on a set S is simply a function whose domain
is a subset of S. An example of a partial function on TV is given by g(n)
= yfn, where the domain of g is the set of perfect squares. If / is a partial
function on S and a e S, then we write f(a)i and say that f{a) is defined
to indicate that a is in the domain of / ; if a is not in the domain of / , we
write f(a)î and say that f(a) is undefined. If a partial function on S has
the domain 5, then it is called total. Finally, we should mention that the
empty set 0 is itself a function. Considered as a partial function on some
set S, it is nowhere defined.
For a partial function / o n a Cartesian product Sx X S2 X ••• X Sn, we
write f(ax,.. .,an) rather than f((a],..., an)). A partial function / o n a
set Sn is called an n-ary partial function on S, or a function of n variables
on S. We use unary and binary for 1-ary and 2-ary, respectively. For /i-ary
partial functions, we often write f(xl9...,xn) instead of / as a way of
showing explicitly that / is n-ary.
4 Chapter 1 Preliminaries

Sometimes it is useful to work with particular kinds of functions. A

function / is one-one if, for all Jt, y in the domain of / , fix) = fiy)
implies x = y. Stated differently, if x # y then f(x) # fiy). If the range of
/ is the set S, then we say that / is an onto function with respect to 5, or
simply that / is onto S. For example, fin) = n2 is one-one, and / is onto
the set of perfect squares, but it is not onto N.
We will sometimes refer to the idea of closure. If S is a set and / is a
partial function on 5, then S is closed under f if the range of / is a subset
of S. For example, N is closed under fin) = n2, but it is not closed under
hin) = 4n (where h is a total function on N).

3. Alphabets and Strings

An alphabet is simply some finite nonempty set A of objects called
symbols. An n-tuple of symbols of A is called a word or a string on A.
Instead of writing a word as iax, a2J..., an) we write simply a ^ ••• an . If
w = 0 ^ ••• ÖW, then we say that n is the length of u and write \u\ = AZ.
We allow a unique null word, written 0, of length 0. (The reason for using
the same symbol for the number zero and the null word will become clear
in Chapter 5.) The set of all words on the alphabet A is written A*. Any
subset of A* is called a language on A or a language with alphabet A. We
do not distinguish between a symbol a e A and the word of length 1
consisting of that symbol. If u,v ^A*, then we write uv for the word
obtained by placing the string v after the string u. For example, if
A = {a, b, c}, u = bab, and u = caa, then

uv = babcaa and vu = caabab.

Where no confusion can result, we write uv instead of u v. It is obvious
that, for all u,
uO = Ou = u,
and that, for all u, L\ w,
uivw) = iuv)w.
Also, if either uv = uw or vu = wu, then v = w.
If w is a string, and n e N, n > 0, we write

w1"1 = uu -- u .
n
We also write w[()] = 0. We use the square brackets to avoid confusion with
numerical exponentiation.
4. Predicates 5

If w E /!*, we write uR for u written backward; i.e., if u = a{a2 ••• a„,

for a},...,an ^ A, then w* = an ••• 0 2 ^i ■ Clearly, 0* = 0 and (WLO* =
vRuR for M,^ e A*.

4. Predicates
By a predicate or a Boolean-valued function on a set 5 we mean a tota/
function P on S such that for each A E 5 , either

P U ) = TRUE or P(a) = FALSE,

where TRUE and FALSE are a pair of distinct objects called truth values.
We often say P(a) is true for P(a) = TRUE, and P(a) is false for
P(a) = FALSE. For our purposes it is useful to identify the truth values
with specific numbers, so we set

TRUE = 1 and FALSE = 0.

Thus, a predicate is a special kind of function with values in N. Predicates
on a set S are usually specified by expressions which become statements,
either true or false, when variables in the expression are replaced by
symbols designating fixed elements of S. Thus the expression

x <5
specifies a predicate on N, namely,

p(x) = M if x = 0,1,2,3,4
\0 otherwise.

Three basic operations on truth values are defined by the tables in Table
4.1. Thus if P and Q are predicates on a set 5, there are also the
predicates ~ P , P & Q, P V Q. ~P is true just when P is false; P & Q is
true when both P and Q are true, otherwise it is false; P V Q is true when
either P or Q or both are true, otherwise it is false. Given a predicate P

Table 4.1

p -p p q p &q p Vq

0 1 1 1 1 1
1 0 0 1 0 1
1 0 0 1
0 0 0 0
6 Chapter 1 Preliminaries

on a set S, there is a corresponding subset R of S, namely, the set of all

elements a G S for which P(a) = 1. We write

R = {a £ES\P(a)}.
Conversely, given a subset R of a given set S, the expression
x <E /?
defines a predicate on 5, namely, the predicate defined by

1 if I E / Î
[0 if x <£ R.

Of course, in this case,

/? = {x<=S\P(x)}.

The predicate P is called the characteristic function of the set R. The close
connection between sets and predicates is such that one can readily
translate back and forth between discourse involving one of these notions
and discourse involving the other. Thus we have

{x G S I P(x) & Q(x)} = {x G S I P(x)} n {* G S I ß(jc)},

{x G 5 I />(*) V ß ( x ) } = {x G S | P U ) } U U G 5 | ß(jc)},
U G 5 I - P U ) } = S - {x G 5 | P(x)}.
To indicate that two expressions containing variables define the same
predicate we place the symbol <=> between them. Thus,
X < 5 ^ X = 0VJC=1 VX = 2VX = 3VX = 4.

The De Morgan identities from Section 1 can be expressed as follows in

terms of predicates on a set S:

P(x)& Q(x) « ~ ( ~ P U ) v ~ ß ( x ) ) ,
P(x) V Q(x) ~~(~P(x)& ~ß(jc)).

5. Quantifiers
In this section we will be concerned exclusively with predicates on Nm (or
what is the same thing, ra-ary predicates on N) for different values of m.
Here and later we omit the phrase "on TV" when the meaning is clear.
5. Quantifiers 7

Thus, let Pit, xx , . . . , xn) be an in + l)-ary predicate. Consider the predi­

cate Qiy, xx,..., xn) defined by

Q(y,xx,...9xn) **P(0,xl,...Jxn) VP(1, *,,...,*„)

V •■• V P i y , ^ x„).

Thus the predicate Qiy, xx,..., xn) is true just in case there is a value of
t < y such that Pit, j t , , . . . , *„) is true. We write this predicate Q as

The expression "(3/)< v " is called a bounded existential quantifier. Similarly,

we write (V/)< yPit, x{,..., xn) for the predicate

P(0,JC1,...,XJ& P O , ^ , . . . , *„)&••■& P(y,xl9...,xn).

This predicate is true just in case Pit, xx,..., xn) is true for all t < y.
The expression "(Vf)< v " is called a bounded universal quantifier. We also
write (3/),. yPit, xx,..., xn) for the predicate that is true just in
case Pit, xx,..., xn) is true for at least one value of t < y and
(V/)< y Pit, xx,..., xn) for the predicate that is true just in case
Pit, xx,..., xn) is true for all values of / < y.
We write

Q(xx,...,xn) « (3t)P(t,xl9...,xn)

for the predicate which is true if there exists some t e N for which
Pit, xx, . . . , xn) is true. Similarly, iVt)Pit, xx, . . . , xn) is true if
Pit, xx,..., xn) is true for all t G TV.
The following generalized De Morgan identities are sometimes useful:

~i3t)^yPit,xx,...,xn) « (Vi)*, ~P(t,xXJ...,xn),

~i3t)Pit,xx,...,xn) « (Vi) ~PU,xX9...,xn).

