Factorization Algebras in Quantum Field Theory 1St Edition Kevin Costello Online Ebook Texxtbook Full Chapter PDF

Factorization Algebras in Quantum
Field Theory 1st Edition Kevin Costello

Visit to download the full and correct content document:
https://ebookmeta.com/product/factorization-algebras-in-quantum-field-theory-1st-edit
ion-kevin-costello/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...
Advanced Topics in Quantum Field Theory 2nd Edition

Mikhail Shifman
https://ebookmeta.com/product/advanced-topics-in-quantum-field-
theory-2nd-edition-mikhail-shifman/
Relativistic Quantum Field Theory Michael T. Strickland
https://ebookmeta.com/product/relativistic-quantum-field-theory-
michael-t-strickland/
Quantum Field Theory 2nd Edition Michael V. Sadovskii
https://ebookmeta.com/product/quantum-field-theory-2nd-edition-
michael-v-sadovskii/
Wilson Lines in Quantum Field Theory 2nd Edition Igor

Olegovich Cherednikov
https://ebookmeta.com/product/wilson-lines-in-quantum-field-
theory-2nd-edition-igor-olegovich-cherednikov/
Problems in Quantum Field Theory With Fully Worked
Solutions 1st Edition Gelis
https://ebookmeta.com/product/problems-in-quantum-field-theory-
with-fully-worked-solutions-1st-edition-gelis/
Statistical Approach to Quantum Field Theory Andreas

Wipf
https://ebookmeta.com/product/statistical-approach-to-quantum-
field-theory-andreas-wipf/
Nonequilibrium Quantum Field Theory 2nd Edition Esteban

A. Calzetta
https://ebookmeta.com/product/nonequilibrium-quantum-field-
theory-2nd-edition-esteban-a-calzetta/
Quantum Field Theory A Diagrammatic Approach 1st

Edition Ronald Kleiss
https://ebookmeta.com/product/quantum-field-theory-a-
diagrammatic-approach-1st-edition-ronald-kleiss/
Quantum Field Theory An Introduction 1st Edition Gordon

Walter Semenoff
https://ebookmeta.com/product/quantum-field-theory-an-
introduction-1st-edition-gordon-walter-semenoff/
Factorization Algebras in Quantum Field Theory
Volume 2
Factorization algebras are local-to-global objects that play a role in classical and
quantum field theory that is similar to the role of sheaves in geometry: they conveniently
organize complicated information. Their local structure encompasses examples such as
associative and vertex algebras; in these examples, their global structure encompasses
Hochschild homology and conformal blocks.
In this second volume, the authors show how factorization algebras arise from
interacting field theories, both classical and quantum, and how they encode essential
information such as operator product expansions, Noether currents, and anomalies.
Along with a systematic reworking of the Batalin–Vilkovisky formalism via derived
geometry and factorization algebras, this book offers concrete examples from physics,
ranging from angular momentum and Virasoro symmetries to a five-dimensional gauge
theory.
K e v i n C o s t e l l o is Krembil William Rowan Hamilton Chair in Theoretical

Physics at the Perimeter Institute for Theoretical Physics, Waterloo, Canada. He is an
honorary member of the Royal Irish Academy and a Fellow of the Royal Society. He has
won several awards, including the Berwick Prize of the London Mathematical Society
(2017) and the Eisenbud Prize of the American Mathematical Society (2020).
O w e n G w i l l i a m is Assistant Professor in the Department of Mathematics and

Statistics at the University of Massachusetts, Amherst.
N E W M AT H E M AT I C A L M O N O G R A P H S
Editorial Board
Jean Bertoin, Béla Bollobás, William Fulton, Bryna Kra, Ieke Moerdijk,
Cheryl Praeger, Peter Sarnak, Barry Simon, Burt Totaro
All the titles listed below can be obtained from good booksellers or from Cambridge University
Press. For a complete series listing visit www.cambridge.org/mathematics.
1. M. Cabanes and M. Enguehard Representation Theory of Finite Reductive Groups

2. J. B. Garnett and D. E. Marshall Harmonic Measure
3. P. Cohn Free Ideal Rings and Localization in General Rings
4. E. Bombieri and W. Gubler Heights in Diophantine Geometry
5. Y. J. Ionin and M. S. Shrikhande Combinatorics of Symmetric Designs
6. S. Berhanu, P. D. Cordaro and J. Hounie An Introduction to Involutive Structures
7. A. Shlapentokh Hilbert’s Tenth Problem
8. G. Michler Theory of Finite Simple Groups I
9. A. Baker and G. Wüstholz Logarithmic Forms and Diophantine Geometry
10. P. Kronheimer and T. Mrowka Monopoles and Three–Manifolds
11. B. Bekka, P. de la Harpe and A. Valette Kazhdan’s Property (T)
12. J. Neisendorfer Algebraic Methods in Unstable Homotopy Theory
13. M. Grandis Directed Algebraic Topology
14. G. Michler Theory of Finite Simple Groups II
15. R. Schertz Complex Multiplication
16. S. Bloch Lectures on Algebraic Cycles (2nd Edition)
17. B. Conrad, O. Gabber and G. Prasad Pseudo-reductive Groups
18. T. Downarowicz Entropy in Dynamical Systems
19. C. Simpson Homotopy Theory of Higher Categories
20. E. Fricain and J. Mashreghi The Theory of H(b) Spaces I
21. E. Fricain and J. Mashreghi The Theory of H(b) Spaces II
22. J. Goubault-Larrecq Non-Hausdorff Topology and Domain Theory
23. J. Śniatycki Differential Geometry of Singular Spaces and Reduction of Symmetry
24. E. Riehl Categorical Homotopy Theory
25. B. A. Munson and I. Volić Cubical Homotopy Theory
26. B. Conrad, O. Gabber and G. Prasad Pseudo-reductive Groups (2nd Edition)
27. J. Heinonen, P. Koskela, N. Shanmugalingam and J. T. Tyson Sobolev Spaces on Metric
Measure Spaces
28. Y.-G. Oh Symplectic Topology and Floer Homology I
29. Y.-G. Oh Symplectic Topology and Floer Homology II
30. A. Bobrowski Convergence of One-Parameter Operator Semigroups
31. K. Costello and O. Gwilliam Factorization Algebras in Quantum Field Theory I
32. J.-H. Evertse and K. Győry Discriminant Equations in Diophantine Number Theory
33. G. Friedman Singular Intersection Homology
34. S. Schwede Global Homotopy Theory
35. M. Dickmann, N. Schwartz and M. Tressl Spectral Spaces
36. A. Baernstein II Symmetrization in Analysis
37. A. Defant, D. Garcı́a, M. Maestre and P. Sevilla-Peris Dirichlet Series and Holomorphic
Functions in High Dimensions
38. N. Th. Varopoulos Potential Theory and Geometry on Lie Groups
39. D. Arnal and B. Currey Representations of Solvable Lie Groups
40. M. A. Hill, M. J. Hopkins and D. C. Ravenel Equivariant Stable Homotopy Theory and the
Kervaire Invariant Problem
Factorization Algebras in
Quantum Field Theory
Volume 2
KEVIN COSTELLO
Perimeter Institute for Theoretical Physics, Waterloo, Ontario
OW E N G W I L L I A M
University of Massachusetts, Amherst
University Printing House, Cambridge CB2 8BS, United Kingdom
One Liberty Plaza, 20th Floor, New York, NY 10006, USA
477 Williamstown Road, Port Melbourne, VIC 3207, Australia
314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre,
New Delhi – 110025, India
103 Penang Road, #05–06/07, Visioncrest Commercial, Singapore 238467
Cambridge University Press is part of the University of Cambridge.

