From Differential Geometry to Non commutative

Geometry and Topology Neculai S. Teleman

download

https://textbookfull.com/product/from-differential-geometry-to-
non-commutative-geometry-and-topology-neculai-s-teleman/

Download more ebook from https://textbookfull.com

We believe these products will be a great fit for you. Click
the link to download now, or visit textbookfull.com
to discover even more!

Differential Geometry and Mathematical Physics Part II

Fibre Bundles Topology and Gauge Fields Gerd Rudolph

https://textbookfull.com/product/differential-geometry-and-
mathematical-physics-part-ii-fibre-bundles-topology-and-gauge-
fields-gerd-rudolph/

A Computational Non commutative Geometry Program for

Disordered Topological Insulators 1st Edition Emil
Prodan (Auth.)

https://textbookfull.com/product/a-computational-non-commutative-
geometry-program-for-disordered-topological-insulators-1st-
edition-emil-prodan-auth/

Algebraic and Differential Methods for Nonlinear

Control Theory Elements of Commutative Algebra and
Algebraic Geometry Rafael Martínez-Guerra

https://textbookfull.com/product/algebraic-and-differential-
methods-for-nonlinear-control-theory-elements-of-commutative-
algebra-and-algebraic-geometry-rafael-martinez-guerra/

Differential geometry of manifolds Second Edition

Lovett

https://textbookfull.com/product/differential-geometry-of-
manifolds-second-edition-lovett/
Differential geometry and Lie groups A computational
perspective Gallier J.

https://textbookfull.com/product/differential-geometry-and-lie-
groups-a-computational-perspective-gallier-j/

Topics in Modern Differential Geometry 1st Edition

Stefan Haesen

https://textbookfull.com/product/topics-in-modern-differential-
geometry-1st-edition-stefan-haesen/

Differential geometry of manifolds Second Edition

Lovett Stephen T

https://textbookfull.com/product/differential-geometry-of-
manifolds-second-edition-lovett-stephen-t/

Basics from Algebraic Geometry Hanspeter Kraft

https://textbookfull.com/product/basics-from-algebraic-geometry-
hanspeter-kraft/

Winding Around The Winding Number in Topology Geometry

and Analysis 1st Edition John Roe

https://textbookfull.com/product/winding-around-the-winding-
number-in-topology-geometry-and-analysis-1st-edition-john-roe/
Neculai S. Teleman

From Differential
Geometry to
Non-commutative
Geometry and
Topology
From Differential Geometry to Non-commutative
Geometry and Topology
Neculai S. Teleman

From Differential Geometry

to Non-commutative
Geometry and Topology
Neculai S. Teleman
Dipartimento di Scienze Matematiche
Università Politecnica delle Marche
Ancona, Italy

ISBN 978-3-030-28432-9 ISBN 978-3-030-28433-6 (eBook)

https://doi.org/10.1007/978-3-030-28433-6

Mathematics Subject Classification (2010): 53-XX

© Springer Nature Switzerland AG 2019

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of
the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation,
broadcasting, reproduction on microfilms or in any other physical way, and transmission or information
storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology
now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
The publisher, the authors, and the editors are safe to assume that the advice and information in this book
are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or
the editors give a warranty, expressed or implied, with respect to the material contained herein or for any
errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional
claims in published maps and institutional affiliations.

This Springer imprint is published by the registered company Springer Nature Switzerland AG.
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
This book is a tribute to the memory of
Professor Enzo Martinelli,
with deep esteem and gratitude.
Neculai S. Teleman
Foreword

Neculai S. Teleman takes the reader on a fascinating expedition exploring the

lands between the smooth and the continuous domains. For a very long time a
continuous function was assumed to be differentiable perhaps with the exception
of a finite or numerable but discrete set of points. The Weierstrass example (1872)
of a continuous but nowhere differentiable function came as a great shock to
mathematicians of the nineteenth century. A similar attitude was prevalent in the
middle of the previous century when a topological manifold was considered to
carry the unique smooth structure inducing the initial topology of the manifold.
Then in 1956 John Milnor showed that on the 7-sphere there are several different
smooth exotic structures. But in 1963 Michael F. Atiyah and Isadore M. Singer
announced their index theorem which asserted that the index of the Laplace operator
(associated with the smooth Riemannian metric) is equal to the topological index
of the smooth manifold. Therefore, the index obtained from the data provided by
the smooth structure is the same as the one provided by the data derived from
the underlying topological structure. These two important and highly non-trivial
results turned attention of many a fine mathematician of the 1960s to the regions
where the smooth meets the continuous. A few years later (1965) S. P. Novikow
published his famous result that the rational Pontrjagin classes of a smooth manifold
are topological invariants. By the end of the decade, it was known that rational
Pontrjagin classes can be defined for any topological manifold (R. Kirby and L.C.
Siebelmann, 1969).
In 1971, I.M. Singer in Future extensions of index theory and elliptic operators
wrote
I was asked to speak on “possible future developments, give conjectures, and speculate
about future advances.” . . . . probably at this very moment a graduate student is busily at
work on a theorem that might change present trends drastically. However, in a limited way,
I will try and fulfill the charge – at least within the context of index theory for elliptic
operators. My main theme is this: by and large this theory assigns an integer to an elliptic
operator on a compact smooth manifold; I expect extensions of the present theory to new
situations; i.e., non-smooth manifolds, non-manifolds of special type, and to a context
where it is natural that integer be replaced by a real number.

vii
viii Foreword

A few years later Neculai S. Teleman arrived at MIT and in 1977 defended his
Ph.D. thesis in which he extended the index theorem to combinatorial manifolds.
Ever since he dedicated all his efforts to developing new tools and results which
permit to extend the validity of the index theorem such as combinatorial Hodge
theory on PL manifolds, (TAMS, 1979, Inv. Math., 1980), “elliptic machinery” and
signature operators on Lipschitz manifolds (Publ. IHES, 1983). The following year
he published a version of the index theorem for general topological manifolds using
new tools, i.e. Kasparov’s realisation of the analytic K-homology group K0 (M)
and Fredholm operators on non-compact manifolds. His quest to generalise various
“smooth” constructions led him to use the existence of the quasiconformal structure
on any topological manifold which finally turned his attention to the techniques and
ideas of non-commutative geometry (1991).
The reader of the book is asked to embark together with the author on this
fascinating journey which leads to the boundaries of our topological knowledge.
Neculai S. Teleman introduces the reader step by step to the basic and then advanced
tools of this fascinating theory: classical differential geometry, e.g. Riemannian
metrics, Laplacian, Hodge theory and the smooth version of the index theorem. Then
to go beyond that the basic elements of non-commutative geometry are presented
as a method of delocalisation of our mathematical perception and natural extension
of the classical geometrical tools. Having covered the known regions of geometry
and topology of manifolds, the knowledgeable and experienced guide shows new
fascinating vistas of non-commutative topology.
It is a very important volume which gathers and presents in an orderly manner
the story of the index theorem. The tools and the results are scattered in numerous
publications. But having them in one place is not the only advantage of this book. We
get much more since the narrative is both objective and passionate at the same time
as the narrator is one of the principal protagonists in this fascinating mathematical
epic.

Kraków, Poland Robert Wolak

April 2019
Preface

This book combines some of the author’s research contributions in index theory
with elements of non-commutative geometry. The book ends with non-commutative
topology.
This work shows that the index formula is a topological statement; on the other
hand, this book gives a re-formulation of the index formula. The re-formulation
should have important consequences.
From the non-commutative geometry side, this book explains how differ-
ent results in index theory, obtained by classical geometry methods (Riemann–
Roch, Hirzebruch [32], Atiyah and Singer [53], Teleman [83, 84]) and by non-
commutative geometry (Connes [82], Connes and Moscovici [96], Connes et
al. [101]), are connected. This has to do with the possibility, offered by different
theories, to exhibit/describe the topological index of elliptic operators. The passage
from classical differential geometry to non-commutative geometry, exemplified by
the index formula done in this book, describes also the way how the author became
interested in, and approached, non-commutative geometry. For this reason, this book
does not intend to give a complete review of all its results or to give a complete list
of its contributors. This book offers the reader a natural path starting with basic
problems of differential geometry and leading to non-commutative geometry and
topology. However, although this book does not intend to present the very significant
developments of non-commutative geometry in all directions, it accompanies the
reader towards present research problems in index theory.
This book has multiple goals:
1. The author did his Ph.D. with Prof. I. M. Singer (1977) working on extending the
index theorem to combinatorial manifolds—a problem that was formulated by I.
M. Singer in his programme article “Prospects of Mathematics” Symposium,
March 16–18, Princeton [58]. One goal of this book is to fill the existing gap
in the mathematical literature in reporting the developments which occurred in
index theory after I. M. Singer presented his 1970-research programme. This
report is certainly not complete.

ix
x Preface

2. It aims to provide a friendly introduction to non-commutative geometry. It studies

index theory from a classical differential geometry perspective up to the point
where classical differential geometry methods become insufficient.
3. Another goal is to present non-commutative geometry as a natural continuation
of classical differential geometry. This is based on the following concepts:
• Non-commutative geometry is abstract index theory.
• Non-commutative geometry is a de-localised theory. In many cases the de-
localisation occurs by replacing C∞ -homomorphisms by operators on L2 .
This change replaces exact relations by L2 -local averages.
• Non-commutative geometry is a multi-distribution theory. The multi-
distribution theory occurs by means of two processes: (1) polarisation of
algebraically higher order expressions; (2) each factor of the polarisation
becomes a mono-distribution theory.
• Hochschild homology—and its variants—replaces the de Rham homology.
4. Beginning with Riemann and up to the present times, index theory has passed
through most chapters of mathematics: algebraic geometry, analytic geometry,
differential geometry, combinatorial geometry, Lipschitz and quasi-conformal
geometry as well as Lp -structures geometry. Another goal of this work is to
show that while the objects of non-commutative geometry of a (classical) space
X live in all powers Xp of X, classical differential geometry objects associated
with X live on the main diagonals of those powers. The possibility of describing
an object classically depends on whether the non-commutative object has a
classical limit, i.e. whether its restriction along the main diagonals are defined;
this depends on whether certain products of distributions can be performed.
The book has the following main mathematical objectives:
(i) it intends to provide a natural link between classical differential geometry and
non-commutative geometry.
(ii) it presents the basic algebraic structure and results which stay at the founda-
tions of non-commutative differential geometry.
(iii) it summarises the basic analytical structures on topological manifolds.
(iv) it shows that the index theorem is a topological statement.
(v) it summarises the results in index theory obtained with non-commutative
geometry methods—local index theorems.
(vi) it proposes prospects in index theory.
(vii) it proposes non-commutative topology.
This book presents the mathematical facts in their chronological order most of
the time. Both the de Rham isomorphism theorem and the Hodge decomposition
theorem show that real cohomology may be extracted analytically from differen-
tiable structures on topological manifolds. These results constitute two of the basic
links between differentiable structures and topological structures. Given, however,
that not all topological manifolds possess differentiable structures, and that the
same topological structure might be compatible with non-equivalent differentiable
Preface xi