The reader may easily verify the following examples:

i3y)ix + y = 4) <=> x < 4,

i3y)ix + y = 4) « ( 3 y ) < 4 U + y = 4),
(VyXxy = 0) « j e = 0,
i3y)<zix + y = 4) ^> ix + z > 4& x < 4).
8 Chapter 1 Preliminaries

6. Proof by Contradiction
In this book we will be calling many of the assertions we make theorems
(or corollaries or lemmas) and providing proofs that they are correct. Why
are proofs necessary? The following example should help in answering this
question.
Recall that a number is called a prime if it has exactly two distinct
divisors, itself and 1. Thus 2, 17, and 41 are primes, but 0, 1,4, and 15 are
not. Consider the following assertion:
n2 — n + 41 is prime for all n e N'.
This assertion is in fact false. Namely, for n = 41 the expression becomes
41 2 - 41 + 41 = 41 2 ,
which is certainly not a prime. However, the assertion is true (readers with
access to a computer can easily check this!) for all n < 40. This example
shows that inferring a result about all members of an infinite set (such as
AO from even a large finite number of instances can be very dangerous. A
proof is intended to overcome this obstacle.
A proof begins with some initial statements and uses logical reasoning to
infer additional statements. (In Chapters 12 and 13 we shall see how the
notion of logical reasoning can be made precise; but in fact, our use of
logical reasoning will be in an informal intuitive style.) When the initial
statements with which a proof begins are already accepted as correct, then
any of the additional statements inferred can also be accepted as correct.
But proofs often cannot be carried out in this simple-minded pattern. In
this and the next section we will discuss more complex proof patterns.
In a proof by contradiction, one begins by supposing that the assertion
we wish to prove is false. Then we can feel free to use the negation of what
we are trying to prove as one of the initial statements in constructing a
proof. In a proof by contradiction we look for a pair of statements
developed in the course of the proof which contradict one another. Since
both cannot be true, we have to conclude that our original supposition was
wrong and therefore that our desired conclusion is correct.
We give two examples here of proof by contradiction. There will be
many in the course of the book. Our first example is quite famous. We
recall that every number is either even (i.e., = In for some n ΠN) or odd
(i.e., = 2n + 1 for some n G N). Moreover, if m is even, m = 2n, then
m2 = An2 = 2 • 2n2 is even, while if m is odd, m = 2n + 1, then m2 =
An2 + An + 1 = 2(2n2 + 2n) + 1 is odd. We wish to prove that the
equation
2 = (m/n)2 (6.1)
7. Mathematical Induction 9

has no solution for m,n G N (that is, that y/2 is not a "rational" number).
We suppose that our equation has a solution and proceed to derive a
contradiction. Given our supposition that (6.1) has a solution, it must have
a solution in which m and n are not both even numbers. This is true
because if m and n are both even, we can repeatedly "cancel" 2 from
numerator and denominator until at least one of them is odd. On the
other hand, we shall prove that for every solution of (6.1) m and n must
both be even. The contradiction will show that our supposition was false,
i.e., that (6.1) has no solution.
It remains to show that in every solution of (6.1), m and n are both
even. We can rewrite (6.1) as
m2 = 2n 2 ,
which shows that m2 is even. As we saw above this implies that m is even,
say m = 2k. Thus, m2 = 4k2 = In2, or n2 = 2k2. Thus, n2 is even and
hence n is even. ■
Note the symbol ■, which means "the proof is now complete."
Our second example involves strings as discussed in Section 3.

Theorem 6.1. Let x <E [a, b}* such that xa = ax. Then x = a[n] for some
n G N.
Proof. Suppose that xa = ax but x contains the letter b. Then we can
write x = a[n]bu, where we have explicitly shown the first (i.e., leftmost)
occurrence of b in x. Then
a[n]bua =aa[n]bu =a[n + l]
bu.
Thus,
bua = abu.
But this is impossible, since the same string cannot have its first symbol be
both b and a. This contradiction proves the theorem. ■

Exercises
1. Prove that the equation (p/q)2 = 3 has no solution for p,q G N.
2. Prove that if x G {a, b}* and abx = xab, then x = (ab)ln] for some
n ΠN.

7. Mathematical Induction
Mathematical induction furnishes an important technique for proving
statements of the form (Vn)P(n), where P is a predicate on N. One
10 Chapter 1 Preliminaries

proceeds by proving a pair of auxiliary statements, namely,

P(0)
and
(Vn)(IfP(n) then P(n + 1)). (7.1)
Once we have succeeded in proving these auxiliary statements we can
regard (Vn)P(n) as also proved. The justification for this is as follows.
From the second auxiliary statement we can infer each of the infinite set
of statements:
IfP(0)thenP(l),
IfP(l) then P(2),
IfP(2)thenP(3),... .
Since we have proved P(0), we can infer P(l). Having now proved P(l) we
can get P(2\ etc. Thus, we see that P(n) is true for all n and hence
(Vn)P(n) is true.
Why is this helpful? Because sometimes it is much easier to prove (7.1)
than to prove (yJn)P(n) in some other way. In proving this second auxiliary
proposition one typically considers some fixed but arbitrary value k of n
and shows that if we assume P(k) we can prove P(k + 1). P(k) is then
called the induction hypothesis. This methodology enables us to use P(k) as
one of the initial statements in the proof we are constructing.
There are some paradoxical things about proofs by mathematical induc­
tion. One is that considered superficially, it seems like an example of
circular reasoning. One seems to be assuming P(k) for an arbitrary k,
which is exactly what one is supposed to be engaged in proving. Of course,
one is not really assuming (VAZ)P(/Î). One is assuming P(k) for some
particular k in order to show that P(k + 1) follows.
It is also paradoxical that in using induction (we shall often omit the
word mathematical), it is sometimes easier to prove statements by first
making them "stronger." We can put this schematically as follows. We
wish to prove (Vn)P(n). Instead we decide to prove the stronger assertion
(\/nXP(n)& Q(n)) (which of course implies the original statement). Prov­
ing the stronger statement by induction requires that we prove
P(0)& 0(0)
and
(Vn)[IfP(n) & Q(n) then P(n + 1) & Q(n + 1)].
In proving this second auxiliary statement, we may take P(k)& Q(k) as
our induction hypothesis. Thus, although strengthening the statement to
7. Mathematical Induction 11

be proved gives us more to prove, it also gives us a stronger induction

hypothesis and, therefore, more to work with. The technique of deliber­
ately strengthening what is to be proved for the purpose of making proofs
by induction easier is called induction loading.
It is time for an example of a proof by induction. The following is useful
in doing one of the exercises in Chapter 6.

Theorem 7.1. For all n <= N we have E" =0 (2/ + 1) = (n + l) 2 .

Proof. For n = 0, our theorem states simply that 1 = l 2 , which is true.
Suppose the result known for n = k. That is, our induction hypothesis is
k

£ ( 2 / + 1) = (* + l) 2 .
/= 0

Then
k+\ k

E (2/ + 1) = E (2/ + 1) + 2 U + 1) + 1
/=o /=o
= (k+ l) 2 + 2(k + 1) + 1
= U + 2) 2 .
But this is the desired result for n = k + 1. ■
Another form of mathematical induction that is often very useful is
called course-of-values induction or sometimes complete induction. In the
case of course-of-values induction we prove the single auxiliary statement

(\fn)[If (\fm)m<nP(m) then P(n)l (7.2)

and then conclude that (Vn)P(n) is true. A potentially confusing aspect of

course-of-values induction is the apparent lack of an initial statement
P(0). But in fact there is no such lack. The case n = 0 of (7.2) is

If (Vm)m<0P(m) then P(Q).

But the "induction hypothesis" (Vra) w < 0 P ( r a ) is entirely vacuous because

there is no m G iV such that m < 0. So in proving (7.2) for n = 0 we really
are just proving P(0). In practice it is sometimes possible to give a single
proof of (7.2) that works for all n including n = 0. But often the case
n = 0 has to be handled separately.
To see why course-of-values induction works, consider that, in the light
of what we have said about the n = 0 case, (7.2) leads to the following
12 Chapter 1 Preliminaries

infinite set of statements:

P(0),
IfP(0) thenPil),
IfP(0)&P(l)thenP(2),
IfP(O) & P ( l ) & P(2) then P(3),

Here is an example of a theorem proved by course-of-values induction.

Theorem 7.2. There is no string x e {a, b}* such that ax = xb.