It furthers the University’s mission by disseminating knowledge in the pursuit of
education, learning, and research at the highest international levels of excellence.
www.cambridge.org
Information on this title: www.cambridge.org/9781107163157
DOI: 10.1017/9781316678664
© Kevin Costello and Owen Gwilliam 2021
This publication is in copyright. Subject to statutory exception
and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.
First published 2021
A catalogue record for this publication is available from the British Library.
Library of Congress Cataloging-in-Publication Data
Names: Costello, Kevin, 1977- author. | Gwilliam, Owen, author.
Title: Factorization algebras in quantum field theory / Kevin Costello, Perimeter Institute
for Theoretical Physics, Waterloo, Ontario, Owen Gwilliam, Max Planck Institute for Mathematics, Bonn.
Other titles: New mathematical monographs ; 31.
Description: Cambridge, United Kingdom ; New York, NY : Cambridge University Press, 2017-. |
Series: New mathematical monographs ; 31 | Includes bibliographical references and index.
Contents: From Gaussian measures to factorization algebras – Prefactorization algebras and
basic examples – Free field theories – Holomorphic field theories and vertex algebras –
Factorization algebras: definitions and constructions – Formal aspects of
factorization algebras – Factorization algebras: examples.
Identifiers: LCCN 2016047832| ISBN 9781107163102 (hardback ; v. 1) | ISBN 1107163102
(hardback ; v. 1) |
ISBN 9781107163157 (hardback ; v. 2) | ISBN 9781316678664 (epub ; v. 2)
Subjects: LCSH: Quantum field theory–Mathematics. | Factorization (Mathematics) | Factors (Algebra) |
Geometric quantization. | Noncommutative algebras.
Classification: LCC QC174.45 .C68 2017 | DDC 530.14/30151272–dc23 LC record available
at https://lccn.loc.gov/2016047832
ISBN – 2 Volume Set 978-1-009-00616-3 Hardback
ISBN – Volume 1 978-1-107-16310-2 Hardback
ISBN – Volume 2 978-1-107-16315-7 Hardback
Cambridge University Press has no responsibility for the persistence or accuracy of
URLs for external or third-party internet websites referred to in this publication
and does not guarantee that any content on such websites is, or will remain,
accurate or appropriate.
To Josie and Dara
&
To Laszlo and Hadrian
Contents
Contents of Volume 1 page xi

1 Introduction and Overview 1
1.1 The Factorization Algebra of Classical Observables 1
1.2 The Factorization Algebra of Quantum Observables 2
1.3 The Physical Importance of Factorization Algebras 3
1.4 Poisson Structures and Deformation Quantization 6
1.5 The Noether Theorem 8
1.6 Brief Orienting Remarks toward the Literature 11
1.7 Acknowledgments 12
PART I CLASSICAL FIELD THEORY 15

2 Introduction to Classical Field Theory 17
2.1 The Euler–Lagrange Equations 17
2.2 Observables 18
2.3 The Symplectic Structure 19
2.4 The P0 Structure 19
3 Elliptic Moduli Problems 20
3.1 Formal Moduli Problems and Lie Algebras 21
3.2 Examples of Elliptic Moduli Problems Related
to Scalar Field Theories 25
to Gauge Theories 28
3.4 Cochains of a Local L∞ Algebra 32
3.5 D-modules and Local L∞ Algebras 34
vii
viii Contents
4 The Classical Batalin–Vilkovisky Formalism 43

4.1 The Classical BV Formalism in Finite Dimensions 43
4.2 The Classical BV Formalism in Infinite Dimensions 45
4.3 The Derived Critical Locus of an Action Functional 48
4.4 A Succinct Definition of a Classical Field Theory 54
4.5 Examples of Scalar Field Theories from Action Functionals 57
4.6 Cotangent Field Theories 58
5 The Observables of a Classical Field Theory 63
5.1 The Factorization Algebra of Classical Observables 63
5.2 The Graded Poisson Structure on Classical Observables 64
5.3 The Poisson Structure for Free Field Theories 66
5.4 The Poisson Structure for a General Classical Field Theory 68
PART II QUANTUM FIELD THEORY 73

6 Introduction to Quantum Field Theory 75
6.1 The Quantum BV Formalism in Finite Dimensions 76
6.2 The “Free Scalar Field” in Finite Dimensions 79
6.3 An Operadic Description 81
6.4 Equivariant BD Quantization and Volume Forms 82
6.5 How Renormalization Group Flow Interlocks with the BV
Formalism 83
6.6 Overview of the Rest of This Part 84
7 Effective Field Theories and Batalin–Vilkovisky Quantization 86
7.1 Local Action Functionals 87
7.2 The Definition of a Quantum Field Theory 88
7.3 Families of Theories over Nilpotent dg Manifolds 99
7.4 The Simplicial Set of Theories 105
7.5 The Theorem on Quantization 109
8 The Observables of a Quantum Field Theory 111
8.1 Free Fields 111
8.2 The BD Algebra of Global Observables 115
8.3 Global Observables 124
8.4 Local Observables 126
8.5 Local Observables Form a Prefactorization Algebra 128
8.6 Local Observables Form a Factorization Algebra 132
8.7 The Map from Theories to Factorization Algebras Is a Map
of Presheaves 138
Contents ix
9 Further Aspects of Quantum Observables 144