structures, there is a fundamental question to ask whether these relations could

be derived purely in topological terms, on all topological manifolds. S. Novikov
proved the fundamental result [41] stating that the rational Pontrjagin classes
of smooth manifolds are topological invariants. Kirby-Siebenmann results [56]
complete Novikov’s theorem by stating that all topological manifolds possess
rational Pontrjagin classes. Going along the same line of ideas, the following
fundamental question arose: whether the Hodge decomposition theorem and index
formula could be stated an all topological manifolds (see e.g., [58]).
This fundamental question was answered using Sullivan’s fundamental result
[75], which states that any topological manifold of dimension = 4 possesess
a Lipschitz or quasi-conformal structure, unique up to Lipschitz/quasi-conformal
isotopy. The first results on Hodge theory on combinatorial manifolds are due
to Teleman [72, 76–78] and to Cheeger [79] on pseudo-manifolds. These results
are successively extended using Sullivan’s result on Lipschitz manifolds [83, 84].
Successive work due to Donaldson and Sullivan [92], Connes et al. [101] extends the
index formula on quasi-conformal manifolds. This book summarises these results.
The first problem the author of this book undertook in his Ph.D. thesis was
extending the Hodge decomposition theorem from smooth manifolds to pseudo-
manifolds, see Teleman [72] and Singer [58]. At that time (1976) this subject was
totally unexplored. In fact, I.M. Singer’s programme [58], published in 1971, poses
the following two questions too:
A “If M is a P L manifold, the L-polynomials are still well defined (Thom [25]),
and one can define from them the rational Pontrjagin classes. The Hirzebruch
signature theorem still holds. Is there an associated elliptic operator (as in the
smooth case) whose index is the signature? On what spaces does it operate? Does
it have a symbol and where does the symbol lie?”
B “. . . Are present analytic techniques strong enough to prove the Hodge theorem
for smooth manifolds by approximations via combinatorial Laplacians associated
with smooth triangulations? If so, this might shed some light on the P L-
problem.”
Looking retrospectively, it is important to remark that extending the Hodge
isomorphism theorem and the index formula to spaces which are not differentiable
manifolds was a daring task; the proof of the Hodge isomorphism theorem (see de
Rham [86]) and the proof of the Atiyah–Singer index formula (see Palais seminar
[42] and Atiyah and Singer [53]) heavily used the differentiable structure.
Combinatorial manifolds are the closest relatives of smooth manifolds. For this
reason, the first problem addressed in trying to extend the index formula to more
general contexts was to investigate whether Hodge theory could be extended from
smooth manifolds to combinatorial manifolds.
Given that the Laplace operator is defined in terms of a chosen Riemannian
metric on the manifold, it was crucial to understand in which respect the special
characteristics of the Riemannian metric should be satisfied if a Hodge type result
is desired. The de Rham [86] proof of the Hodge decomposition theorem showed
that if one wishes to extend the Hodge theory to combinatorial manifolds in such a
xii Preface

way that the main identities of the smooth Hodge theory be preserved, see formula
(1.198), the combinatorial Riemannian metric should satisfy the property that the
local measure around each point x of the manifold (the total local solid measure of
the simplices containing the point x) should be equal to the value of the solid angle
about any point in the Euclidean space of the same dimension.
The following theorem answers this question.
Theorem 1 (Teleman [72, 76]) Let M be a combinatorial manifold of dimension
n. Let g be a combinatorial Riemannian metric on M with the property that for any
point x ∈ M, the total local solid measure of the simplices containing the point
x equals the value of the solid angle about any point in the Euclidean space of
dimension n.
Then the metric defines a smooth differentiable structure on M.
The following fundamental result was known.
Theorem 2 (Milnor [66]) There are obstructions to the existence of smoothings of
combinatorial manifolds, the Milnor [66] obstructions.
These two results lead to the conclusion.
Corollary 1
(i) If one wishes to extend the Hodge theory to general structures, it is necessary
to ignore special requirements on the Riemannian metric.
(ii) Riemannian metrics provide generalised smoothings.
The Laplace operator on smooth manifolds is defined exclusively in terms of a
Riemannian metric on the manifold. On the other hand, the de Rham parametrix for
the Laplacian [86] is expressed purely in terms of the geodesic distance function.
Given that the signature operator plays a basic role in different proofs of the
Atiyah–Singer index theorem (in differential geometry and in the Connes–Sullivan–
Teleman and Hilsum index theorem) in non-commutative geometry it is important
to understand the structure of the signature operator. In the smooth category the
signature operator can be described completely in terms of the distance geodesic
function; this is shown in Sect. 1.12.
This book is organised as follows.
Chapter 1 presents those tools of differential geometry which pass directly into
non-commutative geometry. This is necessary to bring the reader from classical
differential geometry topics to non-commutative geometry by providing a natural
link between the two. This chapter ends with the geometry of the signature operator.
Chapter 2 introduces the first elements of non-commutative geometry. This
chapter reconsiders the tools of differential geometry and finds the minimal
structure necessary to make them work in non-commutative geometry. At this
point we are at the boundary between differential geometry and non-commutative
geometry. It shows that the minimal structure leads naturally to a new definition
of differential forms, homology/cohomology and Chern–Weil theory. This is the
way in which Hochschild homology and cyclic homology appear naturally. The
Preface xiii

basic non-commutative geometry topics are presented here: universal algebra, non-
commutative Chern–Weil theory, and Connes’ long exact sequence.
Chapter 3 establishes the first relations of non-commutative differential
geometry. It presents the basic results concerning Hochschild and cyclic
homology/cohomology. It begins with the computation of the Hochschild homology
of the algebra of smooth functions. The result has to be thought of as the link
between differential geometry and non-commutative geometry. Its proof shows how
combinatorics, topology and geometry interact towards the final result. Hochschild,
cyclic and periodic cyclic homology are introduced and their principal properties
are presented. The results due to Connes and Karoubi on the extension of Chern
character to idempotents, extensions of associative algebras, K-theory and K-
homology are presented. In the same section it is shown that the Chern character
may be extracted from direct connections. Chapter 3 also presents the periodic
cyclic homology bi-complex and homology.
Chapter 4 makes a rapid panorama of the analytic structures on topological
manifolds. These analytical structures classify the various fields of geometry and
are significant in index theory: algebraic geometry, analytic geometry, differential
geometry, combinatorial geometry, Lipschitz, quasi-conformal and Lp geometry.
Research interconnecting these fields, however, is not the norm. Global analysis on
combinatorial, Lipschitz, quasi-conformal and Lp manifolds represent a small part
of research. The majority of research is done in the most regular cases. Differential
geometry benefits from the theory of partial differential equations. In spite of this
reality, understanding the roots of fundamental problems in mathematics is of
maximum importance. Index theory is one of them. It began with the Riemann–
Roch theorem on compact Riemann surfaces. It was later extended by Hirzebruch
on analytic manifolds and then by Atiyah and Singer on differentiable manifolds.
Teleman proved that the index of abstract elliptic operators is a topological
invariant. Presenting the hierarchy of analytic structures on topological manifolds
helps the reader to place correctly the various index theorems. Hence, even though
this chapter is short, we have decided to keep it as a separate chapter to provide a
conceptual framework for the reader.
Chapter 5 presents the index theorems which were obtained with classical
differential geometry methods: the Riemann–Roch index theorem, the Thom index
theorem, Hirzebruch index theorem, Atiyah–Singer index theorem and Teleman
index theorem.
Chapter 6 discusses index theorems which were obtained via non-commutative
geometry methods (local index theorem and applications) due to Connes–
Moscovici, Donaldson–Sullivan, Connes–Sullivan–Teleman and Hilsum.
Chapters 7–9 are devoted to prospects in index theory. In this part a new
formulation of index theory is proposed using the newly introduced local structures
and a few conjectures are formulated. More specifically, index theory is defined on
an arbitrary pair (A , J ) where J is an ideal of the localised ring A . A new
definition of the topological and analytic indices is proposed. The new formulation
is based on the T∗loc (A )-groups which replace the classical algebraic K∗ -theory
groups.
xiv Preface

Chapter 10 introduces non-commutative topology. In this part we define non-

commutative spaces, the homology of the new field and the Chern character of
idempotents. This chapter is short because it is entirely new research that has not
been published before. Nonetheless, we feel this work will provide a useful starting
point for future research.
One of the main themes of this book is locality. The analytic structures (algebraic,
analytical and differentiable) are sufficiently rich to allow writing down a formula
for the topological index theorems. This is not the case of more general structures
(combinatorial, Lipschitz or quasi-conformal structures). The possibility to have a
local formula for the topological index depends on whether the non-commutative
topological index has a classical limit.
The book introduces or recalls the following recent or new structures introduced
by the author: micro-localisation of Hochschild homology [107, 121], direct con-
nection [116, 117], local periodic cyclic homology [125], local index formula [130],
localised algebras [125, 129], T-completion comp. [129], topological periodic
cyclic homology [122], (b̃, d) bi-complex [122], Alexander–Spanier co-homology
of localised rings, topological and analytic index, reformulation of index theory,
local periodic cyclic characteristic classes and non-commutative topology.
The contents of this book are self-contained. That said, in order to keep the size of
this book reasonable, a few chapters are compact. In these cases, readers are invited
to deepen their understanding further by reading the cited references. We hope this
book will serve as a useful resource to introduce these concepts in non-commutative
geometry to a broad audience of readers in the fields of mathematics and physics.

Ancona, Italy Neculai S. Teleman

April 2019
Contents

Part I Spaces, Bundles and Characteristic Classes in Differential

Geometry
1 Spaces, Bundles and Characteristic Classes in Differential
Geometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 3
1.1 Differential Forms .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 3
1.2 de Rham Cohomology . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 6
1.3 Alexander–Spanier Cohomology .. . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 6
1.4 Alexander–Spanier Homology.. . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 9
1.5 Vector Bundles and Connections . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 9
1.6 Characteristic Classes . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 11
1.6.1 Chern Classes . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 11
1.6.2 Pontrjagin Classes . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 12
1.6.3 Chern Character . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 13
1.6.4 Chern–Weil Theory .. . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 15
1.6.5 Geometric Construction of the Chern Character . . . . . . . . . . 17
1.6.6 Levi-Civita and Linear Connections of Idempotents . . . . . 18
1.6.7 Levi-Civita Connections . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 23
1.6.8 Infinitesimal Connections . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 25
1.6.9 Characteristic Classes of Lie Algebra Extensions . . . . . . . . 26
1.7 Direct Connections .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 32
1.8 K ∗ -Theory of Banach Algebras . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 36
1.8.1 K ∗ (C 0 (X)) . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 36
1.8.2 K ∗ -Theory of Banach Algebras . . . . . . .. . . . . . . . . . . . . . . . . . . . 41
1.9 Metric Defined Operators .. . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 42
1.9.1 Laplace Operator . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 43
1.9.2 Smooth Hodge Theory . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 44
1.10 Analytic Preliminaries on Operators . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 47
1.10.1 Symbol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 47
1.10.2 Fredholm Operators . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 50
1.10.3 Analytical Index . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 51

xv
xvi Contents

1.10.4 Dirac Operators .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 53

1.10.5 Signature Operators . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 53
1.11 Statement of the Atiyah–Singer Index Theorem .. . . . . . . . . . . . . . . . . . . 55
1.12 K-Homology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 57
1.12.1 Clifford Algebras . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 57
1.12.2 Kasparov K-Homology . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 59
1.12.3 Even and Odd Spectral Triples . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 61
1.12.4 Schatten Class Operators .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 64
1.12.5 n-Summable Fredholm Modules.. . . . . .. . . . . . . . . . . . . . . . . . . . 65
1.13 Geometry of the Signature Operator on Differentiable
Manifolds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 65
1.13.1 1-Parameter Family of Hodge Decomposition
Operators .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 76
1.13.2 The Cut-Off Function . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 79

