Full download textbook at textbookfull.

com

From Differential Geometry to Non commutative

Geometry and Topology Neculai S. Teleman

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

Download more textbook from https://textbookfull.com

More products digital (pdf, epub, mobi) instant
download maybe you interests ...

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.
Another random document with
no related content on Scribd:
through the booking office on to No. 1 platform, at which stood the
5.5 train for Bristol, which stops for the first time at Reading. He
showed the ticket collector a third-class single ticket for Reading,
and the man hastily opened a carriage door and pushed him in.
“Near shave that, sir!” he remarked.
The passenger was far too short of breath to do anything but
nod his head in reply. He sank down exhausted upon the seat as the
guard blew his whistle and the train drew out of the station. There
were only two men beside himself in the carriage, and they, with
true British indifference, took no further notice of him. The train had
nearly reached Reading when he walked down the corridor to the
lavatory. He did not return to the carriage, but stepped Straight from
the corridor to the platform as the train came to rest. He mingled
unnoticed with the stream passing the barriers. He noticed that the
hands of the clock pointed to ten minutes to six.
Meanwhile Mr. Pargent had stood for an instant upon the refuge
in the middle of Praed Street, a puzzled expression on his face. An
approaching taxi-driver noticed his swaying form and swerved
sharply, just in time to avoid him as he crashed forward into the
road-way. The taxi-driver’s action caused a sudden check to the
traffic, and for an instant there was a chaos of skidding vehicles,
screeching brakes, and blasphemous language. Then the nearest
drivers jumped from their seats and clustered round the prone form.
A majestic policeman, setting a calm and undeviating course through
the tumult, knelt down and turned the body over. Mr. Pargent was
dead.
The newspaper which Mr. Ludgrove bought at Wokingham station
and read carefully during his journey to Waterloo contained nothing
more than a bare statement that the famous poet, Richard Pargent,
had fallen dead in Praed Street, and that, upon the body being
conveyed to the mortuary, a blade was found in it, with the point
entering the heart. Round these facts a formidable structure of
conjecture and reminiscence was built up. There was an obituary
notice, in which Mr. Pargent’s Christian names and the titles of his
published works were incorrectly stated. There was a résumé of the
case of Mr. Tovey, and a photograph of Praed Street, with heavy
black crosses marking the places where the murders had taken
place. Finally there was a leading article, in which the responsibility
for the murders was ingeniously fixed upon the Government, which
in consequence came in for severe criticism. In fact, the murder of
Mr. Pargent was evidently the topic of the day.
Mr. Ludgrove sat with his haversack beside him on the carriage
seat, even his beloved plants unheeded. He read every bit of the
paper over and over again, seeking for some clue which might
account for the crime. The description ended with the words “the
police are, however, in possession of a clue, the nature of which they
are not at present prepared to divulge, but which, they are
confident, will lead to startling developments in the near future.” Mr.
Ludgrove smiled rather cynically as he read them. They seemed to
him to have a somewhat familiar ring.
He had the paper in his hand as he walked into Mr. Copperdock’s
shop to get his key, and his friend caught sight of it. “So you’ve seen
the news, then?” he said excitedly.
Mr. Ludgrove nodded gravely. “I have been reading the account
as I came up in the train,” he replied.
“We didn’t know a thing about it till eight o’clock on Saturday
evening,” continued Mr. Copperdock eagerly. “I was just going across
to the Cambridge Arms when Inspector Whyland came in and asked
me casual-like what I’d been doing all the afternoon. As a matter of
fact neither Ted nor I had hardly left the shop, and so I told him. It
wasn’t till then that he let on that three hours before a gentleman
had been murdered not a hundred yards away, exactly the same as
poor Tovey. Then I remembered that a customer had told me
something about an accident outside Paddington station, but, being
busy, I hadn’t taken any heed of it. And, that reminds me, the
Inspector left a message for you, asking if you would let him know
when you got back. I’ll ring him up from here if you like, I’ve got his
number.”
“I wish you would, I have no telephone, as you know,” replied Mr.
Ludgrove. “Has Mrs. Cooper brought my key back yet?”
Mr. Copperdock felt in his pocket and handed the herbalist a Yale
key. “She brought it in ten minutes ago. She was terribly upset, and
talked about our all being murdered in our beds next. It’s a terrible
thing to have happened just before Christmas like this. People will be
afraid to go out of doors after dark.”
Mr. Ludgrove nodded rather absently, and took the key which his
friend offered him. “You’ll let Inspector Whyland know I’m back?” he
said, and, without awaiting Mr. Copperdock’s reply, he left the shop,
crossed the road, and entered his own premises, a thoughtful frown
upon his face.
Inspector Whyland lost no time in acting upon Mr. Copperdock’s
message. The murder of Mr. Richard Pargent, following so soon upon
that of Mr. Tovey, and perpetrated by the same method, had made a
great sensation, not only among the public, but, which was far more
important to the Inspector, at Headquarters. He had been given a
pretty direct hint that unless he could find some clue within the next
few days, the case would be taken out of his hands and given to
some more capable officer. All the machinery at the disposal of the
police had been put into action, but so far without the slightest
result. And in Inspector Whyland’s despair it seemed to him that the
only hope left of gaining any information was through the herbalist,
with his peculiar inner knowledge of the inhabitants of the district.
He arrived to find Mr. Ludgrove busily engaged in writing up
descriptions of his botanical trophies in a large manuscript book. The
herbalist greeted him warmly, and invited him to a seat in the best
chair. The two sat for a moment in silence, until the Inspector spoke
abruptly.
“Look here, Mr. Ludgrove, what do you know about this man
who’s been killed?”
“Richard Pargent?” replied the herbalist quietly. “I know nothing
about him personally. I have seen his name mentioned once or twice
as a writer of verse, but I doubt if I should have remembered it had
it not been recalled to me by what I read in the papers this
morning.”
“Do you know anything about this poetry of his?” continued the
Inspector.
Mr. Ludgrove smiled. “I am no judge of poetry,” he replied. “But I
can hardly imagine that it was bad enough to inspire anyone with a
desire to murder him.”
Inspector Whyland shrugged his shoulders impatiently. “That’s
just it!” he exclaimed. “There doesn’t seem to have been anything
about the man to provide a motive for his murder. Look here, Mr.
Ludgrove, you’re not the sort of man to go round spinning yarns to
your friends. I’ll make a bargain with you. You get to know all sorts
of things that never come to our ears. Tell me your honest opinion
about these murders, and I’ll tell you what we’ve found out so far.”