9.1 Translation Invariance for Field Theories and Observables 144
9.2 Holomorphically Translation-Invariant Theories
and Observables 148
9.3 Renormalizability and Factorization Algebras 154
9.4 Cotangent Theories and Volume Forms 168
9.5 Correlation Functions 177
10 Operator Product Expansions, with Examples 179
10.1 Point Observables 179
10.2 The Operator Product Expansion 185
10.3 The OPE to First Order in 188
10.4 The OPE in the φ 4 Theory 197
10.5 The Operator Product for Holomorphic Theories 201
10.6 Quantum Groups and Higher-Dimensional Gauge Theories 210
PART III A FACTORIZATION ENHANCEMENT

OF THE NOETHER THEOREM 225
11 Introduction to the Noether Theorems 227
11.1 Symmetries in the Classical BV Formalism 228
11.2 Koszul Duality and Symmetries via the Classical Master
Equation 233
11.3 Symmetries in the Quantum BV Formalism 239
12 The Noether Theorem in Classical Field Theory 245
12.1 An Overview of the Main Theorem 245
12.2 Symmetries of a Classical Field Theory 246
12.3 The Factorization Algebra of Equivariant Classical
Observables 259
12.4 The Classical Noether Theorem 263
12.5 Conserved Currents 268
12.6 Examples of the Classical Noether Theorem 270
12.7 The Noether Theorem and the Operator Product Expansion 279
13 The Noether Theorem in Quantum Field Theory 289
13.1 The Quantum Noether Theorem 289
13.2 Actions of a Local L∞ Algebra on a Quantum Field Theory 294
13.3 Obstruction Theory for Quantizing Equivariant Theories 299
13.4 The Factorization Algebra of an Equivariant
Quantum Field Theory 303
13.5 The Quantum Noether Theorem Redux 304
x Contents
13.6 Trivializing the Action on Factorization Homology 313

13.7 The Noether Theorem and the Local Index Theorem 314
13.8 The Partition Function and the Quantum Noether Theorem 323
14 Examples of the Noether Theorems 325
14.1 Examples from Mechanics 325
14.2 Examples from Chiral Conformal Field Theory 344
14.3 An Example from Topological Field Theory 354
Appendix A Background 360
Appendix B Functions on Spaces of Sections 375
Appendix C A Formal Darboux Lemma 385
References 393
Index 399
Contents of Volume 1
1 Introduction page 1
1.1 The Motivating Example of Quantum Mechanics 3
1.2 A Preliminary Definition of Prefactorization Algebras 8
1.3 Prefactorization Algebras in Quantum Field Theory 8
1.4 Comparisons with Other Formalizations of Quantum Field
Theory 11
1.5 Overview of This Volume 16
PART I PREFACTORIZATION ALGEBRAS 21

2 From Gaussian Measures to Factorization Algebras 23
2.1 Gaussian Integrals in Finite Dimensions 25
2.2 Divergence in Infinite Dimensions 27
2.3 The Prefactorization Structure on Observables 31
2.4 From Quantum to Classical 34
2.6 Further Results on Free Field Theories 39
2.7 Interacting Theories 40
3 Prefactorization Algebras and Basic Examples 44
3.1 Prefactorization Algebras 44
3.2 Associative Algebras from Prefactorization Algebras on R 51
3.3 Modules as Defects 52
3.4 A Construction of the Universal Enveloping Algebra 59
3.5 Some Functional Analysis 62
xi
xii Contents of Volume 1
3.6 The Factorization Envelope of a Sheaf of Lie Algebras 73

3.7 Equivariant Prefactorization Algebras 79
PART II FIRST EXAMPLES OF FIELD THEORIES

AND THEIR OBSERVABLES 87
4 Free Field Theories 89
4.1 The Divergence Complex of a Measure 89
4.2 The Prefactorization Algebra of a Free Field Theory 93
4.3 Quantum Mechanics and the Weyl Algebra 106
4.4 Pushforward and Canonical Quantization 112
4.5 Abelian Chern–Simons Theory 115
4.6 Another Take on Quantizing Classical Observables 124
4.8 Translation–Invariant Prefactorization Algebras 131
4.9 States and Vacua for Translation Invariant Theories 139
5 Holomorphic Field Theories and Vertex Algebras 145
5.1 Vertex Algebras and Holomorphic Prefactorization
Algebras on C 145
5.2 Holomorphically Translation-Invariant Prefactorization
Algebras 149
5.3 A General Method for Constructing Vertex Algebras 157
5.4 The βγ System and Vertex Algebras 171
5.5 Kac–Moody Algebras and Factorization Envelopes 188
PART III FACTORIZATION ALGEBRAS 205

6 Factorization Algebras: Definitions and Constructions 207
6.1 Factorization Algebras 207
6.2 Factorization Algebras in Quantum Field Theory 215
6.3 Variant Definitions of Factorization Algebras 216
6.4 Locally Constant Factorization Algebras 220
6.5 Factorization Algebras from Cosheaves 225
6.6 Factorization Algebras from Local Lie Algebras 230
7 Formal Aspects of Factorization Algebras 232
7.1 Pushing Forward Factorization Algebras 232
7.2 Extension from a Basis 232
7.3 Pulling Back Along an Open Immersion 240
7.4 Descent Along a Torsor 241
Contents of Volume 1 xiii
8 Factorization Algebras: Examples 243

8.1 Some Examples of Computations 243
8.2 Abelian Chern–Simons Theory and Quantum Groups 249
Appendix A Background 273
Appendix B Functional Analysis 310
Appendix C Homological Algebra in Differentiable Vector Spaces 351
Appendix D The Atiyah–Bott Lemma 374
References 377
Index 383
1
Introduction and Overview
This is the second volume of our two-volume book on factorization algebras as

they apply to quantum field theory. In Volume 1, we focused on the theory of
factorization algebras while keeping the quantum field theory to a minimum.
Indeed, we only ever discussed free theories. In Volume 2, we will focus on the
factorization algebras associated with interacting classical and quantum field
theories.
In this introduction, we will state in outline the main results that we prove
in this volume. The centerpiece is a deformation quantization approach to
quantum field theory, analogous to that for quantum mechanics, and the
introduction to the first volume provides an extensive motivation for this
perspective, which is put on solid footing here. Subsequently we explore
symmetries of field theories that fit into this approach, leading to classical
and quantum versions of the Noether theorem in the language of factorization
algebras.
Remark: Throughout the text, we refer to results from the first volume in the
style “see Chapter I.2” to indicate the second chapter of Volume 1. ♦
1.1 The Factorization Algebra of Classical Observables