Part II Non-commutative Differential Geometry

2 Spaces, Bundles, Homology/Cohomology and Characteristic
Classes in Non-commutative Geometry .. . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 85
2.1 Basic Algebra A in Non-commutative Geometry .. . . . . . . . . . . . . . . . . 85
2.2 Bundles in Non-commutative Geometry .. . . . . . .. . . . . . . . . . . . . . . . . . . . 86
2.3 Non-commutative Chern–Weil Theory .. . . . . . . . .. . . . . . . . . . . . . . . . . . . . 86
2.3.1 Preparatory Material: Graded Differential Algebra B̃ . . . . 87
2.3.2 Associative Algebra Extensions and Their Geometry . . . . 89
2.3.3 Connes Chern Character of Algebra Extensions . . . . . . . . . . 91
2.3.4 Karoubi Chern Character of Idempotents . . . . . . . . . . . . . . . . . 97
2.3.5 Connes Chern Character of Spectral Triples . . . . . . . . . . . . . . 104
2.3.6 Pairing n-Summable Fredholm Modules
with K-Theory .. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 109
2.3.7 Pairing of K-Homology with K-Theory—Explicit .. . . . . . 113
2.4 Non-commutative Homology .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 115
2.4.1 The Bar Resolution . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 116
2.4.2 Preliminaries: The Basic Structure . . . .. . . . . . . . . . . . . . . . . . . . 118
2.4.3 Relations Between N, 1 − T , b and b . . . . . . . . . . . . . . . . . . . . 120
2.4.4 Hochschild and Cyclic Homology .. . . .. . . . . . . . . . . . . . . . . . . . 124
2.4.5 Connes’ Exact Sequence . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 126
2.4.6 Chern–Weil Characteristic Classes of Direct
Connections .. . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 128
2.4.7 Non-localised Alexander–Spanier Complex . . . . . . . . . . . . . . 129
2.4.8 Augmentation .. . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 130
2.4.9 The Operator σ . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 132
2.4.10 Universal Graded Differential Algebra . . . . . . . . . . . . . . . . . . . . 137
2.4.11 Non-commutative de Rham Homology and Cyclic
Homology .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 141
Contents xvii

2.4.12 Reduced Cyclic Homology.. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 141

2.4.13 de Rham Homology vs. Reduced Cyclic Homology.. . . . . 149
2.5 Relation Between B and d . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 151
2.6 Non-commutative Cohomology . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 152
2.6.1 Hochschild H H∗ (A ) and Cyclic Hλ∗ (A )
Cohomologies . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 152
2.6.2 Quillen Non-commutative Cohomology, H ∗ (A , V ) . . . . . 154
2.6.3 Non-commutative de Rham Cohomology
with Values in an Algebra, HdR ∗ (A , V ) .. . . . . . . . . . . . . . . . . . 157
2.6.4 Cyclic Co-cycles vs. Closed Graded Trace .. . . . . . . . . . . . . . . 158
2.6.5 Chern Character with Values in Hλev (MN (A ), A ) . . . . . . 162
2.6.6 Chern Character of Idempotents with Value
in Non-commutative de Rham Cohomology . . . . . . . . . . . . . . 163
2.7 Connes Chern Character in K-Homology . . . . . .. . . . . . . . . . . . . . . . . . . . 164
2.8 Connes Karoubi Chern Character in K-Theory.. . . . . . . . . . . . . . . . . . . . 164
2.9 Pairing K-Theory with Alexander–Spanier Cohomology;
Towards Local Index Theorem . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 165
3 Hochschild, Cyclic and Periodic Cyclic Homology .. . . . . . . . . . . . . . . . . . . . 169
3.1 Hochschild Homology of the Algebra of Smooth Functions,
H H∗ (C ∞ (M, R)) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 169
3.1.1 Preliminaries on the Algebra of Smooth Functions .. . . . . . 170
3.1.2 Localisation of Germs Along the Diagonal . . . . . . . . . . . . . . . 171
3.1.3 Micro-Localisation of Hochschild Homology . . . . . . . . . . . . 175
3.1.4 Hochschild Homology of J∗∞ . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 177
3.2 Computation of H H∗ (C ∞ (M)) Using One Derivative . . . . . . . . . . . . 181
3.2.1 The Homology of C̃∗ . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 184
3.3 Division and Hochschild Homology . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 186
3.4 Combinatorics Behind Homology Theories . . . .. . . . . . . . . . . . . . . . . . . . 188
3.4.1 Notation .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 189
3.4.2 Total Difference Operators .. . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 190
3.5 The Karoubi-Type Operator . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 193
3.6 Homological Consequences . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 195
3.7 Pointwise Alexander–Spanier and Hochschild Homology .. . . . . . . . 197
3.8 Derivations.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 200
3.9 Anti-Symmetrisation Map . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 201
3.10 Chevalley–Eilenberg Map . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 202
3.11 Morita Equivalence.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 203
3.12 Connes–Chern Character of Idempotents . . . . . . .. . . . . . . . . . . . . . . . . . . . 206
3.13 S (Chq e) = Chq−2 e . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 206
3.14 Changing the Ground Ring K . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 207
3.15 Separable Algebras .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 208
3.16 Excision .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 210
3.16.1 Shuffle Product . . . .  . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 210
3.16.2 H Hn (A ⊗ B) = p+q=n H Hp (A ) ⊗ H Hq (B) . . . . . 211
xviii Contents

3.17 Tensor Product of Complexes . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 211


3.18 The Cyclic Bi-complex {b, b } and Cyclic Homology .. . . . . . . . . . . . 212
3.19 Computation of H∗λ (C ∞ (M)) . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 215
3.20 H∗λ (A ) .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 217
3.21 Periodic Cyclic Homology .. . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 217
Hev/odd (C ∞ (M)) = HdR
λ,per ev/odd
3.22 (M) . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 220
3.23 Proof of Theorem 3.27 Explicit.. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 222
3.24 Periodic Cyclic Connes–Chern Character of Idempotents . . . . . . . . . 227
3.25 Direct Connections and Chern Character of Idempotents .. . . . . . . . . 228
per,λ
3.26 Connes–Chern Character Ch : K i (A ) −→ Hi (A ) . . . . . . . . . . . 230
per,λ
3.26.1 Chern Character Ch : K 0 (A ) −→ Hev (A ).. . . . . . . . . . 230
per,λ
3.26.2 Chern Character Ch : K 1 (A ) −→ Hodd (A ).. . . . . . . . . . 230

Part III Index Theorems

4 Analytic Structures on Topological Manifolds . . . . . .. . . . . . . . . . . . . . . . . . . . 235
4.1 Topological Manifolds . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 235
4.2 Quasi-Conformal and Lipschitz Manifold Structures . . . . . . . . . . . . . . 236
4.3 Combinatorial and P L Manifold Structures . . . .. . . . . . . . . . . . . . . . . . . . 237
4.4 Differentiable Manifolds .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 238
4.5 Analytic Manifolds .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 239
5 Index Theorems in Differential Geometry .. . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 243
5.1 Riemann–Roch Index Theorem . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 243
5.2 Thom Index Formula .. . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 244
5.3 Thom–Hirzebruch Signature and Index Theorem . . . . . . . . . . . . . . . . . . 245
5.4 Hirzebruch Index Theorem . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 247
5.5 Grothendieck Index Theorem .. . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 248
5.6 Atiyah–Singer Index Formula . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 251
5.7 The Theorem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 252
5.8 Idea of the Proof .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 252
5.9 The Atiyah–Singer Index Formula Holds on Twisted
Signature Operators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 254
5.9.1 Cobordism Invariance of the Analytical Index .. . . . . . . . . . . 254
5.9.2 Hodge Theory and the Analytic Index of Twisted
Signature Operators . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 255
5.9.3 Excision for Twisted Signature Operators .. . . . . . . . . . . . . . . . 255
5.9.4 Cobordism Invariance . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 257
5.9.5 The Analytic and Topological Indexes of Twisted
Signature Operators . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 258
5.9.6 Excisive Triples .. . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 260
Contents xix

5.10 Teleman Index Formula: The Index Formula Is a

Topological Statement . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 267
5.11 Teleman Index Formula on Lipschitz Manifolds . . . . . . . . . . . . . . . . . . . 267
5.11.1 Riemannian Structures: L2 -Cohomology . . . . . . . . . . . . . . . . . 267
p
5.11.2 Strong and Weak Convergence: The Space W1 .. . . . . . . . . . 269
5.11.3 Rellich Lemma on Lipschitz Manifolds . . . . . . . . . . . . . . . . . . . 270
5.11.4 Lipschitz Hodge Theory: The Signature Operator . . . . . . . . 270
5.11.5 The Index of Twisted Signature Operators Is a
Lipschitz Invariant .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 271
5.11.6 The Analytic Index of Dξ+ Is a Lipschitz Invariant . . . . . . . 271
5.11.7 The Index Formula Is a Topological Statement . . . . . . . . . . . 272
5.11.8 Teleman Index Formula on Topological Manifolds .. . . . . . 273
6 Index Theorems in Non-commutative Geometry . . .. . . . . . . . . . . . . . . . . . . . 275
6.1 Connes Moscovici Local Index Theorem . . . . . . .. . . . . . . . . . . . . . . . . . . . 275
6.2 Donaldson, Sullivan Index Theorem . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 278
6.3 Connes, Sullivan, Teleman Index Formula .. . . . .. . . . . . . . . . . . . . . . . . . . 278
6.3.1 Hilsum Index Formula.. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 279
6.4 Index Theorem on Combinatorial Manifolds . . .. . . . . . . . . . . . . . . . . . . . 280
6.4.1 Review of Orthogonal Invariants . . . . . .. . . . . . . . . . . . . . . . . . . . 282