For a moment Mr. Ludgrove made no reply. When at length he
spoke it was with a deep note of earnestness in his voice. “I am
honoured by your confidence, Inspector. As you have guessed, I
have been very much concerned by these murders. I have feared
that they would be traced to one or other of my clients, some of
whom, I regret to say, have little respect for the law. I know much of
the inner history of Praed Street and its neighbourhood, and I regret
to say that much of it is almost incredibly sordid. I refer of course to
its underworld, and not to its respectable inhabitants, who are
greatly in the majority. But of two facts I can assure you. The first is
that in all my experience I have never heard of a crime committed
by any of the class to which I refer except for some specific purpose,
either gain or revenge, and the second is that none of them would
resort to murder except perhaps in a sudden access of passion.”
The Inspector nodded. “That’s very much my own experience of
regular crooks,” he replied. “But you say you’ve thought a lot about
this business. Haven’t you any theory of your own?”
“The only theory I can form is that the murders are the work of
some irresponsible maniac,” said Mr. Ludgrove quietly.
“Would an irresponsible maniac have put a piece of poisoned
glass on the end of Colburn’s pipe?” countered the Inspector quickly.
“No, I am afraid that theory won’t do, Mr. Ludgrove.”
“I think it is the only theory which fits in with the assumption
that these three men were all murdered by the same hand,” replied
the herbalist.
“Yet I don’t think you can get away from that assumption,” said
Inspector Whyland. “I don’t mind telling you something that worries
me a bit, since it’s bound to come out at the inquest. One of the first
things we found when we searched Pargent’s pockets in the
mortuary was this.”
He put his hand in his pocket, then held it out with extended
palm toward Mr. Ludgrove.
The herbalist leant forward with an exclamation of surprise. “Ah,
a white counter with the figure III drawn on it in red ink!” he
exclaimed. “Do you know, Inspector, I wondered if anything of the
kind had been found. That is really interesting.”
“Interesting!” grumbled the Inspector. “Aye, the reporters will
find it interesting enough, you may bet. We couldn’t keep it dark,
even if we wanted to. This fellow Pargent got it by post on Saturday
morning. He told his sister, Miss Clara, that it must be a secret token
from some admirer of his work. He was so pleased about it that he
took it down to Penderworth to show his other sister. It was on his
way back from there that he was killed. Oh yes, it’s interesting
enough. We shall have to say that Tovey and Colburn got counters
like this, and then there’ll be a fearful outcry against us for keeping
it dark. ‘Had the police not displayed this criminal reticence, the
victim would have understood the purport of the warning and
adequate measures could have been taken to protect him.’ Oh yes, I
know what’s coming all right.”
Inspector Whyland gazed savagely at the counter for a few
seconds, then continued abruptly.
“Fortunately the envelope in which this one came had been kept.
The counter was wrapped up in a piece of cheap writing paper, and
put in a square envelope of the same quality. The address was
typewritten. Our people at the Yard are trying to trace the machine it
was done with. I wish them joy of the job. They can tell the make all
right; but there must be thousands of them scattered through
London.”
“If you don’t mind my suggesting it, it doesn’t follow that
because Mr. Pargent received a counter he was murdered by the
same hand that killed Mr. Tovey or Mr. Colburn,” said the herbalist
quietly.
Inspector Whyland glanced at him quickly. “No, it doesn’t exactly
follow,” he agreed. “But we’ll return to that later. I want to show you
something else first. Look here.”
This time he produced from his pocket a paper packet, which he
proceeded to unroll. From it he withdrew two sinister-looking blades,
each stained a dull brown.
“These are the knives with which Tovey and Pargent were killed,”
he said. “Don’t touch them, but tell me if you can tell one from the
other.”
Mr. Ludgrove adjusted his spectacles, and bent over the knives.
They were exactly similar, thin, with an extremely sharp point, about
half an inch wide and eight inches long. Neither of them showed any
sign of having ever possessed a handle.
“No, I can see no difference between them,” agreed Mr.
Ludgrove. “I admit that they strengthen the supposition that Mr.
Tovey and Mr. Pargent died by the same hand. But, admitting that
this was the case, it only bears out my theory that the murders were
the actions of a maniac. If they were not, if both men were killed
with some definite motive, you have to discover some connection
between them. Yet they moved in entirely distinct orbits, and, I
should imagine, had nothing whatever in common.”
Inspector Whyland shook his head with a tolerant smile.
“Theory’s all very well in its way, Mr. Ludgrove,” he said, as he
replaced the knives in his pocket. “I can’t waste my time establishing
connections. There’s evidence enough here to convince a jury that
the same man killed all three of them. And I don’t fancy it’ll be very
long before I lay hands upon him.”
“I sincerely hope your expectations will be realized, Inspector,”
replied the herbalist, in a tone of faint irony which was lost upon
Whyland. “I suppose it is indiscreet to ask whether you have any
suspicions? I confess that for my part I am entirely at a loss. What
about young Snyder, for instance?”
“Master Wal Snyder is out of this,” returned the Inspector shortly.
“He was pinched a week ago for picking pockets in a tube lift, and is
safely under lock and key. But I don’t mind telling you in confidence,
Mr. Ludgrove, that I know the murderer lives somewhere in this
district.”
 Mr. Ludgrove lifted his eyebrows in real or assumed
astonishment. “Indeed?” he exclaimed. “That is indeed a great step
forward. Pray, how did you come to that conclusion?”
If the Inspector perceived the irony in his tone, he paid no heed
to it. “Well, it’s pretty obvious,” he replied. “All three men were
murdered in Praed Street, to begin with. Then, there’s another thing.
The envelope in which the counter was sent to Pargent bore the
post-mark of this district, London, W.2. You can’t get away from it,
everything points to the murderer living somewhere about here.”
Again Inspector Whyland paused. Then suddenly he rose and
stood over the herbalist where he sat in his chair. “There’s another
thing, Mr. Ludgrove,” he said, almost menacingly. “You said just now
that it didn’t follow that because Pargent received a counter he was
killed by the same man that murdered the other two. Do you see
where that leads you? If the third counter was sent by a different
hand, it can only have been sent by one of the very few who knew
of the receipt of the first two. And all of them live in this district, you
will remember.”
Mr. Ludgrove gazed up at him in mock alarm. “They do, indeed,”
he replied. “It makes my blood run cold to think that I am one of
them. Really, I am quite relieved to think that I have an alibi in the
case of this last murder, at least.”
Inspector Whyland laughed shortly. “Oh, for that matter, I’m as
much under suspicion as you,” he said. “But, you see, there are
some people we know who couldn’t produce an alibi for the time
when any of the murders were committed.”