We will start with the factorization algebra associated with a classical field
theory. Suppose we have a classical field theory on a manifold M, given by
some action functional, possibly with some gauge symmetry. To this data we
will associate a factorization algebra of classical observables. The construction
goes as follows. First, for every open subset U ⊂ M, consider the space EL(U)
of solutions to the Euler–Lagrange equations on U, modulo gauge. We work in
perturbation theory, which means we consider solutions that live in the formal
neighborhood of a fixed solution. We also work in the derived sense, which
1
2 Introduction and Overview
means we “impose” the Euler–Lagrange equations by a Koszul complex. In

sophisticated terms, we take EL(U) to be the formal derived stack of solutions
to the equations of motion. As U varies, the collection EL(U) forms a sheaf of
formal derived stacks on M.
The factorization algebra Obscl of classical observables of the field theory
assigns to an open U, the dg commutative algebra O(EL(U)) of functions
on this formal derived stack EL(U). This construction is simply the derived
version of functions on solutions to the Euler–Lagrange equations, and hence
provides a somewhat sophisticated refinement for classical observables in the
typical sense.
It takes a little work to set up a theory of formal derived geometry that
can handle formal moduli spaces of solutions to nonlinear partial differential
equations like the Euler–Lagrange equations. In the setting of derived geom-
etry, formal derived stacks are equivalent to homotopy Lie algebras (i.e., Lie
algebras up to homotopy, often modeled by dg Lie algebras or L∞ algebras).
The theory we develop in Chapter 3 takes this characterization as a definition.
We define a formal elliptic moduli problem on a manifold M to be a sheaf
of homotopy Lie algebras satisfying certain properties. Of course, for the
field theories considered in this book, the formal moduli of solutions to the
Euler–Lagrange equations always define a formal elliptic moduli problem.
We develop the theory of formal elliptic moduli problems sufficiently to define
the dg algebra of functions, as well as other geometric concepts.
1.2 The Factorization Algebra of Quantum Observables

In Chapter 8, we give our main construction. It gives a factorization algebra
Obsq of quantum observables for any quantum field theory in the sense of
Costello (2011b). A quantum field theory is, by that definition, something
that lives over C[[]] and reduces modulo to a classical field theory. The
factorization algebra Obsq is then a factorization algebra over C[[]], and
modulo it reduces to a factorization algebra quasi-isomorphic to the algebra
Obscl of classical observables.
The construction of the factorization algebra of quantum observables is a
bit technical. The techniques arise from the approach to quantum field theory
developed in Costello (2011b). In that book a quantum field theory is defined
to be a collection of functionals I[L] on the fields that are approximately
local. They play a role analogous to the action functional of a classical field
theory. These functionals depend on a “length scale” L, and when L is close to
zero the functional I[L] is close to being local. The axioms of a quantum field
theory are:
1.3 The Physical Importance of Factorization Algebras 3
(i) As L varies, I[L] and I[L ] are related by the operation of “renormal-
ization group flow.” Intuitively, if L > L, then I[L ] is obtained from
I[L] by integrating out certain high-energy fluctuations of the fields.
(ii) Each I[L] satisfies a scale L quantum master equation (the quantum
version of a compatibility with gauge symmetry).
(iii) When we reduce modulo and send L → 0, then I[L]
becomes the interaction term in the classical Lagrangian.
The fact that I[L] is never local, just close to being local as L → 0, means
that we have to work a bit to define the factorization algebra. The essential
idea is simple, however. If U ⊂ M, we define the cochain complex Obsq (U)
to be the space of first-order deformations {I[L] + O[L]} of the collection of
functionals I[L] that define the theory. We ask that this first-order deformation
satisfies the renormalization group flow property modulo 2 . This condition
gives a linear expression for O[L] in terms of any other O[L ]. This idea
reflects the familiar intuition from the path integral that observables are first-
order deformations of the action functional.
The observables we are interested in do not need to be localized at a point,
or indeed given by the integral over the manifold of something localized at a
point. Therefore, we should not ask that O[L] becomes local as L → 0. Instead,
we ask that O[L] becomes supported on U as L → 0.
Moreover, we do not ask that I[L] + O[L] satisfies the scale L quantum
master equation (modulo 2 ). Instead, its failure to satisfy the quantum
master equation defines the differential on the cochain complex of quantum
observables.
With a certain amount of work, we show that this definition defines a
factorization algebra Obsq that quantizes the factorization algebra Obscl of
classical observables.
1.3 The Physical Importance of Factorization Algebras

Our key claim is that factorization algebras encode, in a mathematically clean
way, the features of a quantum field theory that are important in physics.
This formalism must thus include the most important examples of quantum
field theories from physics. Fortunately, the techniques developed in Costello
(2011b) give a cohomological technique for constructing quantum field the-
ories, which applies easily to many examples. For instance, the Yang–Mills
theory and the φ 4 theory on R4 were both constructed in Costello (2011b).
(Note that we work throughout on Riemannian manifolds, not Lorentzian
ones.)
As a consequence, each theory has a factorization algebra on R4 that