Part IV Prospects in Index Theory

7 Algebraic Structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 287
7.1 Algebraic K∗ -Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 287
7.2 Local Algebraic Structures.. . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 288
7.2.1 K∗ -Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 289
7.2.2 Localised Rings . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 290
7.2.3 Local Alexander–Spanier Cohomology . . . . . . . . . . . . . . . . . . . 291
7.2.4 Local Periodic Cyclic Homology: Long Exact
Sequence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 291
loc,per,λ
7.2.5 Local Chern Character Tiloc −→ Hi .. . . . . . . . . . . . . . 292
7.2.6 Local Index Theorem .. . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 293
7.3 Local Algebraic Ti : Theory .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 294
7.3.1 Introductory Considerations.. . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 295
7.3.2 Algebraic Ti and Tiloc -Theory . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 297
7.3.3 Localised Rings . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 297
7.3.4 Local Mayer–Vietoris Diagrams . . . . . . .. . . . . . . . . . . . . . . . . . . . 298
7.3.5 Preparing the Definition of T0loc (A ) and T1loc (A ) . . . . . . . 300
7.3.6 Definition of T0 (Aμ ) and T0loc (A ) . . . .. . . . . . . . . . . . . . . . . . . . 301
7.3.7 T1 (Aμ ) and T1loc (A ) . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 301
7.3.8 T -Completion .. . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 302
7.3.9 Definition of T1 (Aμ ) and T1loc (A ) . . . .. . . . . . . . . . . . . . . . . . . . 303
7.3.10 Induced Homomorphisms .. . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 304
xx Contents

7.3.11 Constructing Idempotents and Invertible Matrices

Over Λμ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 304
7.3.12 K1 (A ) vs. T1 (A ) . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 313
7.3.13 Connecting Homomorphism

∂ : T1loc (Λ ) −→ T0loc (Λ) ⊗ Z[ 12 ] . . . . .. . . . . . . . . . . . . . . . . . . . 316
7.3.14 Six Terms Exact Sequence . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 317
7.3.15 Relative T -Groups: Ti (Aμ , J ) . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 322
7.3.16 Connecting Homomorphism: Second Form .. . . . . . . . . . . . . . 323
8 Topological Index and Analytical Index: Reformulation of
Index Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 329
8.1 Level I: Index Theory at the T∗loc -Theory Level .. . . . . . . . . . . . . . . . . . . 329
8.2 Level II: Index Theory in Local Periodic Cyclic Homology . . . . . . . 330
8.3 Level III: Index Theory Restricted at the Diagonal .. . . . . . . . . . . . . . . . 331
8.4 Topological Expectations . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 332
8.5 Tiloc (C), i = 0, 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 332
8.5.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 332
8.5.2 Notation: Preliminaries .. . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 332
8.5.3 The Main Result . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 333
9 Local Hochschild Homology of the Algebra of Hilbert–Schmidt
Operators on Simplicial Spaces . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 343
9.1 The Main Result . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 343
9.2 Preliminaries and Notation.. . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 343
9.2.1 The Space: Hilbert–Schmidt Kernels and Operators.. . . . . 343
9.3 Hochschild and Local Hochschild Homology
of Hilbert–Schmidt Operators . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 345
9.4 Algebraic Constructions . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 346
9.4.1 Homotopy Operator s: The Splitting . .. . . . . . . . . . . . . . . . . . . . 347
9.4.2 Homotopy Operator S . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 350
9.4.3 Homological Consequences .. . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 354
9.5 Analytic Considerations.. . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 359
9.5.1 Continuous Hochschild Chains Over the Algebra
of Hilbert–Schmidt Operators.. . . . . . . . .. . . . . . . . . . . . . . . . . . . . 359
9.5.2 Continuity of the Hochschild Boundary .. . . . . . . . . . . . . . . . . . 360
9.5.3 Continuity of the Homotopy Operators s . . . . . . . . . . . . . . . . . . 362
9.5.4 Continuity of the Homotopy Operators S . . . . . . . . . . . . . . . . . 363
9.6 Topological Considerations .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 363
9.6.1 Alexander–Spanier Cohomology . . . . . .. . . . . . . . . . . . . . . . . . . . 363
9.6.2 Alexander–Spanier Homology .. . . . . . . .. . . . . . . . . . . . . . . . . . . . 366
9.6.3 Isomorphism Between { C∗I,loc , b }∗
and {C(∗)
AS (X, G), ∂ AS } , G = R, or C . . . . . . . . . . . . . . . . . . . . 367
∗
9.7 Control of the Supports of Hochschild Chains. .. . . . . . . . . . . . . . . . . . . . 369
Contents xxi

9.8 Local Hochschild Homology of the Algebra

of Hilbert–Schmidt Operators . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 371
9.8.1 Preliminaries .. . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 371
9.8.2 Distance Control of Supports vs. Simplicial
Control: The Result . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 373

Part V Non-commutative Topology

10 Non-commutative Topology . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 377
10.1 The Idempotent Π . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 378
10.2 Topological Hochschild Homology . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 378
10.2.1 Local Topological Periodic Cyclic Homology of
the Algebra of Smooth Functions .. . . . .. . . . . . . . . . . . . . . . . . . . 381
10.3 Local Topological Periodic Cyclic Homology of the
Algebra of Arbitrary Functions on a Smooth Manifold . . . . . . . . . . . . 382
10.4 Characteristic Classes of Idempotents in Non-commutative
Topology .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 382
10.5 Rational Pontrjagin Classes of Topological Manifolds .. . . . . . . . . . . . 386
10.5.1 Existence of Direct Connections on Topological
Manifolds . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 386
10.5.2 The Rational Pontrjagin Chain . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 386
10.5.3 The Rational Pontrjagin Class . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 387

References .. .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 389
Index . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . 395
 Part I
Spaces, Bundles and Characteristic Classes
in Differential Geometry
Chapter 1
Spaces, Bundles and Characteristic
Classes in Differential Geometry

Abstract Part II prepares the reader to see how some of the basic notions of differ-
ential geometry pass into non-commutative geometry. The basic notions presented
in the first chapter are reconsidered in the second chapter from a non-commutative
geometry view point. Differential geometry begins with the algebra A = C ∞ (M)
of smooth functions and builds up by adding multiple structures; classical index
theory uses most of these structures. Non-commutative geometry is abstract index
theory; its axioms comprise many of these structures. While differential geometry
is built by summing up different structures, non-commutative geometry reverses
this process. In differential geometry the commutativity and locality assumptions
are built in by means of the construction of differential forms. There are two
basic differences which summarise the passage from differential geometry to non-
commutative geometry: in differential geometry (1) the basic algebra A = C ∞
is commutative, has true derivations (differential fields), and has a topology—the
Fréchet topology; in non-commutative geometry, the basic algebra A is not required
to be commutative nor to have a topology, nor to have derivations, (2) in differential
geometry, the basic algebra A is used to produce local objects; in non-commutative
geometry the locality assumption is removed. Non-commutative geometry finds
and uses the minimal structure which stays at the foundation of geometry: of
differential forms, product of (some) distributions, bundles, characteristic classes,
cohomology/homology and index theory. The consequences of this discovery are
far reaching.

1.1 Differential Forms

In this section we suppose that M is a paracompact smooth manifold of dimension

n. We denote by C ∞ (M) the unital associative algebra of smooth functions on M.
This is a commutative algebra. We denote by T (M) the tangent bundle to M and by
T ∗ (M) its co-tangent bundle. We denote by Γ (T (M)) the space of smooth sections
of the tangent bundle T (M) and by Ω 1 (M) the space of smooth sections in the co-
tangent bundle T ∗ (M). The following result describes local objects in differential
geometry.

© Springer Nature Switzerland AG 2019 3

N. S. Teleman, From Differential Geometry to Non-commutative Geometry
and Topology, https://doi.org/10.1007/978-3-030-28433-6_1
4 1 Spaces, Bundles and Characteristic Classes in Differential Geometry

Proposition 1.1 (Peetre [26]) Let Q be a C ∞ (M)-module and let F : Q −→ Q

be a linear mapping over the algebra of C ∞ (M).
Then F is a differential operator if and only if F decreases supports. We will say
that F is a local operator. In other words, if

F (f . γ ) = f . F (γ ), (1.1)

for any f ∈ C ∞ (M), and any γ ∈ Q, then for any point p ∈ M, one has

(F (γ ))(p) = F (γ (p)) (1.2)

where γ (p) is a jet.

Theorem 1.1
(i) Γ (T (M)) is the space of derivations of the associative algebra C ∞ (T (M)),
i.e. any X ∈ Γ (T (M)) is an R-linear mapping X : C ∞ (M) −→ R which
satisfies the Leibnitz identity

X( f1 . f2 ) = X(f1 ) . f2 + f1 . X(f2 ). (1.3)

(ii) The differential df of the smooth function f is defined by

(( df )( X ))(p) := ( Xf )(p), (1.4)

for any X ∈ Γ (T (M)) and p ∈ M.

(iii) Γ (T (M)) and Ω 1 (M) are C ∞ (M)-modules.
(iv) d 1 = 0.
(v) Taylor’s formula with reminder and Schwartz’ lemma on the commutativity of
second order mixed partial derivatives, applied onto the smooth function f ,
imply

( df )(p) ∈ C ∞ (M)/Ip2 , (1.5)

where Ip ⊂ C ∞ (M) is the ideal consisting of all smooth functions which

vanish at p.
Definition 1.1 A k-differential form ω ∈ Ω k (M) is an k-multilinear C ∞ (M)-
function

ω : ⊗kC ∞ (M) Γ (T (M)) −→ C ∞ (M), (1.6)

which is skew-symmetric

ω(X1 , . . . , Xi , . . . , Xj , . . . ., Xk ) = − ω(X1 , . . . , Xj , . . . , Xi , . . . ., Xk ).
(1.7)
Skew-symmetry is a form of graded commutativity.
1.1 Differential Forms 5

Definition 1.2 For any ω ∈ Ω p (M) and σ ∈ Ω q (M) define the exterior product,
or wedge product ∧

p!q! 
(ω ∧ σ )(X1 , . . . .Xp+q ) := ω(Xσ (1) , . . . ., Xσ (p) ) · τ (Xσ (p+1) , . . . ., Xσ (p+q) ).
(p + q)!
(ω,τ )∈Sp+q
(1.8)

Definition 1.3 The exterior derivative dω ∈ Ω r (M) ∈ Ω r+1(M) is defined by


r=k+1
dω(X1 , . . . , Xi , . . . , Xj , . . . ., Xk+1 ) := (−1)r Xr ω(X1 , . . . , Xˆr , . . . ., Xk+1 )+
r=1

(−1)i+j ω([Xi , Xj ], X1 , . . . , X̂i , . . . ., Xˆj . . . .Xk+1 ).
1≤i<j ≤k+1
(1.9)

The non-commutative geometry does not assume any of the two requirements
(I) the locality and
(II) graded commutativity.
In non-commutative geometry both the space M and the algebra of smooth functions
C ∞ (M) are replaced by an arbitrary associative, unital or non-unital, commutative
or non-commutative, algebra A .
(i) In non-commutative geometry it is not assumed that the algebra A has true
derivations (for example, the algebra of continuous functions does not have
non-trivial derivations),
(ii) Ip , p ∈ M, may not be used for any individual point p,
(iii) the point-wise Taylor formula may not be used,
(iv) the differential forms and their derivatives need a new definition,
(v) the product of differential forms may not follow the classical definition. It is
not postulated that the product of differential forms is graded commutative.
However, in non-commutative geometry the following properties are
preserved
(vi) the collection I of local ideals Ip , p ∈ M, may be recovered from

I := Ker{μ : A ⊗R A −→ A }, (1.10)

where μ denotes the multiplication in A ; this is going to be used into the

construction of non-commutative differential forms,
(vii) the elements of the algebra A constitute a system of generators of the non-
commutative differential forms,
(viii) the non-commutative differentials d produce complexes, i.e. d 2 = 0,
(ix) the graded Leibnitz formula should hold.
6 1 Spaces, Bundles and Characteristic Classes in Differential Geometry

In differential geometry the locality requirement is too restrictive to allow

making products of distributions of any kind; this difficulty persists at all levels
of regularity below the C 1 -regularity level. In non-commutative geometry the
locality requirement is removed; the problem of multiplying distributions has new
possibilities to be accomplished. Although it is not true that the existing non-
commutative geometry succeeds to deal with the problem of multiplying arbitrary
distributions, it is important to state that it is sufficiently powerful to cover the
basic questions raised by the index formula on topological manifolds, if one
takes advantage of the fundamental result due to Sullivan [75], which affirms
that any topological manifold of dimension = 4 admits an unique Lipschitz or
quasi-conformal structure. Sullivan’s theorem together with the non-commutative
geometry techniques produce results below the classical differential geometry of
C 1 -structures level, see e.g. Connes et al. [101].
Given any associative unital algebra A , we anticipate by saying that the idea to
define a complex, whose chains satisfy the conditions (viii) and (ix) only, leads to
the universal complex U ∗ (A ); this will be discussed in Sect. 2.4.10, see [9, 15].
In non-commutative geometry any differential form associated to the algebra A is
the equivalence class of an element of the quotient of a certain sub-algebra of the
universal complex; in the case of topological algebras the projective tensor product
completion will be used, see Sect. 2.4.8.