And with that he turned abruptly on his heel and strode out of
the room.
 Chapter VII.
The Black Sailor
That afternoon Inspector Whyland wrote in his report: “I had a
long interview with Ludgrove. He is a shrewd fellow, and not easy to
get anything out of. I am pretty sure that he has no definite
knowledge on this matter, but he admits himself that some of his
customers are not above petty crime, and it seems to me that he is
the most likely man to hear any underground rumours that may be
flying about.
“I managed to convey to him a pretty broad hint that I suspected
one of his friends. I did this with a double motive. On the one hand,
if he thinks that Copperdock or one of the others had anything to do
with it, he will probably pass the hint on, and somebody may try to
make a bolt for it, which would give them away. On the other hand,
if he is convinced of their innocence, he will be all the more anxious
to pass on to me any information he may get and which may tend to
clear them. It is just a chance, but it may lead to something useful.”
The inspector’s strategy was indeed that of forlorn hope. Rack his
brains as he would, interrogate as he might everyone who could
possibly throw light upon the murder of Richard Pargent, he found
himself up against a stone wall. Nobody could swear to having seen
and recognized the victim from the time he entered the train at West
Laverhurst until he had been seen to stagger and fall on the refuge
in Praed Street. The ticket collector at Reading was almost positive
about him, but, as he explained, he saw so many passengers in the
course of his day’s work that he could not be sure of any particular
one, unless there was something very remarkable about him or her.
All he knew for sure was that if this was the gentleman, he was
alone in the carriage when the train left Reading.
 Nobody at Paddington had noticed him among the streams
flowing from the arrival platforms, and it was impossible to establish
whether or not he had left the station alone. The taxi-driver who had
seen him fall had not seen the blow struck, could not even say
whether anyone else had been standing on the refuge with him or
not. It was a nasty, skiddy evening, and dark at that. He had enough
to do to watch the traffic without worrying about people on refuges.
It wasn’t until he had seen the poor man stagger into the road
almost under his front wheels that he had even noticed him. And
then his attention was fully occupied in pulling up and trying to avoid
collisions.
Finally, neither of the Miss Pargents could throw the faintest
glimmer of light upon the affair. Their brother was not the type of
man to have secrets from his family. Both sisters appeared to know
the inner history of every moment of his life since infancy, and
insisted upon dispensing the knowledge at great length. As to his
having an enemy in the world, the thing was absurd. He was a very
retiring man, who lived only for his work, and had never sought
adventure beyond the narrow bounds of the family circle. Inspector
Whyland came away from his last interview with the sisters—for Miss
Margaret had been prevailed upon to come up to London—convinced
that it would have been impossible for the dead man to have
embarked upon any clandestine enterprise without the knowledge of
one or other of them.
He was equally unsuccessful in finding any link between Richard
Pargent and any of the inhabitants of Praed Street. Their names
were utterly unfamiliar to either of the sisters, and Mr. Pargent’s only
connection with Praed Street appeared to be that he occasionally
passed along it in a bus. Inspector Whyland was convinced that the
same hand that had struck down Mr. Tovey, had killed Richard
Pargent, and probably Mr. Colburn, though the receipt of the counter
was the only link in the latter case. But what could be the motive for
the murder of these three men, so entirely disconnected from one
another? And, if one adopted the theory of a homicidal maniac, the
difficulty was scarcely diminished. Homicidal maniacs either kill
indiscriminately, or, as in the classic case of Jack the Ripper, they
attack a certain class or type. Yet the receipt of the numbered
counters implied some process of definite selection. Upon what
possible grounds could such a selection be based?
The inquest helped the authorities not at all. Nothing beyond
what Inspector Whyland already knew was elicited, and the only
fresh sensation which it provided was the disclosure of the fact that
each of the three victims had received a numbered counter.
Whyland’s forecast of the results of this were proved to have been
only slightly exaggerated. Most of the newspapers of the day
following the inquest appeared with leading articles discussing the
wisdom of official secrecy upon such a matter. A large number of
practical jokers despatched counters bearing the number IV to their
friends, many of whom arrived hot-foot at Scotland Yard demanding
instant police protection. The fatal counter became for the moment
a national symbol. The Opposition newspapers contained cartoons
depicting the Prime Minister opening an envelope labelled “Public
Opinion,” and containing a numbered counter. An enterprising
Insurance Company issued a poster upon which appeared an
enormous representation of the counter, and under it the words,
“You need not fear this if you are insured with the Gigantic.” To find
one in one’s Christmas pudding was an omen of bad luck for the
ensuing year. Then, having played its part as a nine days’ wonder,
the vogue of the numbered counter ceased as suddenly as it had
begun.
Meanwhile, the police had been by no means idle. To Inspector
Whyland it had always seemed that the focus of the whole business
had lain in Mr. Copperdock’s shop. Mr. Copperdock had been an
intimate friend of Mr. Tovey’s; Ted Copperdock was on familiar terms
with Dick Colburn. It had been through Mr. Copperdock that he had
first learned of the receipt of the numbered counters; the pipe which
had been responsible for Mr. Colburn’s death been bought at Mr.
Copperdock’s shop. Inspector Whyland began to regard the three
victims as represented by the extremities of the letter Y, which,
although they are not connected to one another, are each connected
to a central point. But Mr. Copperdock was the central point, this
graphic representation was incomplete until the connection between
Richard Pargent and Mr. Copperdock could be established. And, try
as he would, Inspector Whyland could find absolutely no link
between the two.
Nevertheless, Whyland determined to make a very thorough
investigation into Mr. Copperdock’s habits and associates. He was
rewarded by the discovery of two very interesting facts. A
representative of Scotland Yard, announcing himself as an agent of
the Planet Typewriter Company, the manufacturers of the machine
owned by Mr. Copperdock, and used by his son for writing business
letters, called at the tobacconist’s shop, ostensibly for the purpose of
inspecting the machine. He succeeded in carrying away with him a
specimen of its work, which was submitted to experts. They
reported that, although there were no perceptible peculiarities in any
of the letters which would enable them to identify this specimen with
the address on the envelope sent to Richard Pargent, there was, on
the other hand no discrepancy between the two types, and that
therefore the envelope might well have been typed on Mr.
Copperdock’s machine.