encodes its observables.
1.3.1 Correlation Functions from Factorization Algebras

In the physics literature on quantum field theory, the fundamental objects are
correlation functions of observables. The factorization algebra of a quantum
field theory contains enough data to encode the correlation functions. In this
sense, its factorization algebra encodes the essential data of a quantum field
theory.
Let us explain how this encoding works. Assume that we have a field theory
on a compact manifold M. Suppose that we work near an isolated solution
to the equations of motion, that is, one that admits no small deformations.
(Strictly speaking, we require that the cohomology of the tangent complex
to the space of solutions to the equations of motion is zero, which is a little
stronger, as it means that there are also no gauge symmetries preserving this
solution to the equations of motion.) Some examples of theories where we have
an isolated solution to the equations of motion are a massive scalar field theory,
on any compact manifold, or a massless scalar field theory on the four-torus T 4
where the field has monodromy −1 around some of the cycles. In each case,
we can include an interaction, such as the φ 4 interaction.
Since the classical observables are functions on the space of solution to the
equations of motion, our assumption implies H ∗ (Obscl (M)) = C. A spectral
sequence argument then lets us conclude that H ∗ (Obsq (M)) = C[[]].
If U1, . . . ,Un ⊂ M are disjoint open subsets of M, the factorization algebra
structure gives a map
− : H ∗ (Obsq (U1 )) ⊗ · · · ⊗ H ∗ (Obsq (Un )) → H ∗ (Obsq (M)) = C[[]].
If Oi ∈ H ∗ (Obsq (Ui )) are observables on the open subsets Ui , then O1 · · · On
is the correlation function of these observables.
Consider again the φ 4 theory on T 4 , where the φ field has monodromy
−1 around one of the four circles. In the formalism of Costello (2011b), it is
possible to construct the φ 4 theory on R4 so that the Z/2 action sending φ to
−φ is preserved. By descent, the theory – and hence the factorization algebra –
exists on T 4 as well. Thus, this theory provides an example where the quantum
theory can be constructed and the correlation functions defined.
1.3.2 Factorization Algebras and Renormalization Group Flow

Factorization algebras provide a satisfying geometric understanding of the RG
flow, which we discuss in detail in Chapter 9 but sketch now.
1.3 Factorization Algebras and the Operator Product Expansion 5
In Costello (2011b), a scaling action of R>0 on the collection of theories

on Rn was given. It provides a rigorous version of the RG flow as defined by
Wilson.
There is also a natural action by scaling of the group R>0 (under multiplica-
tion) on the collection of translation-invariant factorization algebras on Rn . Let
Rλ : Rn → Rn denote the diffeomorphism that rescales the coordinates, and
let F be a translation-invariant factorization algebra on Rn . Then the pullback
R∗λ F is a new factorization algebra on Rn .
We show that the map from theories on Rn to factorization algebras on
R intertwines these two R>0 actions. Thus, this simple scaling action on
n
factorization algebras is the RG flow.

In order to define have a quantum field theory with finitely many free
parameters, it is generally essential to only consider renormalizable quantum
field theories. In Costello (2011b), it was shown that for any translation-
invariant field theory on Rn , the dependence of the field theory on the scalar
parameter λ ∈ R>0 is via powers of λ and of log λ. A strictly renormalizable
theory is one in which the dependence is only via log λ, and the quantizations
of the Yang–Mills theory and φ 4 theory constructed in Costello (2011b) both
have this feature.
We can translate the concept of renormalizability into the language of
factorization algebras. For any translation-invariant factorization algebra F on
Rn , there is a family of factorization algebras Fλ = R∗λ F on Rn . Because this
family depends smoothly on λ, a priori it defines a factorization algebra over
the base ring C∞ (R>0 ) of smooth functions of the variable λ. We say this
family is strictly renormalizable if it arises by extension of scalars from a
factorization algebra over the base ring C[log λ] of polynomials in log λ.
The factorization algebras associated with the Yang–Mills theory and the φ 4
theory both have this feature.
In this way, we formulate via factorization algebras the concept of renormal-
izability of a quantum field theory.
1.3.3 Factorization Algebras and the Operator

Product Expansion
One disadvantage of the language of factorization algebras is that the fac-
torization algebra structure is often very difficult to describe explicitly. The
reason is that for an open set U, the space Obsq (U) of quantum observables
on U is a very large topological vector space, and it is not obvious how one
can give it a topological basis. To extract more explicit computations, we
introduce the concept of a point observable in Chapter 10. The space of point
observables is defined to be the limit limr→0 Obsq (D(0,r)) of the space of

quantum observables on a disc of radius r around the origin, as r → 0. Point
observables capture what physicists call local operators; however, we eschew
the term operator as our formalism does not include a Hilbert space on which
we can operate.
Given two point observables O1 and O2 , we can place O1 at 0 and O2 at x
and then use the factorization product on sufficiently small discs centered at 0
and x to define a product element
O1 (0) · O2 (x) ∈ Obsq (Rn ).
The operator product is defined by expanding the product O1 (0) · O2 (x) as a

function of x and extracting the “singular part.” It is not guaranteed that such
an expansion exists in general, but we prove in Chapter 10 that it does exist to
order . This order operator product expansion can be computed explicitly,
and we do so in detail for several theories in Chapter 10. The methods exhibited
there provide a source of concrete examples in which mathematicians can
rigorously compute quantities of quantum field theory.
That chapter also contains the longest and most detailed example in this
book. It has recently become clear (Costello 2013b; Costello et al. 2019)
that one can understand quantum groups, such as the Yangian and related
algebras, using Feynman diagram computations in quantum field theory. The
general idea is that one should take a quantum field theory that has one
topological direction, so that the factorization product in this direction gives us
a (homotopy) associative algebra. By taking the Koszul dual of this associative
algebra, one finds a new algebra that in certain examples is a quantum group.
For the Yangian, the relevant Feynman diagram computations are given in
Costello et al. (2019). We present in detail an example related to a different
infinite-dimensional quantum group, following Costello (2017). We perform
one-loop Feynman diagram computations that reproduce the commutation
relations in this associative algebra. (We chose this example as the relevant
Feynman diagram computations are considerably easier than those that lead to
the Yangian algebra in Costello et al. 2019.)
1.4 Poisson Structures and Deformation Quantization

In the deformation quantization approach to quantum mechanics, the asso-
ciative algebra of quantum operators reduces, modulo , to the commutative
algebra of classical operators. But this algebra of classical operators has a little
more structure: It is a Poisson algebra. Deformation quantization posits that
1.4 Poisson Structures and Deformation Quantization 7
the failure of the algebra of quantum operators to be commutative is given, to