1.2 de Rham Cohomology

The vector spaces Ω k (M) together with the co-boundary homomorphisms d form
the de Rham complex ΩdR ∗ (M) of the smooth manifold M, Ω 0 (M) := C ∞ (M).

The de Rham complex is an example of elliptic complex, see Atiyah and Bott
[51].
Theorem 1.2 (de Rham [86]) The homology of the de Rham complex is isomor-
phic to the singular cohomology H ∗ (M, R).
Karoubi [89] extended de Rham homology in the non-commutative context;
more details will be given in Sect. 2.4.13.

1.3 Alexander–Spanier Cohomology

Alexander–Spanier cohomology is a non-commutative geometry object.

We assume that the topological space M is metrisable; we choose a metric d
on M and we use this metric to define t-neighbourhoods Utk (0 < t) of the main
1.3 Alexander–Spanier Cohomology 7

diagonal in M k+1

Utk := {(x0 , x1 , .., xk ) | d(x0 , x1 ) + d(x1 , x2 ) + . . . + d(xk−1 , xk ) + d(xk , x0 ) < t },

(1.11)

for xi ∈ M.
The system of neighbourhoods Utk form a projective net U with respect to t: if
t1 < t2 , we consider Utk1 Utk2 given by restriction to the smaller space is placed in
the rth-position. Let G be an arbitrary Abelian group.
Let

t (G) := {f | f : Ut −→ G, with f arbitrary f unction.}

k k
CAS, (1.12)

For any f ∈ CAS,

k
t (G) and 0 ≤ r ≤ k + 1 define δr f ∈ CAS, t (G)
k+1

(δr f )(x0 , x1 , .., xk+1 ) := f (x0 , .., xˆr , . . . , xk+1 ). (1.13)

This function is well defined because, if (x0 , x1 , .., xk+1 ) ∈ Utk+1 , the triangle
inequality implies (x0 , .., xˆr , . . . , xk+1 ) ∈ Utk+1 .
k+1
Define d : CAS,
k
t (G) −→ CAS, t (G)

d f := (−1)r δr f. (1.14)
0≤r≤k+1

The homomorphism d passes to the projective limit and satisfies d 2 = 0. Let

k
CAS (U, G) = ProjLimt CAS,
k
t (G). (1.15)

∗ (M, G) := {C ∗ (U, G), d} is called Alexander–

The cohomology complex CAS AS
Spanier complex.
Theorem 1.3 (Alexander–Spanier, See [45])
For any locally finite simplicial or CW complex M, the homology of the
Alexander–Spanier complex is canonically isomorphic to the singular cohomology
H ∗ (M, G).
Remark 1.1 In the Alexander–Spanier complex any cohomology k-chain on M
is represented by a non-local function defined on M k ; this is a feature of non-
commutative geometry.
Definition 1.4 The Alexander–Spanier co-chain φ has cyclic symmetry provided

φ(x0 , x1 , . . . , xq ) = (−1)q φ(x1 , . . . , xq , x0 ).

8 1 Spaces, Bundles and Characteristic Classes in Differential Geometry

The Alexander–Spanier co-chain φ is skew symmetric provided

φ(xσ (0) , xσ (1) , . . . , xσ (q) ) = Sign(σ ) φ(x0 , x1 , . . . , xq ), f or any permutation σ.

∗,ω
Proposition 1.2 (See e.g. [96]) Let CAS (M, G) denote the sub-complex of the
Alexander–Spanier complex consisting of all co-chains which have cyclic symmetry
or are skew-symmetric. Then
(i) the inclusion of H ∗,ω (M, G) into H ∗,ω (M, G) induces isomorphisms in
homology.
(ii) If M is a smooth manifold and H ∗,ω,smoot h(M, G) denotes the sub-complex
of H ∗,ω (M, G) consisting of all smooth, skew-symmetric Alexander–Spanier
co-chains, then the inclusion of C ∗,ω,smoot h(M, G) into H ∗,ω (M, G) induces
isomorphisms in homology.
(iii) The mapping

∂k
f ( x0 , x1 , . . . , xk ) → f ( x0 , x1 , . . . , xk ))| diag dx1i1 ∧ dx2i2 ∧ . . . ∧ dxkik
∂x1i1 ∂x2i2 . . . ∂xkik
(1.16)

is a chain homomorphism ξ∗ from the Alexander–Spanier complex to de Rham

complex which induces isomorphisms in homology

ξk,∗ : HAS
k
(M, R) −→ HdR
k
(M). (1.17)

The next observation is useful in non-commutative geometry.

Remark 1.2 Suppose Cat is a category of analytic structures on topological mani-
folds M with the property that the corresponding associative algebra A of functions
on M is a topological algebra with the property that Cat(M k+1 ) is the projective
completion of the algebraic tensor product algebra ⊗k+1
C Cat(M). This means that
for any f ∈ Cat(M k+1 ), the function f is given by the sum of a series
∞

f = f0i ⊗C f1i ⊗C · · · ⊗C fki . (1.18)
i=1

This condition is satisfied, in particular, if Cat = Top or Cat = C ∞ .

Then

dr (f0 ⊗C f1 ⊗C , . . . , ⊗C fk ) = f0 ⊗C f1 ⊗C , . . . , ⊗C 1 ⊗C . . . ⊗C fk , (1.19)

where the factor 1 is placed into the rth-position.

1.5 Vector Bundles and Connections 9

1.4 Alexander–Spanier Homology

We assume that M is a locally finite simplicial complex. We define the vector spaces

t (R) := {μ | μ is a measure on Ut }.
AS k
Ck, (1.20)

If t1 < t2 , let ρ be the restriction of continuous functions defined on Utk2 to the

sub-space Utk1 . Let μ ∈ Ck,
AS
t1 (R). Then the restriction ρ induces the measure ρ∗ (μ)
k
over the space Ut2 . The homomorphisms

ρ∗ : Ck,
AS
t1 (R) −→ Ck, t2 (R)
AS
(1.21)

define an inductive system of vector spaces

CkAS (M, R) := IndLimt Ck,

AS
t (R). (1.22)

By definition, the real Alexander–Spanier homology complex is

C∗AS (M, R) := {C∗AS (M, R), ∂}. (1.23)

and its homology is denoted H∗AS (M, R). For more information on Alexander–
Spanier homology see Massey [73].
Theorem 1.4
(i) The Alexander–Spanier homology is functorially isomorphic to the real singu-
lar homology of M.
(ii) The integration induces a pairing

k
HAS (M, R) ⊗R HkAS (M, R) −→ R. (1.24)


(iii) The pairing coincides with the canonical pairing H k (M, R) ⊗R
Hk (M, R) −→ R via the isomorphism between the Alexander–Spanier
cohomology/homology and the singular counter-parts.

1.5 Vector Bundles and Connections

The purpose of this section is to discuss how the vector bundles, linear connections
and curvature of classical differential geometry pass into non-commutative geome-
try.
A vector bundle η = (E, π, B) of rank N over the real or complex numbers
consists of the total space E, projection π : E −→ B and base B such that each
10 1 Spaces, Bundles and Characteristic Classes in Differential Geometry

fibre π −1 (x), for any point x ∈ B, is a vector space isomorphic to Kn , where

K = R, or C. It is required that the bundle η be locally trivial, i.e. there exists an
open covering U = {Ui } of the base space and homeomorphisms hi : π −1 (Ui ) −→
Ui × Kn which sends each fibre π −1 (x) isomorphically onto the fibre {x} × Kn , for
any x ∈ Ui .
If all spaces in the definition of η are topological spaces, respectively,
Lipschitz/quasi-conformal manifolds, or smooth manifolds, and the mappings π,
hi are, respectively, in the same category, the bundle η will be called a continuous
bundle, respectively, Lipschitz/quasi-conformal or smooth bundle. Denote by Cat
any of these categories.
If the base space is paracompact simplicial or CW -complex of finite dimension,
then the bundle η may be embedded into a product bundle N := B × KN , with N
sufficiently large, see Milnor [66].
We may choose in the product bundle N the product Hermitian, or Euclidean,
structure, depending on whether K = C, or R.
Let ACat be the associative unital algebra

ACat := {f | f : B −→ K, where f is a Cat f unction} (1.25)

Let px : {x} × KN −→ π −1 (x) be the orthogonal projection, for any x ∈ B. The

collection of matrices p(x) produces an N × N matrix p with entries in ACat .
Theorem 1.5
(i) For any embedded bundle η into a product bundle of rank N there corresponds
an N × N matrix p with entries belonging to the algebra ACat .
(ii) The matrix p is an idempotent (p2 = p); if p is produced by pointwise
orthogonal projections, then p is self-adjoint (p∗ = p) and is called a
projector.
(iii) Vice versa, for any idempotent matrix p with entries in the algebra ACat there
corresponds a bundle η such that, up to an isomorphism, its corresponding
idempotent is p
(iv) 1 − p is also an idempotent; if p is a projector, then 1 − p is the orthogonal
projection onto the orthogonal complement η⊥ of the bundle η.
(v) The direct sum p ⊕ (1 − p) decomposes the ACat -free module of rank N into
a direct sum