This, though disappointingly inconclusive, was to Whyland’s ideas
a point to be remembered should any corroborative evidence come
to light. Among the circle in whose actions he found himself
interested, the Copperdocks were the only people who possessed a
typewriter at all. As a matter of fact, Mr. Copperdock had only
purchased it quite recently, at the instance of Ted, who while stoutly
maintaining that no respectable tradesman could be without one,
had been suspected by his shrewd father of less disinterested
motives. Certainly it had seemed only natural that, the machine once
purchased, the expert Ivy should come in of an evening and show
Ted how to work it; but Mr. Copperdock had a way of winking at his
son behind her back which made that ingenuous youth blush most
embarrassingly.
The second curious fact was discovered by Inspector Whyland
almost by accident. In his study of Mr. Copperdock and his
associates, he had taken to frequenting the saloon bar of the
Cambridge Arms, carefully choosing such times as Mr. Copperdock
was busy in the shop. There was nothing in any way suspicious
about the Cambridge Arms; it lay in one of the streets which run
southwards from Praed Street, and was consequently almost hidden
in a backwater. Its remarkably cosy little saloon bar was patronized
chiefly by the neighbouring small tradesmen, and its habitués were
all well-known to the proprietor, with whom Whyland soon became
on confidential terms.
A couple of days after the inquest on Richard Pargent, Whyland
turned into the Cambridge Arms just as its doors were opened, at
five o’clock. The proprietor met him with a smile, and, having
executed his order, leant over the bar and entered into conversation.
“I never thought, when I opened up last Saturday, that that poor
fellow was being murdered at that very minute,” he began. “Just
about five o’clock, wasn’t it, sir?”
Whyland nodded. “The clock was striking five as they picked him
up,” he replied.
“I remember the evening well,” continued the proprietor. “Though
it wasn’t until nigh upon seven that I heard anything about it. I says
to my wife it was that wet and foggy that even our regulars wouldn’t
be in till later. But there you are, you never can tell. I hadn’t opened
this bar, not more than five minutes, when in comes one of my best
customers, Sam Copperdock, the tobacconist. You may have met
him here, sir?”
Whyland nodded non-committally, and the landlord proceeded.
“I was a bit surprised to see him, sir, because Sam doesn’t
usually come along until eight, unless it happens to be Thursday,
when he closes early. ‘Hullo, Sam, you’re on time to-night,’ I says,
jocular like. ‘Oh, I’ve only come in for a drink,’ he says; ‘The bottle’s
empty at home, and I couldn’t scrounge one from old Ludgrove
opposite, he’s away.’ He has a double, and goes back to his shop.
The bar was pretty nigh empty till about seven, when a fellow comes
in and tells me about the murder.”
Inspector Whyland deftly turned the conversation. He knew that
if he questioned the man his suspicions would be aroused, and this
he was anxious to avoid. Without any appearance of haste he
finished his drink and left the premises. Once round the corner, he
walked swiftly to the spot where Richard Pargent had been
murdered, counting his paces as he went.
Could Mr. Copperdock have committed the murder? It was quite
possible. If he had left his shop a few minutes before five, he would
have had plenty of time to reach the exit from Paddington station,
plunge the knife into his victim, and be at the Cambridge Arms at
five minutes past. But, if he were the criminal, a thousand puzzling
questions presented themselves. How did he know that his victim
would be arriving at Paddington at 4.55 that evening? Above all, why
should Mr. Copperdock, the tobacconist, have any grudge against Mr.
Pargent, the minor poet? And then again, Whyland was convinced
that the murderer of Richard Pargent had been the murderer of
James Tovey. But, at the moment when Tovey had been killed, Mr.
Copperdock was in his sitting-room with his son and the murdered
man’s daughter. The riddle appeared to be insoluble.
Whyland turned abruptly, boarded a bus, and took a ticket to
Oxford Circus. Here he dismounted, and turned into the first picture-
house which presented itself. He had discovered long ago that this
was the surest way of securing freedom from interruption. Here for a
couple of hours he sat motionless, neither seeing the pictures nor
attentive to the strenuous efforts of the orchestra. And as he sat, a
new theory slowly unfolded itself in his brain.
Samuel Copperdock, whom he had studied so carefully, did not
appear to him to be a deliberate criminal. He was not of the type
which harbours revenge, and he did not seem either clever or
painstaking enough to work out the details of a premeditated
murder. Although, if it were true that he had formed a design to
marry Mrs. Tovey, the murder of Mr. Tovey might be said to be
advantageous to him, it was inconceivable that he had had any hand
in it. He could not have struck the blow himself, and, since he was
not a rich man, it was difficult to see what inducement he could offer
to a hired assassin. In all human probability, Mr. Copperdock was
innocent of the murder of James Tovey.
But, on the other hand, this murder, followed so closely by the
mysterious death of his other acquaintance, Colburn, had made a
great impression upon him. He had thought and talked of little else,
and, as it happened, he had known every detail. He had seen the
knife with which James Tovey had been killed; he, in common with
very few others, had known of the receipt of the marked counters.
Suppose that the crimes had had such a powerful effect upon his
imagination that he had felt an irresistible impulse to imitate them?
Such things were known, were in fact a commonplace of the criminal
psychologists.
Upon this assumption, the rest was easy. Mr. Copperdock had
learnt by some accident that this man Richard Pargent would arrive
at Paddington at 4.55 on Saturday, and had selected him as a likely
victim. His imitative faculty fully developed, he had sent him the
numbered counter, and had purchased a knife similar to the one
which he had already seen. Then, just before the train was due, he
had left his shop, met his victim, committed the crime, had a drink
to steady his nerves, and then come home.
The theory was plausible, but Inspector Whyland knew well
enough that so far he had not a fragment of real evidence with
which to support it. But, unless he were to adopt the suggestion of
Mr. Ludgrove, that some homicidal maniac was responsible, Whyland
could see no alternative. The lack of motive was so extraordinarily
puzzling. The murders had been utterly purposeless, and the only
possible theory could be one which took full account of this fact. And
then again, how account for the despatch of the numbered
counters? If the murders had been inspired by irresponsible impulse,
what had been their object?
Still deep in thought, Inspector Whyland left the picture-house
and walked slowly back to the police station through the less
frequented streets. It was a little past eight on a typical December
evening, with a light wet mist which magnified the outlines of the
passers-by and covered the pavements with a shining dampness.
But the weather never had much effect upon Whyland, and, beyond
a passing reflection that on such an evening murder in the streets of
London was not, after all, an enterprise presenting any great
difficulties to a determined man, he paid no great attention to it.
He reached the police station without adventure, ordered a
modest supper, and sat down to deal with the mass of reports which
awaited him. He was thus engaged when a sergeant entered the
room, with the message that a man giving the name of Copperdock
wished to speak to him on the telephone. He rose, walked to the
instrument, and picked up the receiver. “Inspector Whyland
speaking,” he said coldly.