first order in , by the Poisson bracket.
Something similar happens in our story. Classical observables are given
by the algebra of functions on the derived space of solutions to the Euler–
Lagrange equations. The Euler–Lagrange equations are not just any partial
differential equations (PDEs), however: They describe the critical locus of
an action functional. The derived critical locus of a function on a finite-
dimensional manifold carries a shifted Poisson (or P0 ) structure, meaning that
its dg algebra of functions has a Poisson bracket of degree 1. In the physics lit-
erature, this Poisson bracket is sometimes called the BV bracket or antibracket.
This feature suggests that the space of solutions to the Euler–Lagrange
equations should also have a P0 structure, and so the factorization algebra
Obscl of classical observables has the structure of a P0 algebra. We show that
this guess is indeed true, as long as we use a certain homotopical version of P0
factorization algebras.
Just as in the case of quantum mechanics, we would like the Poisson bracket
on classical observables to reflect the first-order deformation into quantum
observables. We find that this behavior is the case, although the statement is
not as nice as that in the familiar quantum mechanical case.
Let us explain how it works. The factorization algebra of classical observ-
ables has compatible structures of dg commutative algebra and shifted Poisson
bracket. The factorization algebra of quantum observables has, by contrast,
no extra structure: it is simply a factorization algebra valued in cochain
complexes. Modulo 2 , the factorization algebra of quantum observables lives
in an exact sequence:
0 → Obscl → Obsq mod 2 → Obscl → 0.
The boundary map for this exact sequence is an operator, for every open
U ⊂ M,
D : Obscl (U) → Obscl (U).
This operator is a cochain map of cohomological degree 1. Because Obsq is not

a factorization algebra valued in commutative algebras, D is not a derivation
for the commutative algebra structure on Obscl (U).
We can measure the failure of D to be a derivation by the expression
D(ab) − (−1)|a| aDb − (Da)b.
We find that this quantity is the same, up to homotopy, as the shifted Poisson
bracket on classical observables.
We should view this identity as being the analog of the fact that the failure
of the algebra of observables of quantum mechanics to be commutative is
measured, modulo 2 , by the Poisson bracket. Here, we find that the failure
of the factorization algebra of quantum observables to have a commutative
algebra structure compatible with the differential is measured by the shifted
Poisson bracket on classical observables.
This analogy has been strengthened to a theorem by Safronov (2018) and
Rozenblyum (unpublished). Locally constant factorization algebras on R are
equivalent to homotopy associative algebras. Safronov and Rozenblyum show
that locally constant P0 factorization algebras on R are equivalent to ordinary,
unshifted Poisson algebras. Therefore a deformation quantization of a P0
factorization algebra on R into a plain factorization algebra is precisely the
same as a deformation of a Poisson algebra into an associative algebra; in this
sense, our work recovers the usual notion of deformation quantization.
1.5 The Noether Theorem

The second main theorem we prove in this volume is a factorization-algebraic
version of the Noether theorem. The formulation we find of the Noether
theorem is significantly more general than the traditional formulation. We will
start by reminding the reader of the traditional formulation before explaining
our factorization-algebraic generalization.
1.5.1 Symmetries in Classical Mechanics

The simplest version of the Noether theorem applies to classical mechanics.
Suppose we have a classical-mechanical system with a continuous symme-
try given by a Lie algebra g. Let A be the Poisson algebra of operators of the
system, which is the algebra of functions on the phase space. Then, the Noether
theorem, as traditionally phrased, says that there is a central extension
g of g
and a map of Lie algebras

g→A
where A is given the Lie bracket coming from the Poisson bracket. This map
sends the central element in g to a multiple of the identity in A. Further, the
image of g in A commutes with the Hamiltonian.
From a modern point of view, this is easily understood. The phase space of
the classical mechanical system is a symplectic manifold X, with a function H
on it, which is the Hamiltonian. The algebra of operators is the Poisson algebra
of functions on X. If a Lie algebra g acts as symmetries of the classical system,
1.5 The Noether Theorem 9
then it acts on X by symplectic vector fields preserving the Hamiltonian

function. There is a central extension of g that acts on X by Hamiltonian vector
fields, assuming that H 1 (X) = 0.
At the quantum level, the Poisson algebra of functions on X is upgraded
to a noncommutative algebra (which we continue to call A), which is its
deformation quantization. The quantum version of the Noether theorem says
that if we have an action of a Lie algebra g acting on the quantum mechanical
system, there is a central extension g of g (possibly depending on ) and a
Lie algebra map g → A. This Lie algebra map lifts canonically to a map of
associative algebras
g → A,
U
sending the central element to 1 ∈ A.
1.5.2 The Noether Theorem in the Language

of Factorization Algebras
Let us rewrite the quantum-mechanical Noether theorem in terms of factor-
ization algebras on R. As we saw in Section I.3.2, factorization algebras on
R satisfying a certain local-constancy condition are the same as associative
algebras. When translation invariant, these factorization algebras on R are the
same as associative algebras with a derivation. A quantum mechanical system
is a quantum field theory on R, and so has as a factorization algebra Obsq
of observables. Under the equivalence between factorization algebras on R
and associative algebra, the factorization algebra Obsq becomes the associative
algebra A of operators, and the derivation becomes the Hamiltonian.
Similarly, we can view Uc (g) as being a translation-invariant factorization
algebra on R, where the translation action is trivial. In Section I.3.6, we give
a general construction of a factorization algebra – the factorization envelope –
associated with a sheaf of dg Lie algebras on a manifold. The associative
algebra Uc (g) is, when interpreted as a factorization algebra on R, the twisted
factorization envelope of the sheaf ∗R ⊗ g of dg Lie algebras on R. We write
this twisted factorization envelope as Uc (∗R ⊗ g).
The Noether theorem then tells us that there is a map of translation-invariant
factorization algebras
Uc (∗R ⊗ g) → Obsq
on R. We have simply reformulated the Noether theorem in factorization-

algebraic language. This rewriting will become useful shortly, however, when
we state a far-reaching generalization of the Noether theorem.
1.5.3 The Noether Theorem in Quantum Field Theory

Let us now phrase our general theorem. Suppose we have a quantum field
theory (QFT) on a manifold X, with factorization algebra Obsq of observables.
The usual formulation of the Noether theorem starts with a field theory with
some Lie algebra of symmetries. We will work more generally and ask that
there is some sheaf L of homotopy Lie algebras on X that acts as symmetries of
our QFT. (Strictly speaking, we work with sheaves of homotopy Lie algebras
of a special type, which we call local L∞ algebras. A local L∞ algebra is
a sheaf of homotopy Lie algebras whose underlying sheaf is the smooth
sections of a graded vector bundle and whose structure maps are given by
multidifferential operators.) Our formulation of the Noether theorem then takes
the following form.
Theorem 1.5.1 In this situation, there is a canonical -dependent (shifted)