(ACat )N = Γ (N) = Γ (η) ⊕ Γ (η⊥ ) (1.26)

and therefore the space Γ (η) of sections of the bundle η is a finite projective
ACat -module.
(vi) Vice versa, to any finite projective ACat -module M there corresponds a bundle
η such that, up to an isomorphism, its associated ACat -module is the module
M.
Discovering Diverse Content Through
Random Scribd Documents
Gordon?" She said this as if the problem were his, not hers, and
showed a relief in this transfer of the responsibility.
"I don't know yet," he said. "Whatever you tell me."
"But you must tell her something; you must make her
understand. It won't do for you to hurt the poor girl's feelings."
"Well, I'll just say that I delivered her message and that you
wouldn't come."
"Oh, Gordon! How could you be so cruel? You certainly would
not be so heartless as to say I wouldn't!"
"Well, then, that you couldn't."
"But she would want a reason, and she'd be entitled to one.
What one could you give her? You must think, Gordon, we must
both think, and decide on something that will help you out. What are
you laughing at?"
"Why, Elizabeth," he said, "it isn't my predicament. It's your
predicament."
He leaned back in his chair comfortably, in an attitude of
irresponsibility.
"How can you sit there," Elizabeth said, "and leave it all to me?"
And then she laughed,--and was grave again.
"Of course," she said. "Well--I'm sure I can't solve it. Poor little
Gusta! She was so pretty and so good, and so--comfortable to have
around--don't you know? Really, we've never had a maid like her.
She was ideal. And now to think of her--in prison! Isn't it awful?"
Marriott sat with half-closed eyes and looked at her through the
haze of his lashes. The room was still; the fire burned slowly in the
black chimney; now and then the oil gurgled cozily in the lamp.
"What is a prison like, Gordon? Is it really such an awful place?"
 Marriott thought of the miserable room in the women's quarters,
with its iron wainscoting, the narrow iron bed; the wooden table and
chair, and he contrasted it with this luxurious library of the Wards.
"Well," he said, turning rather lazily toward the fire, "it's nothing
like this."
"But,"--Elizabeth looked up suddenly with the eagerness of a
new idea,--"can't you get her out on bail--isn't that what it's called?
Can't you get some kind of document, some writ?--yes, that's it."
She spoke with pleasure because she had found a word with a legal
sound. "Get a writ. Surely you are a lawyer clever enough to get her
out. I always thought that any one could get out of prison if he had
a good lawyer. The papers all say so."
"You get in prison once and see," said Marriott.
"Mercy, I expect to be in prison next!" Elizabeth exclaimed.
"Prisons! We seem to have had nothing but prisons for a year or
more. I don't know what started it--first it was that poor Harry
Graves, then Archie, and now it's Gusta. And you talk of them and
John Eades talks of them--and I had to see them one night taking
some prisoners to the penitentiary. I'd never even thought of prisons
before, but since then I've thought of nothing else; I've lived in an
atmosphere of prisons. It's just like a new word, one you never
heard before,--you see it some day, and then you're constantly
running across it. Don't you know? It's the same way with history--I
never knew who Pestalozzi was until the other day; never had heard
of him. But I saw his name in Emerson, then looked him up--now
everything I read mentions him. And oh! the memory of those men
they were taking to the penitentiary! I'll never escape it! I see their
faces always!"
 "Were they such bad faces?"
"Oh, no! such poor, pale, pathetic faces! Just like a page from a
Russian novel!"
The memory brought pain to her eyes, and she suffered a
moment. Then she sat erect and folded her hands with
determination.
"We might as well face it, Gordon, of course. I just can't go; you
see that, don't you? What shall we do?"
"You might try your Organized Charities." His eyes twinkled.
"Don't ever mention that to me," she commanded. "I never
want to hear the word. That's a page from my past that I'm
ashamed of."
"Ashamed! Of the Organized Charities?"
"Oh, Gordon, I needn't tell you what a farce that is--you know it
is organized not to help the poor, but to help the rich to forget the
poor, to keep the poor at a distance, where they can't reproach you
and prick your conscience. The Organized Charities is an institution
for the benefit of the unworthy rich." Her eyes showed her pleasure
in her epigram, and they both laughed. But the pleasure could not
last long; in another instant Elizabeth's hands fell to her lap, and she
looked at Marriott soberly. Then she said, with hopeless conviction:
"I just can't go, Gordon."
Before Marriott could reply there was a sense of interruption; he
heard doors softly open and close, the muffled and proper step of a
maid, the well-known sounds that told him that somewhere in the
house a bell had rung. In another moment he heard voices in the
hall; a laugh of familiarity, more steps,--and then Eades and
Modderwell and Mrs. Ward entered the room. Elizabeth cast at
Marriott a quick glance of disappointment and displeasure; his heart
leaped, he wondered if it were because of Eades's coming. Then he
decided, against his will, that it was because of Modderwell. A
constraint came over him, he suddenly felt it impossible that he
should speak, he withdrew wholly within himself, and sat with an air
of detachment.
The clergyman, stooping an instant to chafe his palms before
the fire, had taken a chair close to Elizabeth, and he now began
making remarks about nothing, his clean, ruddy face smiling
constantly, showing his perfect teeth, his eyes roving over Elizabeth's
figure.
"Well! Well! Well!" he cried. "What grave questions have you
two been deciding this time?"
Elizabeth glanced at Marriott, whose face was drawn, then at
Eades, who sat there in the full propriety of his evening clothes, then
at her mother, seated in what was considered the correct attitude for
a lady on whom her rector had called.
"I think it's good we came, eh, Eades?" the clergyman went on,
without waiting for an answer. "It is not good for you to be too
serious, Miss Elizabeth,--my pastoral calls are meant as much as
anything to take people out of themselves." He laughed again in his
abundant self-satisfaction and reclined comfortably in his chair. And
he rolled his head in his clerical collar, with a smile to show Elizabeth
how he regarded duties that in all propriety must not be considered
too seriously or too sincerely. But Elizabeth did not smile. She met
his eyes calmly.
"Dear me," he said, mocking her gravity. "It must have been
serious."
 "It was," said Elizabeth soberly. "It was--the murder!"
"The murder! Shocking!" said Modderwell. "I've read something
about it. The newspapers say the identification of Koerner by that
poor old woman was complete and positive; they say the shock was
such that she fainted, and that he stood there all the time and
sneered. I hope, Eades, you will see that the wretch gets his deserts
promptly, and send him to the gallows, where he belongs!"
"Marriott here doesn't join you in that wish, I know," said Eades.
"No? Why not?" asked Modderwell. "Surely he--"
"He's going to defend the murderer." Eades spoke in a tone that
had a sting for Marriott.
"Oh!" said Modderwell rather coldly. "I don't see how you can
do such a thing, Marriott. For your own sake, as much as anybody's,
I'm sorry I can't wish you success."
"I wish he hadn't undertaken the task," said Eades.
"I'm sure it must be most disagreeable," said Mrs. Ward, feeling
that she must say something.
"Why do you wish it?" said Marriott, suddenly turning almost
savagely on Eades.
"Why," said Eades, elevating his brows in a superior way, "I
don't like to see you in such work. A criminal practice is the
disreputable part of the profession."
"But you have a criminal practice."
"Oh, but on the other side!" said Modderwell. "And we all expect
so much better things of Mr. Marriott."
"Oh, don't trouble yourselves about me!" said Marriott. "I'm sure
I prefer my side of the case to Eades's."
 The atmosphere was surcharged with bitterness. Mrs. Ward
gave a sidelong glance of pain, deprecating such a contretemps.
"And I'm going to try to save him," Marriott was forging on.
"Well," said Eades, looking down on his large oval polished nails,
and speaking in a tone that would finally dispose of the problem,
"for my part, I revere the law and I want to see it enforced."
"Exactly!" Modderwell agreed. "And if there were fewer delays
in bringing these criminals to justice, there would be fewer lynchings
and more respect for the law."
Marriott did not even try to conceal the disgust with which he
received this hackneyed and conventional formula of thoughtless
respectability. He felt that it was useless to argue with Eades or
Modderwell; it seemed to him that they had never thought seriously
of such questions, and would not do so, but that they were merely
echoing speeches they had heard all their lives, inherited speeches
that had been in vogue for generations, ages, one might say.
"I am sure it must be a most disagreeable task," Mrs. Ward was
saying, looking at her daughter in the hope that Elizabeth might
relieve a situation with which she felt herself powerless to deal.
Marriott seemed always to be introducing such topics, and she had
the distaste of her class for the real vital questions of life. But
Elizabeth was speaking.
"I'm sure that Gordon's task isn't more disagreeable than mine."
"Yours?" Mrs. Ward turned toward her daughter, dreading things
even worse now.
"Yes," replied Elizabeth, looking about in pleasure at the
surprise she had created.
"Why, what problem have you?" asked Modderwell.
 "I've been sent for--to come to the prison to see--"
"Not him!" said Modderwell.
Eades started suddenly forward.
"No," said Elizabeth calmly, enjoying the situation, "his sister."
"His sister!"
"Yes," she turned to her mother. "You know, dear; Gusta. She's
been arrested."
"Oh!" exclaimed Mrs. Ward. "Elizabeth! The idea! What
impertinence! Who could have brought such an insolent message!"
She looked at Marriott, as did the others.
"The idea!" Mrs. Ward went on. "Why, I had no notion he was
her brother. To think of our harboring such people!"
Mrs. Ward stiffened in her chair, with glances from time to time
for Marriott and Elizabeth, in an attitude of chilling and austere social
disapproval; then, as if she had forgotten to claim the reassurance
she felt to be certain, she leaned forward, out of the attitude as it
were, to say:
"Of course you sent the reply her assurance deserved."
"No," said Elizabeth in a bird-like tone, "I didn't. What would
you do, Mr. Eades?"
"Why, of course you could not go to a prison," replied Eades.
"But you could, couldn't you? And you do?"
"Only when necessary."
"But you do, Mr. Modderwell?"
"Only professionally," said Modderwell solemnly, for once
remembering his clerical dignity.
"Oh, professionally!" said Elizabeth with a meaning. "You go
professionally, too, Gordon, don't you? And I--I can't go that way. I
can go only--what shall I say?--humanly? So I suppose I can't go at
all!"
"Certainly not," said Mrs. Ward. "How can you ask such a
question?" She was now too disapproving for words. "I can not
consent to your going at all, so let that end it."
"But, Mr. Modderwell," said Elizabeth, with a smile for her
mother, "we pray, don't we, every Sunday for 'pity upon all prisoners
and captives'?"
"That's entirely different," said Modderwell.
"What does it mean,--'I was in prison and ye visited me'?" She
sat with her hands folded in humility, as if seeking wisdom and
instruction.
"That was in another day," said Modderwell. "Society was not
organized then as it is now; it was--all different, of course."
Modderwell went on groping for justification. "If these people are
repentant--are seeking to turn from their wickedness, the church has
appointed the clergy to visit them and give them instruction."
"Then perhaps you'd better go!" Elizabeth's eyes sparkled, and
she looked at Modderwell, who feared a joke or a trap; then at
Eades, who was almost as deeply distressed as Mrs. Ward, and then
at Marriott, whose eyes showed the relish with which he enjoyed the
situation.
"I don't think she wishes to see me," said Modderwell, with a
significance that did not have a tribute for Gusta. No one disputed
him, and there was silence, in which Eades looked intently at
Elizabeth, and then, just as he seemed on the point of speaking to
her, he turned to Marriott and said:
"You certainly don't think that a proper place for her to go?"
 "Oh," said Marriott, "don't refer to me; I'm out of it. I've been, I
brought the message--it's--it's up to Elizabeth."
"Well," said Eades, turning to Elizabeth, "you surely can't be
seriously considering such a thing. You don't know, of course, what
kind of place that is, or what kind of people you would be going
among, or what risks you would be exposing yourself to."
"There would be no danger, would there?" said Elizabeth in her
most innocent manner. "There would be plenty of policemen at
hand, wouldn't there,--in case of need?"
"Well, I don't think you'd willingly elect to go among policemen,"
said Eades.
"Perhaps you three would go with me?" suggested Elizabeth.
"I'd be safe then--all I'd lack would be a physician to make my
escort completely representative of the learned professions."
"The newspaper men would be there," said Eades, "you may be
sure of that, and the publicity--"
At the word "publicity" Mrs. Ward cringed with genuine alarm.
"Do you find publicity so annoying?" asked Elizabeth, smiling on
the three men.
"Elizabeth!" said Mrs. Ward, "I do wish you'd stop this
nonsense! It may seem very amusing to you, but I assure you it is
not amusing to me; I find it very distressing." She looked her
distress, and then turned away in the disgust that was a part of her
distress. "It would be shocking!" she said, when she seemed to them
all to have had her say.
"I'm sorry to shock you all," said Elizabeth meekly. "It's very
kind of you, I'm sure, to act as mentors and censors of my conduct.
I feel sufficiently put down; you have helped me to a decision. I
have decided, after hearing your arguments, and out of deference to
your sentiments and opinions, to--"
They all looked up expectantly.
"--to go," she concluded.
She smiled on them all with serenity; and they looked at her
with that blank helplessness that came over them whenever they
tried to understand her.
 X