“Is that you, Inspector?” came an excited voice, which Whyland
recognized at once. “Can you come and see me at once? I’ve
something very important to tell you.”
Whyland hesitated for a moment. Whatever it was that Mr.
Copperdock had to say to him, a visit to his house could afford him
an excellent opportunity for keeping his eyes open for some clue in
support of his theory. “Yes, I’ll come along straight away,” he replied.
With a word to the sergeant, he left the station and walked rapidly
to Praed Street.
It was about a quarter to ten when he arrived and rang the bell
outside Mr. Copperdock’s shop. The tobacconist was obviously
awaiting him, and the door was opened immediately. With a brief
word of greeting, Mr. Copperdock led the way upstairs to the sitting-
room. Then, shutting the door carefully behind him, he turned
dramatically, “I’ve seen that there black sailor!” he exclaimed in a
hoarse whisper.
Inspector Whyland looked at him sharply. When the reward had
originally been offered, he had been overwhelmed by a flood of
people who had claimed to have seen the “black sailor,” but of late
this elusive figure had ceased to occupy the popular imagination.
Was this a vindication of his theory, showing that Mr. Copperdock’s
brain was completely under the influence of the crime?
“You have seen the black sailor? Sit down and tell me all about it,
Mr. Copperdock,” he replied gravely.
But Mr. Copperdock was far too excited to sit down. “I see him
plain as I see you now,” he exclaimed. “Just after I left the
Cambridge Arms——”
But Inspector Whyland put out his hand and forced him gently
into the nearest chair. “Now, look here, Mr. Copperdock, if your
information is to be of any use to me, you must tell it to me in the
proper order. Begin by telling me how you spent the evening.”
 “Soon as I shut up shop at eight o’clock, I goes off to the
Cambridge, same as I always does,” replied the tobacconist
protestingly. “Ted went out too, he’s taking Ivy to the pictures, and
they won’t be back for an hour or so yet. I has a couple there,
sitting in front of the fire, and it wasn’t until nigh on closing time
that I left. There wasn’t anybody there as was coming home my
way, so I starts off home alone. And I hadn’t gone more than a
hundred yards when I runs slap into the black sailor!”
“What do you mean when you say you ran into him?” interrupted
Whyland.
“Just what I says. I saw someone come round the corner, and
before I could move we ran into one another. He was a great big
fellow, and I could see him perfectly plain, we was right under the
lamp. He bent down and looked straight at me, and you could have
knocked me down with a feather. It was the black sailor, all right. He
was muffled up in a great heavy coat, but I could see his beard, and
his woollen cap, and a great red scar down the right side of his face.
He glared at me as though he were going to hit me, and I hollered
out, being sure he was going to do me in. Then all at once he turns
his back and walks off at a tremendous pace. After a bit I follows
him, hoping to meet a copper. But he suddenly turns in to the
Edgware Road, and by the time I gets there I’d lost him.”
“H’m,” said Inspector Whyland. “That must have been nearly an
hour ago now.” He paused for a moment, then, realizing that in Mr.
Copperdock’s present excited state it would be impossible to get
anything coherent out of him, he rose and took his leave.
“Well, good night, Mr. Copperdock,” he said abruptly. “It’s a pity
you didn’t catch your man. I’d better get back to the station and
warn my people to look out for him.”
As he left the shop, he noticed that Mr. Ludgrove’s door stood
open, and as he crossed the road, the herbalist himself appeared in
the entrance. “Good evening, Inspector,” he said politely.
Obeying a sudden impulse, Whyland stopped and returned the
greeting. “Good evening, Mr. Ludgrove,” he replied. “May I come in
for a moment?”
 “By all means, come along,” replied the herbalist, leading the way
into his sanctum. “I do not drink much myself, but perhaps you will
take some refreshment yourself?”
He busied himself with a bottle and a glass, while the Inspector
drew up a chair to the blazing fire. Then, when his host had seated
himself, Whyland glanced at him with a slight smile playing about his
lips.
“Our friend Mr. Copperdock has seen the black sailor,” he
remarked.
“Seen the black sailor!” exclaimed the herbalist incredulously.
“When and where—if I may ask?”
“Listen, and I’ll tell you all about it,” replied Whyland. “About an
hour ago outside the Cambridge Arms.”
Mr. Ludgrove laughed softly at something in the tone of the
Inspector’s voice, then suddenly became serious.
“I do not think that I should place too great importance upon this
incident,” he said. “Mr. Copperdock is a most excellent person, and I
have the greatest respect for him. But I have often suspected that
his visits to the Cambridge Arms have—shall I say, a pernicious
effect upon the accuracy of his observations.”
“He seemed a bit excited, certainly,” replied Whyland. “But he
was quite positive that he had seen the man.”
Mr. Ludgrove shook his head. “Now, this is strictly between
ourselves, Inspector,” he said. “I make an invariable practice of
taking a walk every evening, as soon as my clients give me the
opportunity. This evening I went out soon after eight, and walked in
the direction of Hyde Park. As it happened, I was returning past the
Cambridge Arms, when I saw a figure which was unmistakably that
of Mr. Copperdock leave the entrance of that house. I hastened my
steps to overtake him, when suddenly he stopped, shouted, and
after a moment or two, set off in the direction of Edgware Road. I
could not imagine what was the matter, since but for us two the
road was deserted as far as one could see for the mist. I was within
twenty or thirty yards of him, and I certainly saw no black sailor.”
It was Inspector Whyland’s turn to smile significantly. “It does
not altogether surprise me to learn that Mr. Copperdock suffers from
hallucinations,” he said.
 Chapter VIII.
At No. 407, Praed Street
Although the unexplained murders which had taken place in
Praed Street were soon forgotten by the general public, their shadow
hung heavily over the neighbourhood in which they had been
committed. The fact that the three deaths had taken place within
comparatively few hundred yards of one another could not fail to
have its effect upon the local imagination. A very noticeable change
came over the usual cheerful and careless life of Praed Street. The
evening pavements were no longer blocked by a strolling, noisy
crowd. Women and children were rarely to be seen abroad after
dark. Even men, traversing the street upon their lawful occasions,
had a way of keeping close to the inner side of the pavements, and
crossing the road upon the approach of any unfamiliar form.