Proof. Consider the following predicate: If x e {a,b}* and |JC| = n, then
ax ¥= xb. We will show that this is true for all n e N. SO we assume it true
for all m < k for some given k and show that it follows for k. This proof
will be by contradiction. Thus, suppose that \x\ = A: and ax = xb. The
equation implies that a is the first and b the last symbol in x. So, we can
write x = aw/7. Then

i.e.,
aw = ub.
But |u| < |jt|. Hence by the induction hypothesis au ¥= wo. This contradic­
tion proves the theorem. ■
Proofs by course-of-values induction can always be rewritten so as to
involve reference to the principle that if some predicate is true for some
element of N, then there must be a least element of TV for which it is true.
Here is the proof of Theorem 7.2 given in this style.
Proof. Suppose there is a string x G {a, b}* such that ax = xb. Then
there must be a string satisfying this equation of minimum length. Let x
be such a string. Then ax = xb, but, if \u\ < |x|, then au # ub. However,
ax = xb implies that x = aub, so that au = ub and \u\ < \x\. This contra­
diction proves the theorem. ■

Exercises
1. Prove by mathematical induction that Y,?={ i = n(n + l ) / 2 .
2. Here is a " p r o o f by mathematical induction that if x, y e /V, then
x = y. What is wrong?
7. Mathematical Induction 13

Let
fx úx>y
maxU,y) = < .
\y otherwise
for x, y e N. Consider the predicate
(Vx)(V)>)[// max(x, y) = n, thenx = y].
For n = 0, this is clearly true. Assume the result for n = k, and let
max(x, y) = k + 1. Let jct = x - 1, yx = y - 1. Then max(x t ,3;,) = k.
By the induction hypothesis, .*, = y , and therefore A: = jf, + 1 =
y, + 1 =>'.
3. Here is another incorrect proof that purports to use mathematical
induction to prove that all flowers have the same color! What is
wrong?
Consider the following predicate: If S is a set of flowers containing
exactly n elements, then all the flowers in S have the same color. The
predicate is clearly true if n = 1. We suppose it true for n = k and
prove the result for n = k + 1. Thus, let S be a set of k + 1 flowers. If
we remove one flower from S we get a set of k flowers. Therefore, by
the induction hypothesis they all have the same color. Now return the
flower removed from S and remove another. Again by our induction
hypothesis the remaining flowers all have the same color. But now
both of the flowers removed have been shown to have the same color
as the rest. Thus, all the flowers in S have the same color.
4. Show that there are no strings x, y <E {a, £>}* such that xay = ybx.
5. Give a "one-line" proof of Theorem 7.2 that does not use mathemati­
cal induction.
This page intentionally left blank
 Part 1

Computability
This page intentionally left blank
 2
Programs and
Computable Functions

1. A Programming Language
Our development of computability theory will be based on a specific
programming language ,¥\ We will use certain letters as variables whose
values are numbers. (In this book the word number will always mean
nonnegative integer, unless the contrary is specifically stated.) In particu­
lar, the letters
X\ X2 X3 •••

will be called the input vanables of <9\ the letter Y will be called the
output variable of <9\ and the letters

will be called the local variables of ćf\ The subscript 1 is often omitted; i.e.,
X stands for Xx and Z for Zv Unlike the programming languages in
actual use, there is no upper limit on the values these variables can
assume. Thus from the outset, 5? must be regarded as a purely theoretical
entity. Nevertheless, readers having programming experience will find
working with òf very easy.
In äf we will be able to write "instructions" of various sorts; a
"program" of Sf will then consist of a list (i.e., a finite sequence) of
17
18 Chapter 2 Programs and Computable Functions

Table 1.1

Instruction Interpretation

V <- V + 1 Increase by 1 the value of the variable V.

V <— V - 1 If the value of V is 0, leave it unchanged; otherwise decrease by 1 the
value of V.
IF V =£ 0 GOTO L If the value of V is nonzero, perform the instruction with label L next;
otherwise proceed to the next instruction in the list.

instructions. For example, for each variable V there will be an instruction:

V ^ V+ 1
A simple example of a program of J/7 is
X ^ X +1
X *- X + 1
''Execution" of this program has the effect of increasing the value of X by
2. In addition to variables, we will need "labels." In <9' these are
Ax B] Cx Dx E] A2 B2 C2 D2 E2 A^ —.
Once again the subscript 1 can be omitted. We give in Table 1.1 a
complete list of our instructions. In this list V stands for any variable and
L stands for any label.
These instructions will be called the increment, decrement, and condi­
tional branch instructions, respectively.
We will use the special convention that the output variable Y and the
local variables Z, initially have the value 0. We will sometimes indicate the
value of a variable by writing it in lowercase italics. Thus x5 is the value of
x5.
Instructions may or may not have labels. When an instruction is labeled,
the label is written to its left in square brackets. For example,
[B] Z <-Z - 1
In order to base computability theory on the language <9\ we will
require formal definitions. But before we supply these, it is instructive to
work informally with programs of y\

2. Some Examples of Programs

(a) Our first example is the program
[A] X<-X-l
Y <- Y+ 1
IF X * 0 GOTO A
2. Some Examples of Programs 19

If the initial value x of X is not 0, the effect of this program is to copy x

into Y and to decrement the value of X down to 0. (By our conventions
the initial value of Y is 0.) If x = 0, then the program halts with Y having
the value 1. We will say that this program computes the function

1 if x = 0
/<*> = ( x otherwise.
This program halts when it executes the third instruction of the program
with X having the value 0. In this case the condition X # 0 is not fulfilled
and therefore the branch is not taken. When an attempt is made to move
on to the nonexistent fourth instruction, the program halts. A program will
also halt if an instruction labeled L is to be executed, but there is no
instruction in the program with that label. In this case, we usually will use
the letter E (for "exit") as the label which labels no instruction.
(b) Although the preceding program is a perfectly well-defined pro­
gram of our language <9\ we may think of it as having arisen in an attempt
to write a program that copies the value of X into Y, and therefore
containing a "bug" because it does not handle 0 correctly. The following
slightly more complicated example remedies this situation.
[A] IF A^ ^ 0 G O T O B
Z ^ Z + 1
I F Z ^ O GOTO E
[B] X«-X-\
Y <- Y + 1
Z ^ Z + 1
IF Z # 0 GOTO A

As we can easily convince ourselves, this program does copy the value of
X into Y for all initial values of X. Thus, we say that it computes the
function fix) = x. At first glance. Z's role in the computation may not be
obvious. It is used simply to allow us to code an unconditional branch. That
is, the program segment
Z <-Z + 1 (2 j)
IF Z * 0 GOTO L
has the effect (ignoring the effect on the value of Z) of an instruction
GOTOL
such as is available in most programming languages. To see that this is true
we note that the first instruction of the segment guarantees that Z has a
nonzero value. Thus the condition Z ^ 0 is always true and hence the next
instruction performed will be the instruction labeled L. Now GOTO L is
Random documents with unrelated
content Scribd suggests to you:
condemnatory remark of Lewie Baronald.

'Their Highnesses,' replied the latter calmly, 'have no right to leave their
ports open to the King's rebels, in disregard of friendship and honour, and in
defiance of the remonstrances of his ambassadors.'

'Permit me to dispute your right, as a soldier in the service of their

Highnesses, to censure them.'

Baronald's nether lip quivered at this retort, and the Countess and
Dolores exchanged glances of uneasiness; for politics had become so
embittered by the American Squadron, which had recently captured H.M.'s
ships Serapis and Scarborough, having taken them into the Texel, and when
General Yorke claimed those ships and their crews, the Dutch refused to
restore them, and soon after Commodore Fielding fired upon their squadron
under Count Bylandt, and took him into Portsmouth; so war was looked for
daily, while the Scots Brigade were yet serving under the Dutch colours.

'Do not let us think or speak of such things, Cousin Maurice,' said
Dolores imploringly; 'I tremble at the idea of Britain invading us, if this sort
of work goes on. What have we to do with her colonists and their quarrels?'

'Invade us, indeed!' said Morganstjern, with angry mockery; 'if our
swords fail us we can open the sluices, as we did in the days of Louis XIV.,
and drown every man and mother's son!'

'But how should we escape ourselves?' asked the Countess.

'Good generalship would take care of that; and then how about your
Scots Brigade?' asked Morganstjern, turning abruptly to Baronald.

'The Bulwark of Holland, we have never failed her yet,' replied the latter
haughtily; 'but to draw the sword upon our own countrymen is certainly a
matter for consideration.'

'In Holland, perhaps, but not in America.'

 'Let this subject cease,' said the Countess imperatively, while fanning
herself with an air of excessive annoyance; and now Morganstjern,
beginning to find himself de trop, bowed himself out, and with vengeance
gathering in his heart, withdrew to an estaminet, or tavern, where he knew
that he would find his friend the Heer van Schrekhorn, and whither we may
perhaps follow him.

To Lewie Baronald, who was naturally destitute of much personal

vanity, it had hitherto seemed rather strange that the Countess had permitted
his attentions to her daughter at all, though he was known to be the heir of
his uncle; and now that he had all the joy of knowing himself to be her
accepted lover, his soul trembled within him at the prospect of having to
announce General Kinloch's utter hostility to the mother.

Lewie had more than once observed that the Countess always smiled, or
laughed outright, when his uncle the General was spoken of, as if she
considered him somewhat of a character—an excentrique.

'You have seen the General, I presume, since you were here last?' said
she.