Though Elizabeth, as long as Eades and Modderwell were there, had

chosen to satirize her predicament, and had experienced the
pleasure of shocking them by the decision she reached, she found
when they had gone that night, and she was alone in her room, that
it was no decision at all. The situation presented itself in all
seriousness, and she found that she must deal with it, not in any
whimsical spirit, but in sober earnestness. She found it to be a real
problem, incapable of isolation from those artificialities which were
all that made it a problem. She had found it easy and simple
enough, and even proper and respectable to visit the poor in their
homes, but when she contemplated visiting them in the prisons
which seemed made for them alone, and were too often so much
better than their homes, obstacles at once arose. As she more
accurately imagined these obstacles, they became formidable. She
sat by the table in her room, under the reading-lamp that stood
among the books she kept beside her, and determined to think it
out. She made elaborate preparations, deciding to marshal all the
arguments and then make deductions and comparisons, and thus,
by a process almost mathematical, determine what to do. But she
never got beyond the preparations; her mind worked, after all,
intuitively, she felt rather than thought; she imagined herself, in the
morning, going to the police station, confronting the officers, finally,
perhaps, seeing Gusta. She saw clearly what her family, her friends,
her set, the people she knew, would say--how horrified they would
be, how they would judge and condemn her. Her mother, Eades and
Modderwell accurately represented the world she knew. And the
newspapers, in their eagerness for every detail touching the tragedy,
however remotely, would publish the fact! "This morning Miss
Elizabeth Ward, daughter of Stephen Ward, the broker, called on the
Koerner girl. Fashionably dressed--" She could already see the cold
black types! It was impossible, unheard of. Gusta had no right--ah,
Gusta! She saw the girl's face, pretty as ever, but sad now, and
stained by tears, pleading for human companionship and sympathy.
She remembered how Gusta had served her almost slavishly, how
she had sat up at night for her, and helped her at her toilet, sending
delicious little thrills through her by the magnetic touch of her soft
fingers. If she should send for Gusta, how quickly she would come,
though she had to crawl!
And what, after all, was it that made it hard? What had decreed
that she, one girl, should not go to see another girl who was in
trouble? Such a natural human action was dictated by the ethics and
by the religion of her kind and by all the teachings of her church,
and yet, when it was proposed to practise these precepts, she found
them treated cynically, as if they were of no worth or meaning. That
very evening the representatives of the law and of theology had
urged against it!
At breakfast her mother sat at table with her. Mrs. Ward had
breakfasted an hour earlier with her husband, but she had a kindly
way of following the members of her family one after another to the
table, and of entertaining them while they ate. She had told her
husband of Elizabeth's contemplated visit to the prison, and then
had decided to say nothing of it to Elizabeth, in the hope that the
whim would have passed with the night. But Mrs. Ward could not
long keep anything in her heart, and she was presently saying:
"I hope, dear, that you have given up that notion of going to
see Gusta. I hope," she quickly added, putting it in the way she
wished she had put it at first, "that you see your duty more clearly
this morning."
"No," said Elizabeth, idly tilting a china cup in her fingers, and
allowing the light that came through the tall, broad windows to fill it
with the golden luminosity of the sun, "I don't see it clearly at all. I
wish I did."
"Don't you think, dear, that you allow yourself to grow morbid,
pondering over your duty so much?"
"I don't think I'm morbid." She would as readily have admitted
that she was superstitious as that she was morbid.
"You have--what kind of conscience was it that Mr. Parrish was
talking about the other night?" Mrs. Ward knitted the brows that life
had marked so lightly.
"New England, I suppose," Elizabeth answered wearily. "But I
have no New England conscience, mama. I have very little
conscience at all, and as for my duty, I almost never do it. I am
perfectly aware that if I did my duty I should lead an entirely
different life; but I don't; I go on weakly, day after day, year after
year, leading a perfectly useless existence, surrounded by wholly
artificial duties, and now these same artificial duties keep me from
performing my real duty--which, just now, seems to me to go and
see poor little Gusta."
 Mrs. Ward was more disturbed, now that her daughter saw her
duty, than she had been a moment before, when she had declared
she could not see it.
"I do wish you could be like other girls," she said, speaking her
thought as her habit was.
"I am," said Elizabeth, "am I not?"
"Well," Mrs. Ward qualified.
"In all except one thing."
Mrs. Ward looked her question.
"I'm not getting married very fast."
"No," said Mrs. Ward.
Elizabeth laughed for the first time that morning.
"You dear little mother, I really believe you're anxious to get rid
of me!"
"Why, Elizabeth!" said Mrs. Ward, lifting her eyes and then
lowering them suddenly, in her reproach. "How can you say such a
thing!"
"But never mind," Elizabeth went on:

"'If no one ever marries me I sha'n't mind very much;

I shall buy a squirrel in a cage and a little rabbit hutch.
I shall have a cottage in a wood, and a pony all my own,
And a little lamb quite clean and tame that I can take to town.
And when I'm getting really old--at twenty-eight or nine--
I shall buy a little orphan girl and bring her up as mine."
She smiled as she finished her quotation, and then suddenly sobered
as she said:
"I'm twenty-seven already!"
"Who wrote that?" asked Mrs. Ward.
"Alma-Tadema."
"Oh! I thought Mr. Marriott might have done it. It's certainly
very silly."
Nora had brought her breakfast, and the action recalled Gusta
to Elizabeth.
"What did papa say--about my going to the prison?"
"He said," Mrs. Ward began gladly, "that, of course, we all felt
very sorry for Gusta, but that you couldn't go there. He said it would
be absurd; that you don't understand." Mrs. Ward was silent for a
moment, knowing how much greater the father's influence was than
her own. She was glad that Elizabeth seemed altogether docile and
practicable this morning.
"There's a good girl now," Mrs. Ward added in the hope of
pressing her advantage home.
Elizabeth gave a little start of irritability.
"I wish you wouldn't talk to me in that way, mama. I'm not a
child."
"But surely your father knows best, dear," the mother insisted,
"more than--we do."
"Not necessarily," said Elizabeth.
"Why! How can you say so!" exclaimed Mrs. Ward, who bowed
to all authority as a part of her religion.
"Papa takes merely the conventional view," Elizabeth went on,
"and the conventional view is taken without thought."
 "But--surely--" Mrs. Ward stammered, in the impotence of one
who, easily convinced without reasons, has no reasons at command-
-"surely--you heard what Mr. Modderwell and Mr. Eades said."
"Their view is conventional," said Elizabeth, "and proper." She
gave a little curl of her lip as she spoke this last word.
"Well, I'm sure, dear, that we all wish to be proper, and Mr.
Modderwell and Mr. Eades--"
"Oh! Don't quote those two men to me! Two such prigs, such
Pharisees, I never saw!"
Mrs. Ward looked at her daughter in a new horror. "Why,
Elizabeth! I'm surprised--I thought that Mr. Eades especially--"
"Well, don't you think Mr. Eades especially at all! He's not
especially; he thinks he is, no doubt, and so does everybody else,
but they have no right to, and hereafter Mr. Eades can't come here--
that's all!" Her eyes were flashing.
Mrs. Ward ventured no further just then, but presently resumed:
"Think what people would say!"
"Oh, mother! Please don't use that argument. I have often told
you that I don't care at all what people say."
"I only wish you cared more." She looked at Elizabeth helplessly
a moment and then broke out with what she had been tempted all
along to say.
"It's that Gordon Marriott! That's what it is! He has such
strange, wild notions. He defends these criminals, it seems. I don't
see how he can approve their actions the way he does."
"Why, mother!" said Elizabeth. "How you talk! You might think I
was a little child with no mind of my own. And besides, Gordon does
not approve of their actions, he disapproves of their actions, but he
recognizes them as people, as human beings, just like us--"
"Just like us!" exclaimed Mrs. Ward, withdrawing herself wholly
from any contact with the mere suggestion. "Just like us, indeed!
Well, I'd have him know they're not like us, at all!"
Elizabeth saw how hopeless it was to try to make her mother
understand Marriott's attitude, especially when she found it difficult
to understand it herself.
"Just like us, indeed!" Mrs. Ward repeated. "You are certainly
the most astonishing girl."
"What's the excitement?"
It was Dick, just entering the room. He was clean-shaved, and
glowing from his plunge, his face ruddy and his eyes bright. He was
good-humored that morning, for he had had nearly five hours of
sleep. His mother poured his coffee and he began eating his
breakfast.
"What's the matter, Bess?" he asked, seizing the paper his
father had laid aside, and glancing at it in a man's ability to read and
converse with women at the same time.
"Why, she threatens to go to the jail," Mrs. Ward hastened to
reply, in her eagerness for a partizan in her cause. "And her father
and Mr. Modderwell and Mr. Eades have all advised her that it would
be improper--to say nothing of my own wishes in the matter."
Dick, to his mother's disappointment, only laughed.
"What do you want to go there for? Some of your friends been
run in?"
"Yes," said Elizabeth calmly.
 "That's too bad! Why don't you have Eades let 'em out,--you
certainly have a swell pull with him."
"You have just had Mr. Eades's opinion from mama."
"Who is your friend?"
"Gusta."
Dick's face was suddenly swept with scarlet, and he started--
looked up, then hastily raised his coffee-cup, drained its last drop,
flung his napkin on his plate, and said:
"Oh, that girl that used to work for us?"
"Yes."
"Well, mother's right."
Mrs. Ward looked her gratitude.
"Of course, you can't go."
"I can't?"
He had risen from the table, and Elizabeth's tone impressed
him.
"Look here," he said peremptorily. "You just can't go there,
that's all there is about it!"
"Why not?"
"Because you can't. It wouldn't do, it wouldn't be the thing; you
ought to know that."
"But why?" Elizabeth persisted. "I want a reason."
"You don't mean to say you seriously consider it?" asked Dick in
real alarm.
"Yes, I do."
Dick suddenly grew excited, his eyes flamed, and he was very
red.
"Look here, Bess," he said. "You just can't, that's all."
 "Can't I?" she said, and she gave a little laugh. It was not her
usual pleasant laugh.
"No, you can't." He spoke more than insistently, he spoke
angrily. He snatched out his thin gold watch and glanced at it. "I've
not got time to discuss this thing. You just can't go--that's all there is
to it."
Elizabeth rose from the table calmly, went out of the room, and
Dick, after a hesitant moment, ran after her.
"Bess! Bess!"
She stopped.
"See here, Bess, you must not go there to see that girl. I'm
surprised! She isn't the sort, you understand! You don't know what
you're doing. Now look here--wait a minute!" He caught her by the
arm. "I tell you it's not the thing, you mustn't!"
He was quite beside himself.
"You seem greatly excited," she said.
He made a great effort, controlled himself, and, still holding her,
began to plead.
"Please don't go, Bess!" he said. "Please don't!"
"But why--why?" she insisted.
"Because I say so."
"Humph!"
"Because I ask it. Please don't; do it for me, this once. You'll be
sorry if you do. Please don't go!"
His eyes were full of the plea he was incoherently stammering.
He was greatly moved, greatly agitated.
"Why, Dick," she said, "what is the matter with you? You seem
to take this trifle very much to heart. You seem to have some special
interest, some deep reason. I wish you'd tell me what it is. Why
shouldn't I go to see poor Gusta? She's in trouble--she was always
good to me."
There was a sudden strange wild expression in his face, his lips
were slightly parted. The moments were flying, and he must be off.
"Oh, Bess," he said, "for God's sake, don't go!"
He implored her in his look, then snatching out his watch ran to
the hall, seized his hat and top-coat, and went out, flinging on his
coat as he ran, and leaving the door flying wide behind him.
Elizabeth stood looking after him. When she turned, her mother was
in the room.
"What can be the matter with Dick?" said Elizabeth. "I never
saw him so excited before. He seemed--" She paused, and bit her
lip.
"Well, my dear," said Mrs. Ward calmly, "you see now, I hope,
just how the world regards such a wild action. It was his love and
respect for his sister, of course."