Christmas passed in this atmosphere of intangible fear; the old
year died and the new year came in with a welcome spell of clear
and frosty weather. But, in spite of the fact that no further tragedies
occurred, the shadow still lay heavily upon the district. Ludgrove,
listening to the whispered and entangled stories of his clients,
became more and more certain that, even in the hidden depths of
the underworld, there was no knowledge of the agency by which the
crimes had been committed.
Had such knowledge existed, it must inevitably have been
divulged to him. The police, under the direction of Inspector
Whyland, were engaged in passing a fine-toothed comb through the
Paddington district, and the minor offenders disturbed in this process
were as concerned as a colony of ants unearthed by a spade. Mr.
Ludgrove was visited furtively late at night by anxious people
seeking advice how to conceal the evidence of their misdemeanours
from the prying eyes of the police. He questioned each of these
closely, but the more he did so the more he became convinced that
none of them had the slightest inkling of the perpetrator of the
murders.
Another section of his clients, however, equally furtive and
mysterious, had clues in plenty, which they seemed to think entitled
them to some reward. These, after assuring themselves that no one
could overhear them, would produce an incoherent story of how one
of their neighbours must be the criminal. He had been heard to utter
threats that he would do some one in some day, he had been in
Praed Street on the night when Mr. Tovey was murdered, in the
neighbourhood of Paddington station when Mr. Pargent was killed.
Others, again, had met a muffled figure brandishing a knife, or had
seen a dark man with a beard standing in the shadow thrown by a
projecting wall. Two or three searching questions were always
sufficient to prove that their suspicions were baseless.
The murders in Praed Street were thus peculiar in the annals of
crime. In nearly every case of a crime being committed, there are
others besides the criminal who know all the facts. The police know
this, but their great difficulty is to secure evidence sufficiently
convincing to lead to a conviction. There is a certain esprit de corps
in the constant strife between the professional criminal and the
police, and those who know are careful to keep their knowledge to
themselves. But in this case, had there been any of the inhabitants
of the district in the secret, Mr. Ludgrove would have obtained some
hint of it. It seemed conclusive that the criminal was either working
alone and independently, or came from some other district.
This was the state of affairs towards the end of January. The
police were completely baffled; Inspector Whyland, although he still
favoured the theory he had evolved in the picture-house, had failed
to find any evidence upon which he could act. He hardly knew
whether to attribute Mr. Copperdock’s story of his meeting with the
black sailor to pure hallucination, or to an attempt to divert suspicion
from himself. In either case, it could be made to fit in with the
theory that he had murdered Richard Pargent under the influence of
what the psychologists called an imitative complex. Whyland
redoubled his activities, and a most accurate watch was kept upon
Mr. Copperdock’s movements. A month had elapsed since the
murder of Richard Pargent when Mr. Jacob Martin, the prosperous
wine merchant of the Barbican, opening his morning post in his
comfortable office, came upon a typewritten envelope, marked
“Private and Confidential.” He ripped it open with the paper cutter
which lay upon his desk, and unfolded the letter it contained. It also
was typewritten, on plain paper. Only the signature “John Lacey”
was in ink.
Mr. Martin, a portly, grey-haired man of between fifty and sixty,
read the letter through twice, with a gathering frown upon his face.
He had prospered exceedingly since the day when he had first set
up in business for himself as a wine merchant, so much so that his
competitors wondered at his success. There had, from time to time
been rumours, quickly suppressed, to the effect that Mr. Martin had
other means of livelihood than his ostensible business. But Mr.
Martin was a remarkably astute man, and had hitherto managed to
avoid undue inquisitiveness.
And now, like a bolt from the blue, came this extraordinary letter.
Confound John Lacey, and his prying habits, whoever he might be.
That little cellar under the back office! How well he remembered it.
It had been the scene of many most profitable transactions, had
harboured treasures for which the police of two continents had
searched in vain. For Mr. Martin was a receiver of stolen goods, not a
mere general practitioner of the art, but a specialist whose services
were utilized only by the aristocracy of thieves, and whose particular
function was turning into cash only the most valuable jewellery.
With an exclamation of annoyance Mr. Martin turned once more
to the letter. “Dear Sir, I venture to address you upon a subject
which will no doubt interest you,” it ran. “I have recently acquired a
lease of the premises No. 407, Praed Street, which I understand,
were in your occupation up till some fifteen years ago. Since you
vacated them, these premises have been occupied by a clothier, who
did not require the extensive cellarage, which was boarded up. In
the course of certain alterations which I am having made to the
premises, the entrance to the cellars has once again been opened.
In the course of my investigations I have made a most interesting
discovery in the small cellar beneath the back office.
“I should perhaps report this discovery to the authorities, but I
have thought it best to consult you upon the matter before doing so.
This letter should reach you by the first post to-morrow (Saturday)
morning. I should be glad if you could make it convenient to meet
me on the premises at 2 p.m. that day, when the workmen who are
at present employed there will have gone. I can then show you my
discovery, and we can consult upon the most suitable steps to be
taken in the matter. Should you not find this convenient, I shall
assume that the subject does not interest you, and shall report the
matter forthwith to the police. Yours faithfully, John Lacey.”
A blackmailing letter, without a doubt, Mr. Martin could see at a
glance. But what on earth could the fellow have discovered? Fifteen
years ago, when Mr. Martin moved from Praed Street to the
Barbican, he had taken the utmost care to destroy every trace of the
purpose to which that little cellar had been put. He racked his brains
to try and think of anything that could possibly have been
overlooked, but without success. Even those ingeniously contrived
recesses in the walls had been torn out, leaving nothing but the bare
brick. Mr. Martin had done the work with his own hands, not caring
to trust anyone else with so delicate a matter. Yet, after all, in spite
of all his care, he must have left some clue, which this infernal fellow
Lacey had somehow blundered upon.
He brushed aside as absurd the suggestion that “Lacey” could
conceal the identity of one of those with whom he had done
business, who knew the secret of the cellar and had adopted this
clumsy plan to blackmail him. The circle of his secret clients was a
very narrow one; he knew them all and was well aware that
blackmail was neither in their line nor in their interest. The expert in
acquiring jewellery concentrated upon his particular art; he did not
descend to blackmail, especially of the one man through whose
agency he was able to dispose of his spoil. No, there was no doubt
about it, a slip had been made, a slip which had lain undiscovered all
these years, through the circumstance that his successor, the
clothier, had found no use for the cellars and had boarded up the
entrance to them.
There was nothing for it but to carry out the suggestion
contained in the letter, and meet the man in Praed Street at two
o’clock. He must find out the nature of this mysterious discovery,
and take steps to ensure that all evidence of it be securely
destroyed. Lacey, of course, would demand some compensation for
holding his tongue. The sum he demanded would depend upon the
importance of what he had found. Well, they would discuss terms,
alone in that empty house, in the very cellar which Mr. Martin
remembered so well.