'Yes, madam,' replied Lewie, painfully colouring.

'And told him of your love for Dolores—of your engagement to her, in
fact?'

'Yes, madam; he seems to have suspected it for some time past.'

'Suspected—that sounds unpleasant.'

Lewie played with the feather in his regimental hat, and his colour
deepened, when the Countess said:

'Then he is coming to wait upon me, I presume?'

'Alas—no, madam.'

'Indeed—why?'
 'He is averse to the society of ladies.'

'In fact, is a woman-hater, I have heard.'

Lewie smiled feebly, and felt himself in a foolish predicament; so the

Countess spoke again.

'He had some disappointment in early life, I believe, and never got over
it.'

'Yes, madam.'

'Poor fellow! and of Dolores——'

'He will not hear me speak with patience.'

'How grotesque!' exclaimed the Countess, laughing heartily.

'Ah, could he but see her!' exclaimed the young man, regarding the
upturned face of her he loved with something of adoration.

'Your uncle the General is very cruel, Lewie,' said Dolores; 'he is a
veritable ogre!'

'He is the king of good fellows, but——'

'But what?'

'He has never seen you.'

'And never shall,' she said petulantly, opening and shutting her fan.

'Nay, dearest Dolores, do not say so.'

'The ogre, or worse!' exclaimed the girl, with a pout on her sweet lips.

'Nay—no worse—only a man,' said the Countess, laughing excessively;

'he thinks of us only as women, but to be shunned—avoided—dreaded. It is
very droll!'
 And looking down, she played with the étui and appendages that hung
from her girdle, her tiny watch with the judgment of Solomon embossed on
its case; and as she did so, Lewie thought her hand as white and dimpled as
that of Dolores.

To him it certainly seemed strange that the opposition of his uncle

seemed only to provoke—not the pride or the indignation of the lively
Countess—but her laughter and amusement.

'And if he gets me banished on foreign service to the Dutch West

Indies!' he urged rather piteously.

'My poor Lewie,' said she, patting his cheek with her fan, 'I must see
what can be done; meantime, we must be patient and wait. From all I have
heard and know, an early disappointment at the hands of one he loved only
too well, has shaken his faith in human goodness and integrity, and now he
is soured, suspicious and sarcastic.'

'But only so far as women are concerned.'

'True; and I suppose he is like a French writer, who says that "of all
serious things, marriage is the most ridiculous;" but men are not infallible,
especially men like your uncle the General—errare humanum est. Let us be
patient a little, and all will come right in the end.'

But Dolores and her lover would only sigh a little impatiently as her
hand stole into his, and the twilight of evening deepened around them.

CHAPTER V.

'THE BULWARK OF HOLLAND.'

 And now while the lovers are waiting in hope, while the General is
'nursing his wrath to keep it warm,' and determined upon their separation;
and while Maurice Morganstjern is plotting what mischief he may work
them, we shall briefly tell the story of the Scots Brigade, and how it came to
be called 'the Bulwark of Holland.'

Lewie Baronald, his uncle the General, and all others belonging to the
Scots Brigade, had a good right to be proud of doing so, as it had a glorious
inheritance of military history, second only to that of the 1st Royal Scots;
and though its memory should have been immortal, its records now lie
rotting in a garret of the Town House of Amsterdam; and even in Scotland
little is remembered of it, save its march:

'The Lowlands o' Holland

Hae parted my love and me.'

Yet the drums of that Brigade have stirred the echoes of every city and
fortress between the mouths of the Ems in the stormy North Sea, and the
oak forests of Luxembourg and the Ardennes; and between the ramparts of
Ostend and the banks of the Maese and Rhine.

In 1570 the fame of Maurice, Prince of Nassau, drew to his standard

many of those Scots whose swords were rendered idle by peace with
England, and it was by their aid chiefly, that he drove out his Spanish
invaders. Among those who took with them the bravest men of the Borders,
were that Sir Walter Scott of Buccleugh who exasperated Queen Elizabeth
by storming the Castle of Carlisle; his son Walter, the first Earl of
Buccleugh; Sir Henry Balfour of Burleigh; Preston of Gourton; Halkett of
Pitfiran, and other commanders, named by Grose, Stewart, Hay, Douglas,
Graham, and Hamilton.

These formed the original Scots Brigade in the army of Holland, and
some of the battalions must have been kilted, as Famiano Strada, the Jesuit,
states that at the battle of Mechlin they fought 'naked'—nudi pugnant Scoti
multi.

In 1594, on the return of the States ambassadors, whither they had gone
to congratulate James VI. on the birth of his son, they took back with them
1,500 recruits for the Brigade, which five years after fought valiantly in
repelling the Spaniards at the siege of Bommel.

The year 1600 saw its soldiers cover themselves with glory at the great
battle of the Downs, near Nieuport, and in the following year at the siege of
Ostend, which lasted three years, and in which 100,000 men are said to
have perished on both sides, and where so many of Spinola's bullets 'stuck
in the sandhill bulwark that it became like a wall of iron, and dashed fresh
bullets to pieces when they hit it;' and so great was the valour of the Brigade
at the siege of Bois-le-Duc, that Frederick Henry, Prince of Orange, styled it
'The Bulwark of the Republic.'

It consisted then of three battalions—those of Buccleugh, Scott, and

Halkett.

The bestowal of some commissions on Dutch officers caused much

discontent during the time of the Prince of Orange, afterwards William III.,
with whom (after being demanded by James VII. without effect in 1688) the
Brigade came over to Britain for a time, and

served at the siege of Edinburgh Castle and the battle of Killiecrankie.

In 1747, by the slaughter at Val and the terrible siege of Bergen-op-

Zoom, the Brigade was reduced to only 330 men, but the Hague Gazette
records how they drove the French from street to street, and of all the glory
won thus, the greatest fell to two lieutenants named Maclean, the sons of
the Laird of Torloisk, who were complimented by Count Lowendhal, who
commanded the enemy, by whom they were taken.

The men of the Brigade were ever good soldiers, yet strict and God-
fearing Presbyterians, who would rather have had their peccadilloes known
to a stern General like Kinloch, than to the regimental chaplain.

And it might be said of this Brigade, as it used to be said of the Scots

Greys, that the members of it retained a kind of regimental dialect coeval
with the days of its formation, when the language was rather different from
the present Scotch; so, in the Irish brigades of France and Spain the
strongest and purest old Irish was found to the last.
 As a sequel to this brief account of the Brigade in Holland, we may sum
up the story of its service in the British army—though that service was
brilliant—in little more than a paragraph.

After a long and angry correspondence between the Governments of

Holland and Britain, the Brigade—save some fifty officers who had formed
ties in Holland, or elected to remain there—was transferred to the service of
the latter, when a rupture took place between them at the time of the
American War, and was taken to Edinburgh, clad in the Dutch uniform, and
about 1794, it adopted the red coat, and there in George's Square, when
drawn up under Generals Dundas and Kinloch, received its new colours at
the hands of the Scottish Commander-in-Chief, when it was numbered as
the 94th Regiment; and under these colours it fought gallantly at
Seringapatam, winning the elephant as a badge; at the storming of Ciudad
Rodrigo, and in all the battles of the Peninsular War, after which it was
disbanded at Edinburgh in 1818.

Then another 94th was embodied five years afterwards, on which

occasion, as we tell in our novel, 'The King's own Borderers,' 'the green
standard of the old Brigade of immortal memory was borne through the
streets from the Castle of Edinburgh by a soldier of the Black Watch,' thus
identifying the new regiment with the old; but now even the number of the
former has passed away, as under the new and helplessly defective army
organisation scheme, it is muddled up with the 88th Regiment under a new
name.

And now, having told what the 'Bulwark of Holland' was, we shall
return to the fortunes of Lewie Baronald and his fiancée Dolores.

CHAPTER VI.

AT THE GOLDEN SUN.

 'So—so! this Scottish adventurer stands between me and the girl I love;
between me and my own flesh and blood; more than all, between me and
the fortune of Dolores!' muttered Morganstjern—himself a penniless
adventurer and knave to boot, as he strode through the streets with his left
hand in the hilt of his sword and his right tightly clenched. 'I have a right to
hate and dread him—the right to remove him, too, by fair means or foul!'

The latter were the only means he could think of, as he had a
wholesome dread of Lewie Baronald's skill with his sword, and the bucks of
the Scots Brigade were not wont to stand on trifles when they resorted to
that weapon; and in this mood of mind he rejoined his friend the Heer van
Schrekhorn, whom he found at an estaminet called the Gond Zon, or
'Golden Sun,' in a narrow and gloomy street near the Klooster Kerk.