XI

"No, don't say anything more. I've thought it all out; my duty's clear
now, I must go." Elizabeth laid her hand on her father's shoulder,
and though he turned from the great desk at which he sat in his
private office, he hesitated. "Come on."
"That conscience of yours, Bess--" he began, drawing down the
lid of his desk.
 "Yes, I know, but I can't help it."
"How did you decide at last to go?" asked Ward, as they walked
rapidly along in the crowded street.
"Well, it tortured me--I couldn't decide. It seemed so difficult,--
every one--mama, our dear Modderwell, Mr. Eades, Dick--he nearly
lost his reason, and he did lose his temper--thought it impossible.
But at last I decided--"
"Yes?"
"--just to go."
Elizabeth gave a little laugh at this not very illuminating
explanation.
"I didn't know what the proprieties were," she went on. "Our
little code had not provided rules--what to wear, the chaperonage,
and all that, you know. And then,"--she abandoned her irony,--"I
thought of you."
"As a last resort, eh?" said Ward, looking fondly into her face,
flushing behind her veil in the keen November air. She drew close to
him, put her hand on his arm.
"Yes," she said, "and as a first resort, as a constant, never-
failing resort."
She gave his arm a little squeeze, and he pressed her hand to
his side in silence.
"Do you know where it is?" Elizabeth asked presently.
"Oh, yes, I was there once."
"When?"
"When that boy of mine was arrested--Graves."
"Yes, I remember."
 "I wonder," he said after a pause, and he paused again at the
question he seemed to fear--"whatever became of him!"
She had never told him of that day at the charity bureau; she
wondered if she should do so now, but she heard him sigh, and she
let it pass.
"Yes," he went on as if she had been privy to his rapid train of
thought, "I suppose such things must be; something must be done
with them, of course. I hope I did right."
At the Central Station they encountered a young policeman,
who, when he saw Ward, evidently recognized him as a man of
affairs, for he came forward with flattering alacrity, touching his
helmet in the respect which authority always has ready for the rich,
as perhaps the real source of its privilege and its strength. The
young policeman, with a smile on his pleasant Irish face, took Ward
and Elizabeth in charge.
"I'll take yez to the front office," he said, "and let yez speak to
the inspector himself."
When McFee understood who Ward was, he came out instantly,
with an unofficial readiness to make a difficult experience easy for
them; he implied an instant and delicate recognition of the
patronage he saw, or thought it proper to see, in this visit, and he
even expressed a sympathy for Gusta herself.
"I'm glad you came, Mr. Ward," he said. "We had to hold the
poor girl, of course, for a few days, until we could finish our
investigation of the case. Will you go up--or shall I have her brought
down?"
"Oh, we'll go up," said Ward, wondering where that was, and
discovering suddenly in himself the usual morbid desire to look at
the inmates of a prison. The sergeant detailed to conduct them led
them up two broad flights of stairs, and down a long hall, where, at
his step, a matron appeared, with a bunch of keys hanging at her
white apron. Elizabeth went with none of the sensations she had
expected. She had been surprised to find the police station a quiet
place, and the policemen themselves had been very polite, obliging
and disinterested. But when the matron unlocked one of the doors,
and stood aside, Elizabeth felt her breast flutter with fear.
The sergeant stood in the hall, silent and unconcerned, and
when the matron asked him if he would be present at the interview
he shook his head in a way that indicated the occasion as one of
those when rules and regulations may be suspended. Ward, though
he would have liked to go in, elected to remain outside with the
sergeant, and as he did this he smiled reassuringly at Elizabeth, just
then hesitating on the threshold.
"Oh, just step right in," said the matron, standing politely aside.
And Elizabeth drew a deep breath and took the step.
She entered a small vestibule formed of high partitions of
flanged boards that were painted drab; and she waited another
moment, with its gathering anxiety and apprehension, for the
matron to unlock a second door. The door opened with a whine and
there, at the other end of the room in the morning light that
struggled through the dirty glass of the grated window, she saw
Gusta. The girl sat on a common wooden chair that had once been
yellow, her hat on, her hands gloved and folded in her lap, as if in
another instant she were to leave the room she somehow had an air
of refusing to identify herself with.
 "She's sat that way ever since she came," the matron
whispered. "She hasn't slep' a wink, nor e't a mouthful."

"She's sat that way ever since she

came"
 Elizabeth's glance swept the room which was Gusta's prison, its
walls lined higher than her head with sheet-iron; on one side a
narrow cot, frowsy, filthy, that looked as if it were never made,
though the dirty pillow told how many persons had slept in it--or
tried to sleep in it. There was a wooden table, with a battered tin
cup, a few crusts and crumbs of rye bread, and cockroaches that
raced energetically about, pausing now and then to wave their
inquisitive antennæ, and, besides, a cheap, small edition of the
Bible, adding with a kind of brutal mockery the final touch of squalor
to the room.
Gusta moved, looked up, made sure, and then suddenly rose
and came toward her.
"I knew you'd come, Miss Elizabeth," the girl said, with a relief
that compromised the certainty she had just expressed.
"I came as soon as I could, Gusta," said Elizabeth, with an
amused conjecture as to what Gusta might think had the girl known
what difficulties she had had in getting there at all.
"Yes," said Gusta, "thank you, I--"
She blushed to her throat. They stood there in the middle of
that common prison; a sudden constraint lay on them. Elizabeth,
conscious of the difficulty of the whole situation, and with a little
palpitating fear at being in a prison at all--a haunting apprehension
of some mistake, some oversight, some sudden slip or sliding of a
bolt--did not know what to say to Gusta now that she was there. She
felt helpless, there was not even a chair to sit in; she shuddered at
the thought of contact with any of the mean articles of furniture,
and stood rigidly in the middle of the room. She looked at Gusta
closely; already, of course, with her feminine instinct, she had taken
in Gusta's dress--the clothes that she instantly recognized as being
better than Gusta had ever before worn--a hat heavy with plumes, a
tan coat, long and of that extreme mode which foretold its early
passing from the fashion, the high-heeled boots. Her coat was open
and revealed a thin bodice with a lace yoke, and a chain of some
sort. An odor of perfume enveloped her. The whole costume was
distasteful to Elizabeth, it was something too much, and had an
indefinable quality of tawdriness that was hard to confirm, until she
saw in it, somehow, the first signs of moral disintegration. And this
showed in Gusta's face, fuller--as was her whole figure--than
Elizabeth remembered it, and in a certain coarseness of expression
that had scarcely as yet gone the length of fixing itself in lines.
Elizabeth felt something that she recoiled from, and her attitude
stiffened imperceptibly. But not imperceptibly to Gusta, who was a
woman, too, and had an instant sense of the woman in Elizabeth
shrinking from what the woman in her no longer had to protect itself
with, and she felt the woman's rush of anger and rebellion in such a
relation. But then, she softened, and looked up with big tears. She
had a sudden yearning to fling herself on Elizabeth's breast, but
leave was wanting, and then, almost desperately, for she must
assert her sisterhood, must touch and cling to her, she seized
Elizabeth's hand and held it.
"Oh, Miss Elizabeth," she said, "I oughtn't to 'av' sent for you. I
know I had no right; but you was always good to me, and I had no
one. I've done nothing. I've done nothing, honest, honest, Miss
Elizabeth, I've done nothing. I don't know what I'm here for at all;
they won't tell me. And Archie, too, it must have something to do
with him, but he's innocent, too. He hasn't done nothing either.
Won't you believe me? Oh, say you will!"
She still clung to Elizabeth's hand, and now she pressed it in
both her own, and raised it, and came closer, and looked into
Elizabeth's face.
"Say you believe me!" she insisted, and Elizabeth, half in fear, as
though to pacify a maniac, nodded.
"Of course, of course, Gusta."
"You mean it?"
"Surely I do."
"And you know I'm just as good as I ever was, don't you?"
"Why--of course, I do, Gusta." It is so hard to lie; the truth, in
its divine persistence, springs so incautiously to the eyes before it
can be checked at the lips.
The tears dried suddenly in Gusta's blue eyes. She spoke
fiercely.
"You don't mean it! No, you don't mean it! I see you don't--you
needn't say you do! Oh, you needn't say you do!"
She squeezed Elizabeth's hand almost maliciously and Elizabeth
winced with pain.
"You--you don't know!" Gusta went on. And then she hesitated,
seemed to deliberate on the verge of a certain desperation, to pause
for an instant before a temptation to which she longed to yield.
"I could tell you something," she said significantly.
A wonder gathered in Elizabeth's eyes. Her heart was beating
rapidly, she could feel it throbbing.
"Do you know why I sent for you--what I had to tell you?"
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!

textbookfull.com