central extension of L, and a map of factorization algebras,
Uc (L) → Obsq ,
from the twisted factorization envelope of L to the factorization algebra of
observables of the quantum field theory.
Let us explain how a special case of this statement recovers the traditional
formulation of the Noether theorem, under the assumption (merely to simplify
the notation) that the central extension is trivial.
Suppose we have a theory with a Lie algebra g of symmetries. One can show
that this implies the sheaf ∗X ⊗ g of dg Lie algebras also acts on the theory.
Indeed, this sheaf is simply a resolution of the constant sheaf with stalk g.
The factorization envelope U(L) assigns the Chevalley–Eilenberg chain
complex C∗ (Lc (U)) to an open subset U ⊂ X. This construction implies that
there is a map of pre-cosheaves Lc [1] → U(L). Applied to L = ∗X ⊗ g, we
find that a g-action on our theory gives a cochain map
∗c (U) ⊗ g[1] → Obsq (U)
for every open. In degree 0, this map 1c ⊗g → Obsq can be viewed as an n−1-
form on X valued in observables. This n − 1 form is the Noether current. (The
other components of this map contain important homotopical information.)
If X = M × R, where M is compact and connected, we get a map
g = H 0 (M) ⊗ Hc1 (R) ⊗ g → H 0 (Obsq (U)).
This map is the Noether charge.
1.6 Brief Orienting Remarks toward the Literature 11
We have seen that specializing to observables of cohomological degree 0 and

the sheaf L = ∗ ⊗ g, we recover the traditional formulation of the Noether
theorem in quantum field theory. Our formulation, however, is considerably
more general.
1.5.4 The Noether Theorem Applied to Two-Dimensional

Chiral Theories
As an example of this general form of the Noether theorem, let us consider the
case of two-dimensional chiral theories with symmetry group G.
In this situation, the symmetry Lie algebra is not simply the constant sheaf
with values in g but the sheaf 0,∗
⊗ g, the Dolbeault complex on valued
in g. In other words, it encodes the sheaf of g-valued holomorphic functions.
This sheaf of dg Lie algebra acts on the sheaf of fields 1/2,∗ (,R) in the
evident way.
Our formulation of the Noether theorem then tells us that there is some
central extension of 0,∗ ⊗ g and a map of factorization algebras

Uc 0,∗
⊗ g → Obsq
from the twisted factorization envelope to the observables of the system of free
fermions.
In Section I.5.5, we calculated the twisted factorization algebra of 0,∗
⊗ g,
and we found that it encodes the Kac–Moody vertex algebra at the level
determined by the central extension.
Thus, in this example, our formulation of the Noether theorem recovers
something relatively familiar: In any chiral theory with an action of G, we
find a copy of the Kac–Moody algebra at an appropriate level.
1.6 Brief Orienting Remarks toward the Literature

Since we began this project in 2008, we have been pleased to see how themes
that animated our own work have gotten substantial attention from others:
• Encoding classical field theories, particularly in the BV formalism, using
L∞ algebras (Hohm and Zwiebach 2017; Jurčo et al. 2019b,a).
• The meaning and properties of derived critical loci (Vezzosi 2020; Joyce
2015; Pridham 2019).
• The role of shifted symplectic structures in derived geometry and enlarged

notions of deformation quantization (Pantev et al. 2013; Calaque et al.
2017; Ben-Bassat et al. 2015; Brav et al. 2019; Pridham 2017; Melani and
Safronov 2018a,b; Safronov 2017; Toën 2014).
• Factorization algebras as a natural tool in field theory, particularly for
topological field theories (Scheimbauer 2014; Kapranov et al. 2016; Benini
et al. 2019, 2020; Beem et al. 2020).
We are grateful to take part in such a dynamic community, where we benefit
from others’ insights and critiques and we also have the chance to share our
own. This book does not document all that activity, which is only partially
represented by the published literature anyhow; we offer only a scattering of
the relevant references, typically those that played a direct role in our own
work or in our learning, and hence exhibit an unfortunate but hard-to-avoid
bias toward close collaborators or interlocutors.
Our work builds, of course, upon the work and insights of generations of
mathematicians and physicists who precede us. As time goes on, we discover
how many of our insights appear in some guise in the past. In particular, it
should be clear how much Albert Schwarz and Maxim Kontsevich shaped our
views and our approach by their vision and by their results, and how much we
gained from engaging with the work of Alberto Cattaneo, Giovanni Felder, and
Andrei Losev.
There is a rich literature on BRST and BV methods in physics that we hope
to help open up to mathematicians, but we do not make an attempt here to
survey it, a task that is beyond us. We recommend Henneaux and Teitelboim
(1992) as a point for jumping into that literature, tracking who cites it and
whom it cites. A nice starting point to explore current activity in Lorentzian
signature is Rejzner (2016), where these BRST/BV ideas cross-fertilize with
the algebraic quantum field theory approach.
1.7 Acknowledgments
The project of writing this book has stretched over more than a decade. In
that course of time, we have benefited from conversations with many people,
the chance to present this work at several workshops and conferences, and
the feedback of various readers. The book and our understanding of this
material is much better due to the interest and engagement of so many others.
Thank you.
We would like to thank directly the following people, although this list is
undoubtedly incomplete: David Ayala, David Ben-Zvi, Dan Berwick-Evans,
Damien Calaque, Alberto Cattaneo, Lee Cohn, Ivan Contreras, Vivek