There he found him seated in a quiet corner, smoking, drinking

schiedam and water, while intently studying some profitable and useful
gambling tricks with a pack of not overclean cards.

'I have just been trying some ruses or tricks at lansquenet,' said he, as
the tapster brought fresh glasses and more liquor; 'it is the grandest of old
gambling games, like those that are of French origin. Look here, to begin
with: the cards being shuffled by the dealer, and cut by one of the party, two
are dealt out and turned up on the left hand of the dealer, so; he then takes
one and places a fourth, the réjouissance card, in the middle of the table, so.
On this——'

'Enough of this, Van Schrekhorn,' said Morganstjern impatiently. 'I did

not come here to be taught lansquenet,' he added, as he threw his sword,
hat, and gloves on a side-table, and flung himself wearily into a chair.

'Oh—so you have just come from the house of la belle Dolores?'

'Yes,' replied the other with an imprecation.

'And left her well and happy, I hope?' said the Heer mockingly.

I left her with that fellow of the Scots Brigade.'

 ''Sdeath! the more fool you. Why not keep your ground? Were you not
there before him?'

'Yes, and did my best to win her favour—even her mere regard.'

'In vain?'

'As usual. In fact, I think these two are affianced—or nearly so. I never
so bestirred myself about a woman before,' said Morganstjern, as he drank
at a draught a crystal goblet of schiedam and water, and refilled it.

'How is the Countess affected towards you?'

'But indifferently. Indeed, she only tolerates me as a kinsman, and, I

suppose, has encouraged or permitted Baronald's addresses to her daughter,
because he is the heir of General Kinloch, while I am heir to that only of
which nothing can deprive me.'

'And that is?'

'A grave—six feet of earth,' replied Morganstjern, grinding his teeth

unpleasantly.

'Come—you have always the guilders you win at roulette.'

'Because they are so won, there is the greater necessity that I should
have those of my cousin Dolores.'

'Which also reminds me that you owe me a good sum of money—cash

lent, and lost at play.'

'Why the devil remind me of that just now?' replied, or rather asked,
Morganstjern, savagely; and then for a little time the two smoked moodily
in silence.

The would-be lover of Dolores had long been subjected to a run of evil
fortune at the gaming-table. 'So long as there is the beacon of hope,' says a
writer, 'life is able to show up a gleam now and then of rose-colour; but
when adverse circumstances render any change impossible, life becomes
intolerable.' And to this verge of desperation had Maurice Morganstjern
come.

It was a source of keen irritation to him, to find that his rival—favoured

by the Countess—could be with Dolores daily, while he—her cousin—
could only visit her at stated times; and that all the advances he made to her
seemed utterly futile and hopeless now.

'Nearly penniless as I am, Schrekhorn,' said he; 'I might have waited
patiently, but have never had a gleam of hope.'

'If you waited a hundred years, it would be all the same, while she is
under the influence of this fellow's voice, eye, and society.'

'What would you have me do?'

'Remove him, or remove her!' replied Schrekhorn with a fierce Dutch

oath.

'More easily said than done. With her money, by the henckers! how I
should enjoy myself all day long and do nothing!'

'About all you ever cared to do,' rejoined his friend, who was rather
disposed to treat him mockingly.

'Don't attempt to act my Mentor.'

'Why?'

'Because I should make but a sorry Telemachus.'

'Then it seems all a settled thing between them!'

'What?'

'How dull you are! Marriage?'

'Ach Gott! it looks like it.'

 'Then you have been a trifle late of taking the field?'

'Nay,' replied Morganstjern, smoking his meerschaum with vicious

energy. 'I was the first, but when this fellow Baronald came, I found myself
instantly at a discount.'

'Jilted—eh?'

'Nay, I was never on such a footing with her that she could treat me so,
because she was ever utterly indifferent.'

'Then it is too late for fair means now, but not for foul.'

Then, after a pause, the Heer said in his mocking tone;

'If money is your object, and you openly avow that it is so, why not
propose to the mother, if the daughter won't have you? She is rich enough
and certainly handsome enough, and only some fifteen years older than
yourself. She is a widow, and all the world knows how easily widows are
won. 'Sblood! cut in for her, and leave the girl to the Scot.'

Morganstjern thought for a minute, and then uttering one of his

imprecations, added:

'No—no—NO! I shall be thwarted by no man!'

'Right!' exclaimed the other; 'I like this spirit—give me your hand.'

'This infernal Scots Brigade has married at the Hague and Amsterdam
more than fifty Dutch girls within the last few years, and all of them rich.'

'Der Duyvel!'

'Many of them girls of the first rank.'

'Thousand duyvels!' said the Heer with a mocking laugh.

'Is it not enough that these Scots—the Bulwark of the Republic as they
boast themselves——'
 'And have done so since the old siege of Bois-le-Duc—well?' asked the
Heer.

'Is it not enough, I say, that they should assume our glory in war, and
win our guilders in peace, but they must carry off our prettiest girls too?'

'They do not assume your glory, but win their own,' said the Heer, who
had some contempt for his companion; 'their guilders have been hardly won
on many a Dutch and Flemish battlefield; and if the pretty girls of Haarlem
and the Hague prefer them to Walloons, they are right.'

Morganstjern's brow grew black.

'I am no Walloon,' said he, huskily.

'I did not say so,' said Van Schrekhorn; then he added, 'I have some
news for you, and a hint to make thereon. Dolores van Renslaer is to be at
the ridotto given by the wife of the Sixe van Otterbeck, the Minister of
State, on the night after next.'

'That I know, and of course this pestilent Scot will be there too.'

'No; on that night he is on duty at the Palace of the Prince of Orange.'

'Well—what about all this?'

'Listen,' said Van Schrekhorn, leaning forward on the table and lowering
his voice almost to a whisper, while the colour in his bloated visage
deepened, and an expression of intense cunning stole into his watery
bloodshot eyes: 'let us carry her off as her sedan bears her from the ridotto!'

'To where?'

'Listen. I know a skipper whose ship is now in the Maese, and almost
ready to sail for the coast of France. She is anchored off Maesluis now; let
us once get her on board and the Hoek van Holland will soon be left astern,
and the girl your own, unless you are a greater fool than I think you.'

Morganstjern made no immediate reply, so his tempter spoke again.

 'Once on board that ship, her honour will be compromised, and marriage
alone can restore it. Let her be once on board that ship with you, I say, and
she cannot be so blind as not to see that she will have gone a great deal too
far to draw back.'

'Right!' exclaimed Morganstjern, as a glance of triumph came into his

eyes. 'I have a political mission to France, and it will be supposed that she
has eloped with me, and befooled the Scot Baronald. With all her contempt
and scorn of me, she little knows that her fate is to become my wife—my
wife—mine! Once that, and then let her look to herself!' he added as a
savage expression mingled with the triumph that sparkled in his shifty eyes,
and he smote the table with his clenched hand.

'The distance from the Hague to Maesluis is only eleven miles—a few
pipes, as the people say,' resumed Schrekhorn; 'my friend shall have a boat
waiting us at a quiet spot among the willows that fringe the shore, near a
deserted windmill on the river-bank; and then we shall take her on board.
Once under hatches, her fate will soon be sealed.'

'How can I thank you?'

'By refunding what you owe me out of the guilders of Dolores,' replied
the Heer, as he and Morganstjern shook hands again; but the latter became
silent for a time.

He knew the Heer van Schrekhorn to be a rascal capable of committing

any outrage, and also that he had personally a special grudge at Lewie
Baronald. Dolores was beautiful. What if this scheme so speciously
arranged, was one for his own behoof, to carry her off, leaving the onus of
the abduction on the shoulders of him—Morganstjern—after passing a
sword through his body among the willows near the old mill on the Maese.

But this grave suspicion was only a passing thought, and he thrust it
aside.

'This may preclude your return for some time, and compromise you
with the authorities,' said the Heer.
 'Their reign will soon be over; and when a French army comes to the
assistance of the Dutch patriots, the Prince of Orange may find himself a
fugitive in England.'

'But we must be wary; not for all the gold and silver bars in the Bank of
Amsterdam would I be in your shoes if we fail. The Burgomasters are
worse than the devil to face, and we may find ourselves behind the grilles of
the Gevangepoort or the Rasp-haus, as brawlers.'

'A thousand duyvels!—fail? don't think of it.'