Mr. Martin stroked his chin reflectively. What sort of fellow was
this Lacey, he wondered? He himself was a powerful man, proud of
his own powers of intimidation. It ought to be simple enough for him
to destroy the evidence first then to tell Lacey to go to the devil and
do his worst. He unlocked a drawer of his desk and drew from it a
small automatic pistol, which he slipped into his pocket. It was with
a smile that he remembered that Praed Street had recently acquired
a sinister reputation. Well, if any accident should happen——
Mr. Martin’s office closed at one o’clock on Saturdays. He left a
few minutes before that hour, having put John Lacey’s letter in his
pocket and examined the mechanism of his automatic. Then he
consumed a hasty lunch at a chop-house near-by, and walked down
the street to Aldersgate Metropolitan station, where he took a ticket
to Paddington. He emerged into Praed Street at about five minutes
to two, and began to walk along that thoroughfare in the direction of
Number 407.
Praed Street, at this time on a fine afternoon, wore its usual air
of busy activity. The exodus from central London was still at its
height, the crowds still passed along in bus or taxi towards the
portals of Paddington station. On the other hand, the inhabitants of
the neighbourhood, released from work, were beginning their week-
end shopping. The pavement was crowded, and Mr. Martin, who had
some distance to walk to his destination, looked about him curiously.
Since he had put up his shutters for the last time, fifteen years ago,
he had rarely revisited Praed Street, and even then he had only
traversed it in a taxi, going to or from Paddington. The street had
changed a little since he had been an inhabitant of it. Certainly there
were fresh names over many of the shops, and, here and there,
there had been alterations to the buildings he remembered. He
glanced at a board on the opposite side of the street. Elmer
Ludgrove. That was new since his day. The rather musty-looking
shop was conspicuous among its neighbours by being closed. Praed
Street, as a rule, does most of its business on Saturday afternoon.
Mr. Martin walked on a step or two, then hesitated for an instant.
So Sam Copperdock, from whom he bought his cigarettes in those
early days, was still here. Mr. Martin had long ago abandoned
cigarettes for cigars, but he felt a momentary impulse to go in and
renew the old acquaintance. Sam Copperdock had been a good
customer of his, he remembered. Many a bottle of whiskey had he
sold him. But on second thoughts he refrained. Perhaps it would be
just as well, in the light of what might happen during the next half
hour or so, that his visit to Praed Street remained unsuspected.
Mr. Martin walked on, and in a few minutes found himself outside
Number 407. It was much as he remembered it, but that it was lying
empty, and showed signs of being refitted for a new tenant. The
windows were almost obscured with patches of white-wash, but
peering through them he could see trestles, timber, shavings, all the
litter which betrays the presence of the carpenter. So far, then, the
letter was correct. Mr. Martin, after a furtive glance round about him,
walked up to the door and tried it. It was locked.
He drew out his watch and consulted it. It was nearly five
minutes past two, but he remembered that he had noticed that
morning that it was a trifle fast. Lacey, whoever he was, would
doubtless be along in a moment. Mr. Martin hoped he would be
quick; he had no wish to be recognized outside the premises by any
of his old acquaintances. A sudden thought struck him. Perhaps
Lacey was already inside, waiting for him. There was an electric
bellpush by the side of the door. Mr. Martin pressed it firmly, but
could hear no answering ring above the roar of the traffic. Then all
at once he became aware that a small girl was standing by his side,
looking up at him with an appraising expression.
 As he turned to look at her, she made up her mind and spoke. “Is
your name Mr. Martin?” she asked.
Mr. Martin frowned down at her. “What’s that got to do with
you?” he replied.
“ ’Cos if you are, Dad told me to give you the key,” she continued,
quite unabashed. “And there’s a message from Mr. Lacey that you’re
to go in and wait for him, as he’ll be a few minutes late.”
“Oh, I see,” said Mr. Martin with a smile. “Yes, my dear, I am Mr.
Martin. How did your father come to have the key?”
She jerked her head towards a sweet shop two or three doors
away. “We lives there,” she replied curtly. “Mr. Briggs the builder,
what’s working here, leaves it with us because he’s my uncle. His
men come and get it in the morning. And Mr. Lacey, him what’s
coming here, telephoned just now to say you was coming to see the
drains and I was to give you the key.”
“Oh yes, that’s right,” said Mr. Martin, completely reassured.
“Thank you, my dear, I’ll see that either Mr. Lacey or myself brings
you back the key so that your uncle’s men may get in on Monday
morning. That’s the shop, there, isn’t it?”
He waited until the child had disappeared into the shop, then
opened the door of Number 407 with the key which she had given
him, walked in and closed the door behind him. Here he paused for
a moment, thinking rapidly. He felt that he was up against a brain as
agile as his own, and that the coming interview would demand all
his skill if he were to bring it to a favourable issue. This man Lacey
had taken steps to advertise the fact that Mr. Martin had entered the
premises, obviously as a precaution against foul play. His automatic
would be of no use to him, he would have to rely upon his brains
instead.
Then suddenly it flashed upon him that he had been given a very
fortunate opportunity. It was not to be supposed that Lacey had left
his discovery, whatever it was, lying about the cellar in full view of
the workmen who had been there that morning. But, if he could
have a quiet look round by himself, he might gain some clue to the
nature of the discovery, and so be prepared to meet Lacey when he
arrived.
 Mr. Martin was not anxious for Lacey to come upon him unawares
while he was searching the cellar. Fortunately, the possibility of such
an event could easily be guarded against. He bolted the door top
and bottom, and noticed with satisfaction that the bolts were brand
new, evidently part of the alterations which the new tenant was
making to the premises. Then, after glancing round the ground floor,
and making sure that nobody was hidden there, he made his way to
the top of the stairway which led down to the cellars. The upper part
of the house had no connection with the shop, and was approached
by a separate entrance.
He saw at once that the letter had been correct as to the
alterations being made. During his tenancy the cellars had been
lighted by gas, but now electric light was being installed throughout.
Mr. Martin could see the run of the wires and the fittings for the
lamps, but he noticed that the lamps themselves had not yet been
inserted. This fact did not worry him. He had a box full of matches in
his pocket, and one of these he lit as he cautiously descended the
stairs. The builders’ litter was here, as everywhere, and he picked
his way among it until he stood in the main vault below the shop. It
smelt close and musty, as though it had only recently been opened
up after a long period of disuse. Mr. Martin, holding a lighted match
above his head, glanced rapidly round. Nothing had been altered
here since he remembered it, except that a new brass switch stood
behind the door, and two empty lamp sockets projected downwards
from the ceiling.