Had Maurice Morganstjern known the intentions of General Kinloch

towards his nephew, and the plans he had formed to separate him from
Dolores, he might have patiently awaited the events of the next few days;
but as he was ignorant of them, he and the malevolent Heer van Schrekhorn
laid all their plans for the abduction of the girl with caution, confidence, and
extreme deliberation, before they quitted the Golden Sun that night.

CHAPTER VII.

THE GENERAL'S SECRET.

Next day, when Lewie Baronald, apparelled in all his regimental

bravery, was setting forth to visit Dolores, he was summoned by General
Kinloch, who, after working himself up to a certain degree of sternness or
firmness, real or assumed, for the occasion, said:

'Stay, young man, I pray you, as we must have some conversation

together.'

Lewie took off his Khevenhüller hat, and fearing that some
animadversions were coming, played a little irresolutely with its upright
scarlet feather.
 'Your name has gone in for foreign service, Lewie,' said the General.

'To whom, sir?'

'The Director-General of Infantry.'

'Sent by you, uncle?'

'Yes, sir, by me.'

'You might at least have consulted with me in this matter. How cruel of
you, uncle, under all the circumstances!' exclaimed Lewie, with sudden
bitterness and intense anger.

'You will come to think it kindness in time, boy; I seek but to save you
from what I, in my time, underwent.'

'If I refuse to go?'

'Refuse, and compromise your honour and mine—yea, the honour of the
Brigade itself! My dear Lewie, when you have lived in this world as long as
I——'

'Why, uncle, you are only forty!'

'Not yet twice your age, certainly—well?'

'If detailed for the Colonies, anywhere, separation from Dolores will be
the death of me!' exclaimed the young man passionately.

'No, it won't; nor of Dolores either. So you are very much in love with
her?' asked the General with a scornful grin.

'God only knows how purely I love her!' exclaimed the nephew in a low
concentrated voice.

'Nature is full of freaks, certainly!'

'How?'
 'She has varied the annals of the old fighting line of the Baronalds of
that Ilk, by having them varied by something else.'

'By what?'

'A moonstruck fool!'

'This is eccentricity combined with unwarrantable interference and

military tyranny,' cried Lewie, as he stuck his hat on his head and drew
himself haughtily up; then in a moment his mood changed, for he loved this
kinsman to whom he owed so much, and he said with an air of dejection,
'How shall I ever tell Dolores of what you have done to us both? I cannot
sail for the Cape or the Caribbean Isles, and leave her bound to me! I must
release her from her promise, though I know that she would wait a lifetime
for me.'

'Poor fool that you are, Lewie! Do you forget the adage, "Out of sight,
out of mind"? You think that, like Penelope, she will wait your return in
hope, in love, and all the rest of it? You may be like Ulysses, but never was
there a Penelope among women.'

The General indulged in many more doubting and slighting remarks

upon women, particularly on their faith and constancy; and while he was
running on thus, grief struggled with rage and indignation for mastery in the
heart of Lewie, which seemed to stand still at this sudden wrench, and the
prospect of an abrupt and protracted separation from Dolores—a separation
that might be for years—every moment of which would be an agonised
heart-throb, it seemed to him then!

How hard, how cruel, that they should be thus separated, and forced to
drink, as it were, of the bitter waters of Marah, because this stern soldier
hated all women so grotesquely, as the Countess had said, viewing them all
through the medium of one; while Lewie and Dolores were so young that
all the world seemed too small to contain the measure of their joy, and now
—now, thought was maddening!

He would resign, 'throw up his pair of colours,' as the phrase was then;
but his uncle had compromised him, by sending in his name to the Director-
General of Infantry!

Already in anticipation he imagined and rehearsed their parting; already

he saw her tears, her blanched face, and heard himself entreating her not to
forget him, while vowing himself to be true to her—each regarding the
other mournfully and yearningly, hand clasped in hand, lip clinging to lip;
then came the void of the departure; the seas to plough, and the years that
were to come with all their doubts and longing.

It was too bad—too bad; he owed his uncle much—all in the world
indeed; but this stroke—this harsh interference, ended all between them for
ever!

Overwhelmed with dejection he cast himself into a chair; there the

General regarded him wistfully, and placing a hand kindly on his shoulder,
said:

'Lewie, shall I tell you of what once happened to me?'

But, full of his own terrible thoughts, Lewie made no reply.

'It may have been that evil followed me,' said the General, looking
down, with a hand placed in the breast of his coat.

'Evil?' repeated Lewie.

'Yes. When a boy I shot in the wood of Thomineau the last crane that
was ever seen in Scotland, and my old nurse predicted that a curse would
follow me therefor; thus, I never see a crane on a house-top here that I don't
remember her words. Now listen to what happened to me when I was on
detachment in the Dutch West India Islands. I belonged then to the battalion
of Charles Halkett Craigie, who six years ago died Lieutenant-Governor of
Namur, and we garrisoned Fort Nassau, or New Amsterdam as it is called
now. There,' continued the General, alternately and nervously toying with
his sword-knot and shirt-frill, 'I was silly enough to fall in love with the
daughter of a wealthy merchant, a Dutch girl, like your Dolores, with some
of the old Castilian blood in her, though a lineal descendant of the great
Dutch family of Van Peere, to whom, in 1678, Berbice was granted by the
States-General as a perpetual and hereditary fief. She possessed great
beauty, and what proved more attractive still, a hundred sweet and winning
ways, with the art of saying pretty and even daring little things, that
endeared her to all—to none more than me. I was a great ass, of course; but,
heavens, what a coquette she was!'

'What was her name?' asked Lewie, with just the smallest amount of
interest.

'Excuse me telling, as I have sworn never to utter it again; nor do I wish

it to go down in the annals of our family. She wound herself round my
heart; my soul, my existence, seemed to be hers. My love for her became a
species of idolatry; but poverty tied my tongue, and I dared not speak of it,
till one evening, which I shall never forget, the secret left me abruptly,
drawn from me by herself. We were lingering in the garden of her father's
villa near the Berbice river, and the stars were coming out, one by one, in
the deep blue sky above us. The hour was beautiful—all that a lover could
wish; and around us the atmosphere was fragrant with the perfume of
flowers—among those wonders of the vegetable world—the gigantic water-
lilies, each leaf of which is six feet in diameter. I was soon to leave for
Holland on duty, and my heart was wrung at the prospect of a separation.

'I had her hand in mine: my secret was trembling on my lips; and gazing
into her eyes which were of a golden-brown colour, like that of her hair, I
said very softly:

'"If your eyes have at all times an expression so sweet, so beautiful and
winning, what must they seem to the man who reads love in them—love for
himself!"

'"Can you not read it now?" she asked in a low voice, as she cast her
long lashes down.

'I uttered her name and drew her close to me, my heart beating wildly
the while, in doubt whether this was one of the daring little speeches I
spoke of.
 'Taking her sweet little face between my hands, I kissed her eyes and
forehead, on which she said, in her low cooing voice:

'"I wonder if you will ever think of me after you are gone?"

'"Darling, do you think there will ever be a day of my life when I will
not think you! Oh, the thought of our parting is worse than death to me!"

('A fellow feeling makes us wondrous kind,' thought Lewie, becoming

fully interested now.)

'"We are jesting," said she; "do not say this."

'"There is no need to tell you that I love you," said I, "for you know that
I do—dearly, fondly: that this love will last with life, end with death;" and
much more rubbish I said to the same purpose, adding, "And you, if quite
free, could you love me?"

'"I love you now; have I not admitted as much?"

'So it all came about in that way,' said the General; 'umph—what an ass
I was! May you never live to be deceived as that girl deceived me! I thought
our passion was mutual; and then perhaps she thought so too—all
perfidious though she was!

'But how happy—how radiantly happy I was for a time, till a Dutch
squadron came to anchor off the bar of the Berbice river, and in one of the
lieutenants thereof she discovered, or said she discovered, a kinsman; and
from that moment a blight fell upon me, and I discovered that she was
variable as the wind. Her attentions seemed divided for a time; at last they
were no longer given to me. Her smiles were for the stranger; she sang to
him, played to him, and talked to him only. At home or abroad, riding or
driving, or boating on the river, he was ever by her side when not on board
his ship.

'What rage and mortification were in my heart! The rules of the service
alone prevented me calling him to a terrible account, though indeed he was
not to blame.
 'When I attempted to reason or remonstrate with her, she laughed; then
after a time became indignant. We parted in anger, and I felt fury and death
in my heart when she tossed my engagement-ring at my feet.

'Once again we met, alone, and by the merest chance. How my pulses
throbbed as our eyes met, and she coyly presented her hand, which I was
craven enough, and fool enough, to fondle!

'"Oh, what have I done," said I, "that you should treat me thus? that you
should tread my heart under your feet, and leave me to long years of sorrow
and repining?"

'Then she laughed, and snatched her hand away, while once again my
soul seemed to die within me.

'"Do you love this kinsman?" I asked her fiercely; and never till my last
hour shall I forget her reply, or the almost cruel expression of her face.

'"Yes; I love him—love him with my whole heart, and as I never loved
you!"

'Turning away, she left me—left me rooted to the spot. Yet she had some
shame, or compunction, left in her after all; for next day came a would-be
piteous letter of explanation, that she had given this lieutenant a promise to
please her father when he was dying—her father who was his guardian;
how she had never had the courage to tell me so at first; that she did not
dream I loved her so much; that I must learn to forget her, though she would
never forget me; and so—a thousand devils!—there was an end of it.

'A few weeks after I saw her marriage in the papers, to the Lieutenant—
d—n his name—to her and her fortune of ever so many thousand guilders.

'I tore her farewell letter into minute fragments, and set to work to adopt
her advice.'

'What was that?' asked Lewie.

 'To forget her; and to do so I threw myself into my profession. I never
looked upon her face again, and I thanked God when I heard our drums
beating as we marched out of Fort Nassau, and when the accursed shore of
the Berbice river faded into the evening sea! Now, Lewie, have I not the
best of reasons for mistrusting women, and seeking to save you from the
fangs of this little ogress—this Dolores?'

'Ah, you know not her of whom you speak thus!' exclaimed Lewie.

'Nor am I likely to do so. Shun her, nephew! a girl, doubtless, with a fair
face, and a heart as black as Gehenna! Be firm, Lewie Baronald!—firmness
is a great thing, as you will find when you come to be a general officer and
as old as I am.'

Lewie had done his duty like a man and a soldier—like one worthy of
the glorious old Brigade—among the savages in the old Cape War; but it
was cruel, absurd, and, to use the Countess van Renslaer's phrase,
'grotesque,' that he should now be treated like a child, and in the most
momentous matter of his life and happiness too!

'I was weak enough—idiot enough, to wish I might die, then and there,
when that girl deceived me,' resumed his uncle bitterly; 'but I knew that I
must live on and on; I was very young, and thought I might live for forty
years with that pain in my heart at night and in the morning. It is twenty
years since then, and though the pain is dead, I suppose, I cannot laugh at it
yet, or the memory of Mercedes.'

'Mercedes! was that her name—Mercedes?'

'The devil—it has escaped me!'

'So that is the name which is not to go down in the annals of the family?'

'Precisely so.'

'But surely, dear uncle, after all these years, you must have forgiven
her? Besides, she may be dead.'
 'Dead to me, certainly! Forgiven her—well, perhaps I may have
forgiven her; but what can make a mere mortal forget a wrong, a cruelty, or
an injury?'

'Then you will not yield, but insist that I shall go abroad?'

'I will not yield an inch, and march you shall!' replied the General, as he
turned on his heel and left him.

'My darling Dolores—the first and only love of my life!' exclaimed the
young man passionately; 'how can he—how dare he—act thus towards us?
But that I love him, I think, I may soon come to hate him!'

He rushed away in search of Dolores; but she and the Countess were
from home. He was on duty at the Palace next day, and Dolores was to be at
the ridotto; thus, ere they could meet, events were to transpire which were
altogether beyond the conception of both.

CHAPTER VIII.

THE RIDOTTO.

The 'ridotto,' the Italian word then fashionable for an entertainment of

music and dancing, at the huge old red-brick villa of the Heer van
Otterbeck, Minister of State, in the vicinity of the Hague, was one of the
gayest affairs of the season.

The Prince of Orange (whose son afterwards became King of the

Netherlands) was not present, but all the rank, the wealth, and beauty of the
Hague were represented; and among those present were many officers of
the Scots Brigade, including the Earl of Drumlanrig, General Dundas, in
after years the captor of the Cape of Good Hope; and there too was the
Conservator of Scottish Privileges at Campvere, John Home, then the
celebrated author of the nearly forgotten tragedy of 'Douglas.'

A band of the Dutch Guards furnished music on the lawn, and there
dancing was in progress in the bright sunshine of the summer afternoon;
and, in the fashion of the time, many of the guests were arrayed in what
they deemed the costume of Arcadian shepherds and shepherdesses.

People danced early in the evenings of the eighteenth century, and were
abed about the time their descendants now begin to dress for a ball. Ices
were unknown; no wine was dispensed, but the liveried servants of the
Heervan Otterbeck regaled his guests on coffee, green tea, orange tea, and
many kinds of cakes and confectionery in the intervals of the dancing, in
which Dolores (all innocent and unaware of the plots in progress against her
peace, even her honour and liberty—one of them born of avarice, wounded
vanity, and foiled desire) indulged joyously and with all her heart.

For the information of the ladies of the present day we shall detail the
dress worn by Dolores on that evening as described in the Hague Gazette,
and they may imagine how charming she looked:

'Her body and train were silver tissue, with a broad silver fringe; her
petticoat was white satin covered with the richest crape, embroidered with
silver, fastened up with bunches of silver roses, tassels, and cords. Her
pocket-holes were blonde, her stockings were blue, clocked with silver, and
her hair was twisted and plaited in the most beautiful manner around a
diamond comb.'

Seated under a tree, flushed with a recent dance, she was alternately
playing with her fan and silver pomander ball, with a crowd of admirers
about her, and looking alike pure and bright, with 'a skin as though she had
been dieted on milk and roses.'

'No wonder it is, perhaps, that Lewie loves me,' thought the girl, as she
looked at the reflection of her own sweet face in a little bit of oval mirror in
the back of her huge Dutch fan; 'I am pretty!'
 She might have said 'lovely,' and more than lovely; and then she smiled
consciously at her own vanity.

Under the genial influence of her surroundings the heart of the girl was
full of happiness, and had but one regret that Lewie Baronald was not there.
Yet, she thought, 'to-morrow I shall see him—to-morrow be with my
darling, who at this moment is thinking of me.'

And amid the brilliance of the scene, so rich in the variety of colour and
costume, the strains of the music and beauty of the old Dutch pleasure-
grounds, she almost longed to be alone, with the grass, the birds, the
insects, and the flowers—alone in the sweet summer evening with the
perfume of the roses, the jasmine, and the glorious honeysuckle around her.

On one hand, about a mile distant, was the Hague, with all its Gothic
spires and pointed gables; on the other spread the landscape so usual in that
country of cheese and butter—church-towers and wind-mills, bright
farmhouses, long rows of willow-trees, their green foliage ruffling up white
in the passing breeze; the grassy dykes and embankments, a long continuity
of horizontal lines, which seemed so tame and insipid to the mountaineers
of the Scots Brigade, and to all but the Dutch themselves.

Among the groups around her, Dolores, as usual now, heard the growing
political quarrel between Great Britain and Holland openly and freely
discussed, together with the consequent and too probable departure of the
Scots Brigade from the latter for ever. That seemed almost a settled thing—
a certainty, if the quarrel became an open one, and the probabilities wrung
the girl's affectionate heart.

How would all this affect her lover and herself? Alas! she knew not that
the doom of the former for foreign service was nearly a fixed thing now!
And she was fated to receive her first mental shock that evening, all
unwittingly, from the Earl of Drumlanrig, who drew near her, and with the
stately manner of the time lifted his hat with one hand, and with the other
touched her hand as he bowed over it.

The golden light of the setting sun fell full upon her hair, flecking its
bronze with glorious tints, and giving her beauty a brilliance that, to the
Earl's appreciative eye, was very striking.

'You look like one of Watteau's beauties, waiting to hear herself

addressed in the language of Love,' said the old peer, smiling.

'Love has three languages, my lord,' observed Dolores.

'Three?'

'The pen, the tongue, and the eyes.'

'True; but I am too old to use any of these now,' said the Earl, shaking
his powdered head.

'The evening is a lovely one,' observed Dolores, after a pause.

'And the landscape yonder, as it stretches away towards Delft, is

wonderfully steeped in sunshine; and but for its flatness——' the Earl
paused.

'Your Scottish eyes cannot forgive that,' said Dolores laughing, as she
recalled some of Lewie Baronald's complaints on the same subject; 'but
people cannot live on scenery.'

'So the great Samuel Johnson has written.'