The way to the little cellar below the back office lay along a
narrow passage, and this Mr. Martin followed, striking a fresh match
as he went. Every detail came back to him, as though he had last
passed this way yesterday, instead of fifteen years ago. The door of
the little cellar closed with a spring; as he pushed it open the spring
creaked with a note which he remembered so well, and the door
closed behind him with the same muffled bang. Again he raised his
match and looked around him anxiously. The cellar appeared to be
the same as he had left it, even to a mildewed packing case against
the wall, which had sometimes been used as a table. Only here, as
in the larger cellar, the electricians had been at work. There was a
lamp holder on the ceiling, but this time it held a lamp. It was
evidently by the light of this lamp that Lacey had made his discovery.
There was no reason why he should not use it too.
The match had burnt down very near his fingers, but as he
dropped it he caught sight of the switch. Without troubling to light
another match he groped his way towards it and turned it on. There
was a flash and a sharp report, but the cellar remained in complete
darkness. Mr. Martin, with a gasp, leapt for the door in sudden panic.
Someone hidden in the cellar must have fired at him. Then, after a
moment, the absurdity of the idea forced itself upon him. The report
had not been nearly loud enough for a pistol, and the cellar was far
too small to conceal anybody. With a short laugh, but with trembling
fingers, he felt for his matchbox and struck a light.
Then at once he saw what had happened. The globe had burst
as he turned on the current, and only the butt remained in the
holder. He turned to look at the switch, and then for the first time he
noticed that something white was hanging from it by a string. He
bent down and held his match close to it. It was a white counter,
bearing the figure IV in red ink.
With a sudden access of fear he clutched at it, in order to
examine it more closely. The string broke, and he stared at the
counter in his hand, fascinated. He began to realize his position. He
had been lured to this empty house, and the assassin was no doubt
even now in the passage outside, waiting to plunge his deadly knife
into him as he emerged. Thank heaven, he had brought his
automatic with him!
Mr. Martin was no coward. His plan of defence was quickly made.
Throwing the match away, he took his stand on the far side of the
cellar, facing the door, his automatic in his hand. If his assailant
came in, he would hear the creak of the spring, and could fire before
the man could reach him.
For a second or two he waited, tense and with ears strained for
the slightest sound. It was very quiet in the cellar; the rumble of the
traffic in Praed Street came to him but faintly. Then, all at once, he
caught his breath as a pungent and unfamiliar smell reached him.
Suddenly he found himself choking, dizzy, his senses leaving him. An
overpowering impulse to escape from the cellar swept through him,
drowning his fear of what might be awaiting him outside. He
staggered towards the door, his hands clawing wildly over its damp
surface. There used to be a string by which one pulled it open, but
Mr. Martin could not find it. Vainly he strove against the awful
numbness which closed in upon him. He must find the string, must
——
With a crash Mr. Martin fell down at the foot of the door. The little
cellar resumed its accustomed silence.
 Chapter IX.
A Strange Affair
It was not until shortly after eight o’clock on Monday morning
that Inspector Whyland, who had arrived early at the police station,
received the startling intelligence that there was a dead man lying in
the cellar of Number 407, Praed Street. He immediately jumped into
a taxi, and was met in the door-way of the empty house by a man
who introduced himself as Mr. Houlder, the builder who was carrying
out certain alterations for the new tenant, Mr. Lacey.
“My men found him when they came in this morning,” he said.
“They telephoned to me and I told them to fetch a policeman. He
rang you up, I understand. I think I can explain how the man got in,
and who he is.”
“Thanks very much, Mr. Houlder,” replied Whyland. “I think we’ll
have a look at him first, if you don’t mind. Will you lead the way?”
They descended to the cellar together, Mr. Houlder leading the
way with an electric torch. As they arrived at the passage leading to
the small cellar, a constable appeared and saluted.
“Ah, you’re in charge here, I suppose?” said the Inspector. “You
haven’t touched anything, I hope?”
“Not since I’ve been here, sir, but I understand that the body was
moved accidentally before I came, I’ve got the man who found him
here, sir.”
A man, who had been hidden in the gloom behind the policeman,
came forward at this. He explained that he was the electrician who
had been wiring the cellars. He had very nearly finished the job on
the previous Saturday, and had left about noon. The carpenters were
still working on the ground floor when he left. He returned just
before eight this very morning, found the carpenters already at
work, and went down to the cellars to put the finishing touches to
the job. The door of the small cellar appeared to be jammed, and he
pushed against it to open it. When he had got it open far enough to
squeeze through, he found the body of a man lying against it. He
had immediately run upstairs and told the foreman, and had waited
there till the constable came and asked him to show him the body.
Inspector Whyland dismissed Mr. Houlder and the electricians,
and went on into the cellar with the constable. “There’s something
queer about this business, sir?” said the latter, as soon as they were
alone. “I had a look round while I was waiting for you, and the first
thing I saw was one of them white counters, same as the others
had. There it lies, sir, I didn’t touch it.”
They were in the cellar by now, and Whyland glanced at the
counter, lying in the beam of light which the constable had thrown
upon it. “With the figure IV on it, I’ll wager,” he muttered. “Yes, I
thought so. We’ll try for finger marks, but I’ll bet it’s no good. Now,
let’s have a look at this dead man.”
The constable turned his lamp on the prostrate form, and
Whyland knelt down and gazed at it intently. The body was lying
doubled up as it had fallen, and was quite cold.
“H’m,” said Whyland, rising to his feet. “We can’t do much more
till the doctor comes. You don’t know who he is, I suppose?”
“Mr. Houlder said he believed his name was Martin, sir,” replied
the constable cautiously.
“Well, you stay here till the doctor comes. I’ll go and have a chat
with Houlder. Just throw your light over the floor for a minute. Hullo,
what’s this?”
He strode across the cellar and carefully picked up a small
automatic. “Hasn’t been fired,” he muttered. “Now I wonder who
that belongs to? Just see if you can spot anything else. Don’t touch
it, if you do.”
He went upstairs again, and drew Mr. Houlder aside into a quiet
corner. Mr. Houlder’s story was a very simple one. His foreman had
orders to leave the key to the house with Mr. Briggs, the
confectioner, three doors off. Mr. Briggs was Houlder’s brother-in-
law. The reason for this arrangement was that whoever came on the