'Who is he?' asked Dolores.

'A great lexicographer—a wonderful English savant—who believes in a

ghost in London, yet discredited the late earthquake at Lisbon. I think I
have seen you at the Vyverberg with Lewis Baronald of my battalion; he
has the honour of being known to you.'

'He visits us,' replied Dolores, the flower-like tints of her sweet face
growing brighter as the Earl spoke.

'He is a fine and handsome fellow, young Baronald; but it is strange that
he should wish to quit the Hague when it possesses such peculiar
attractions,' said the Earl markedly, and with a courteous bow.
 'Quit the Hague!' repeated Dolores, as if she had not heard him aright.

'I do not know whether the desire to do so, has any connection with his
uncle's scheme for the recapture or restitution to Holland of the Island of
Goree, off the coast of Senegal, in defiance of the old Treaty of Nimeguen,
which gave it to France, a scheme which will win him the favour of their
Mightinesses; but young Baronald's name was sent, through me this
morning, to the Director-General of Infantry, for instant foreign service.'

'Foreign service!' whispered Dolores, in an almost breathless voice,

while her white throat gave a sharp nervous gasp, and her long lashes
drooped over her beautiful eyes. 'Surely, my lord, this must be some
mistake. Lewie—he had no desire to leave Holland, in any way—he
dreaded nothing so much as the departure of the Brigade to Britain; and this
—this——'

'No mistake, I assure you,' interrupted the Earl, all unaware of the
astonishment he was exciting and the pain he was inflicting, and both of
which he must have perceived had not the Heer van Otterbeck, fortunately
for Dolores, approached at that moment, and tapping and proffering his
Sèvres china snuff-box, 'buttonholed' him on the inevitable subjects, the
quarrel between Britain and Holland, Paul Jones in the Texel, and
Commodore Fielding's conduct in firing on the Dutch fleet in the Channel,
which the Commodore did with hearty goodwill.

But for Dolores, the charms of the ridotto had vanished now; and in sore
perturbation of spirit and anxiety of heart, she bade her host and hostess a
hurried farewell, summoned her sedan, and took her departure homeward.

The lights, the music—the music of Lulli; the minuet de la cour, and the
gaiety of the ridotto, faded away behind her as the heiress took the
somewhat lonely road that led to the villa of her mother.

She was escorted to her sedan by an officer of the Brigade, a friend of

Lewie's, who, as he closed the roof of it over her, thought that she looked
like—as he vowed to some others—'a lovely queen in wax-work done up in
a glass-case.'
 CHAPTER IX.

THE ABDUCTORS.

What was this mystery concerning the movements and intentions of

Lewie Baronald, on which the Earl of Drumlanrig had so abruptly but
unconsciously thrown a light?

When last they met and parted, Lewie had given no hint of any desire
for foreign service, and certainly, with the relations then existing between
himself and her, it was the last thing to be thought of.

'Oh,' thought Dolores, 'that I were at home to consult mamma on this

amazing subject!'

Her bearers seemed to crawl; she narrowly opened and shut her fan
again and again in her impatience, and stamped her little foot on the floor of
the sedan in her irritation and anxiety.

Yes! that horrid General—that odious uncle, the eccentric woman-hater,

was no doubt at the bottom of it, and had thus resolved to separate Lewie
from her, and hot tears started to her eyes at the thought.

Though in the immediate vicinity of the Hague, the road was as lonely
as those who awaited her thereon could have wished. The blue dome of
heaven, a dome studded with diamonds—each itself a world—was
overhead; and steady and silvery was the light of the uprisen moon, above
the far expanse of the level landscape.

Suddenly Dolores heard the sound of voices; there were threats on one
hand and expostulation on the other. The sedan, with a violent jolt, was
suddenly deposited on the ground, and its bearers were dashed aside, as she
supposed, by foot-pads. Then a shriek of dismay escaped Dolores, when a
man, whose face was half-concealed by a crape mask, threw up the roof of
the sedan, opened the door and attempted to drag her out by the hand.

She saw another similarly masked, and a caleche, with a pair of horses,
close by.

Never dreaming of outrage for a moment, she thought that she must be
the victim of some extraordinary mistake, till she recognised the voice of
Maurice Morganstjern, when her alarm and astonishment instantly changed
to indignation.

'Maurice,' she exclaimed, 'for whom do you mistake me? What outrage
is this?'

'No mistake at all, my pretty cousin; will you please to take your seat in
this caleche?' he replied deliberately.

'For what purpose?'

'Time will show, beloved Dolores.'

'Loose my hand. I wish none of your fair words; they are ever hateful
and unwelcome to my ear: more so than ever when you come thus—as you
must be—intoxicated,' she added, believing this to be the case.

'Beware, cousin—beware! You know how I love you, and yet you spurn
me. Come, Schrekhorn, and help me to lift her into the caleche. For all the
past bitterness I shall have a sweet revenge; and, Dolores, you will learn to
love me, when you will have none else in this world to cling to.'

On seeing the Heer van Schrekhorn, of whose character she had heard
something, approach her, the girl looked wildly round in terror: the road
was lonely; her home was at some distance, yet the lights in its windows
were visible; but no help was nigh. She now perceived that nothing less
than her forcible abduction was daringly intended; but what lay in the future
beyond that, she could scarcely realise.
 Her first fears returned with double force, for she knew the recklessness
of the two men at whose mercy she found herself. How lovely and helpless
she looked!

Ruffian and coward though he was, Maurice Morganstjern was a

consummate egotist, and her continued indifference and contempt of him
had deeply wounded his amour propre, and roused a spirit of revenge.

'It is useless to fight against Fate, Cousin Dolores; and Fate decrees that
you are to be mine!' said he, firmly grasping her hand.

'Oh that I were a man!' exclaimed Dolores.

'For what purpose?'

'To strike you to the earth for your insolence and daring.'

'In that case I would not seek to carry you off; so, I thank Heaven that
you are not a man, sweet cousin!' He placed his face close to hers, and
lowering his voice, said through his clenched teeth: 'Listen to me, Dolores;
you have, I fear, plighted yourself to the Scotsman Baronald in ignorance of
yourself, and now I am here to rescue you from the death in life to which
your girlish folly would doom you. I will soon teach you to forget that artful
interloper, if you ever thought seriously about him, which I cannot believe,
and our marriage will alter all your ideas.'

These references to her lover infuriated Dolores, who was a high-

spirited girl; but he wound his arms round her despite all her efforts. With
all her strength she kept him, however, at arms' length, exclaiming:

'I hate you—oh, how I hate you!'

'Cease this nonsense, cousin; a day is coming when you will love me as
much as you may think you hate me now!'

'And what will cause the change?' she asked scornfully.

'Marriage.'
 'Why waste time thus?' asked the Heer van Schrekhorn, who had not yet
spoken, and who listened to all this with manifest impatience and
uneasiness; 'we know not who may come upon us; so into the caleche with
her at once!' he added with an oath.

''Sdeath, but she is as strong as I am!' exclaimed Morganstjern, as he

strove to drag her from the sedan.

Her slender figure stood very erect, and with tiny hands she strove to
free herself from his odious grasp; but the scorn, indignation, and
passionate resentment that flashed in her dark eyes and curled her tender
lips, now gave place to much of genuine fear of her assailants and how far
their daring might carry them, especially when the Heer laid his brutal
hands upon her; and uttering a wild cry she clung to the sedan, and without
a resort to extreme violence would not be torn from it.

Meanwhile the driver of the caleche, who was in ignorance of the

purpose his employers had in view, looked on somewhat scared, and was
thinking of how he might, in the future, be handled by the Burgomaster or
other authorities.

Dolores suddenly found her strength give way, and felt about to faint,
when she heard a loud and wrathful exclamation as Morganstjern was
dashed aside on one hand, Schrekhorn knocked down in a heap on the other,
and there towered between her and them a tall military-looking man,
wearing a Khevenhüller hat, and having a scarlet roquelaure wrapped round
him.

The latter he instantly threw off, and drew his sword, on which the
driver of the caleche whipped up his horses, and fled at full speed towards
the Hague, leaving his employers to get out of the affair as they best could.

The first impulse of the two conspirators was to unsheath their swords
also; but their second was to pause ere attempting to use them, as they
recognised in their assailant an officer of the Scots Brigade, and one of high
rank apparently by his gold aiguilette.

'Protect me, sir—save me!' implored Dolores.

Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.

Let us accompany you on the journey of exploring knowledge and

personal growth!

ebookfinal.com