Dhand, Chris Douglas, Chris Elliott, John Francis, Davide Gaiotto, Dennis
Gaitsgory, Sachin Gautam, Ezra Getzler, Greg Ginot, Ryan Grady, André
Henriques, Rune Haugseng, Theo Johnson-Freyd, David Kazhdan, Si Li,
Jacob Lurie, Takuo Matsuoka, Aaron Mazel-Gee, Pavel Mnev, David Nadler,
Thomas Nikolaus, Fred Paugam, Dmitri Pavlov, Toly Preygel, Eugene
Rabinovich, Kasia Rejzner, Kolya Reshetikhin, Nick Rozenblyum, Pavel
Safronov, Claudia Scheimbauer, Graeme Segal, Jackie Shadlen, Yuan Shen,
Jim Stasheff, Stephan Stolz, Dennis Sullivan, Matt Szczesny, Hiro Tanaka,
Peter Teichner, David Treumann, Alessandro Valentino, Brian Williams,
Philsang Yoo, and Eric Zaslow for helpful conversations. K.C. is particularly
grateful to Graeme Segal for many illuminating conversations about quantum
field theory over the years. O.G. would like to thank the participants in
the Spring 2014 Berkeley seminar and Fall 2014 MPIM seminar for their
interest and extensive feedback. Particular thanks are owed to Brian Williams
and Eugene Rabinovich: their work and collaborations using the Noether
machinery have substantially enhanced that part of this book.
Our work has taken place at many institutions – Northwestern University;
the University of California, Berkeley; the Max Planck Institute for Mathemat-
ics; the Perimeter Institute for Theoretical Physics; and the University of Mass-
chusetts, Amherst – which each provided a supportive environment for writing
and research. Research at the Perimeter Institute is supported by the Gov-
ernment of Canada through Industry Canada and by the Province of Ontario
through the Ministry of Economic Development and Innovation. The Perimeter
Institute is located within the Haldimand Tract, land promised to the Six
Nations (Haudenosaunee) peoples. K.C.’s research at the Perimeter Institute
is supported by the Krembil Foundation and by the NSERC discovery grant
program. We have also benefited from the support of the NSF: K.C. and O.G.
were partially supported by NSF grants DMS 0706945 and DMS 1007168,
and O.G. was supported by NSF postdoctoral fellowship DMS 1204826 and
DMS 1812049. K.C. was also partially supported by a Sloan fellowship and a
Simons fellowship in the early days of this long project.
Finally, during the period of this project, we have depended upon the
unrelenting support and love of our families. In fact, we have both gotten
married and had children over these years! Thank you Josie, Dara, Laszlo, and
Hadrian for being models of insatiable curiosity, ready sources of exuberant
distraction, and endless founts of joy. Thank you Lauren and Sophie for joining
and building our lives together. Amid all the tumult and the moves, you always
encourage us and you always share your wit, wisdom, and warmth. And it
makes all the difference.
PART I
Classical Field Theory

2
Introduction to Classical Field Theory
Our goal here is to describe how the observables of a classical field theory
naturally form a factorization algebra. More accurately, we are interested in
what might be called classical perturbative field theory. “Classical” means that
the main object of interest is the sheaf of solutions to the Euler–Lagrange
equations for some local action functional. “Perturbative” means that we
will only consider those solutions that are infinitesimally close to a given
solution. Much of this part of the book is devoted to providing a precise
mathematical definition of these ideas, with inspiration taken from deformation
theory and derived geometry. In this chapter, then, we will simply sketch the
essential ideas.
2.1 The Euler–Lagrange Equations

The fundamental objects of a physical theory are the observables of a theory,
that is, the measurements one can make in that theory. In a classical field
theory, the fields that appear “in nature” are constrained to be solutions to
the Euler–Lagrange equations (also called the equations of motion). Thus, the
measurements one can make are the functions on the space of solutions to
the Euler–Lagrange equations.
However, it is essential that we do not take the naive moduli space of
solutions. Instead, we consider the derived moduli space of solutions. Since we
are working perturbatively – that is, infinitesimally close to a given solution –
this derived moduli space will be a “formal moduli problem” Lurie (2010,
n.d.). In the physics literature, the procedure of taking the derived critical
locus of the action functional is implemented by the Batalin–Vilkovisky (BV)
formalism. Thus, the first step (Section 3.1.3) in our treatment of classical field
theory is to develop a language to treat formal moduli problems cut out by
17
18 Introduction to Classical Field Theory
systems of partial differential equations on a manifold M. Since it is essential

that the differential equations we consider are elliptic, we call such an object a
formal elliptic moduli problem.
Since one can consider the solutions to a differential equation on any open
subset U ⊂ M, a formal elliptic moduli problem F yields, in particular, a sheaf
of formal moduli problems on M. This sheaf sends U to the formal moduli
space F(U) of solutions on U.
We will use the notation EL to denote the formal elliptic moduli problem of
solutions to the Euler–Lagrange equation on M; thus, EL(U) will denote the
space of solutions on an open subset U ⊂ M.
2.2 Observables
In a field theory, we tend to focus on measurements that are localized in
spacetime. Hence, we want a method that associates a set of observables to
each region in M. If U ⊂ M is an open subset, the observables on U are
Obscl (U) = O(EL(U)),
our notation for the algebra of functions on the formal moduli space EL(U)
of solutions to the Euler–Lagrange equations on U. (We will be more precise
about which class of functions we are using later.) As we are working in the
derived world, Obscl (U) is a differential-graded commutative algebra. Using
these functions, we can answer any question we might ask about the behavior
of our system in the region U.
The factorization algebra structure arises naturally on the observables in a
classical field theory. Let U be an open set in M and V1, . . . ,Vk a disjoint
collection of open subsets of U. Then, restriction of solutions from U to each
Vi induces a natural map
EL(U) → EL(V1 ) × · · · × EL(Vk ).
Since functions pull back under maps of spaces, we get a natural map
Obscl (V1 ) ⊗ · · · ⊗ Obscl (Vk ) → Obscl (U)
so that Obscl forms a prefactorization algebra. To see that Obscl is indeed a

factorization algebra, it suffices to observe that the functor EL is a sheaf.
Since the space Obscl (U) of observables on a subset U ⊂ M is a commuta-
tive algebra and not just a vector space, we see that the observables of a classi-
cal field theory form a commutative factorization algebra (i.e., a commutative
algebra object in the symmetric monoidal category of factorization algebras).
2.4 The P0 Structure 19
2.3 The Symplectic Structure

Above, we outlined a way to construct, from the elliptic moduli problem
associated with the Euler–Lagrange equations, a commutative factorization
algebra. This construction, however, would apply equally well to any system
of differential equations. The Euler–Lagrange equations, of course, have the
special property that they arise as the critical points of a functional.
In finite dimensions, a formal moduli problem that arises as the derived
critical locus (Section 4.1) of a function is equipped with an extra structure:
a symplectic form of cohomological degree −1. For us, this symplectic form
is an intrinsic way of indicating that a formal moduli problem arises as the
critical locus of a functional. Indeed, any formal moduli problem with such a
symplectic form can be expressed (nonuniquely) in this way.
We give (Section 4.2) a definition of symplectic form on an elliptic moduli
problem. We then simply define a classical field theory to be a formal elliptic
moduli problem equipped with a symplectic form of cohomological degree −1.
Given a local action functional satisfying certain nondegeneracy properties,
we construct (Section 4.3.1) an elliptic moduli problem describing the corre-
sponding Euler–Lagrange equations and show that this elliptic moduli problem
has a symplectic form of degree −1.
In ordinary symplectic geometry, the simplest construction of a symplectic
manifold is as a cotangent bundle. In our setting, there is a similar construction:
Given any elliptic moduli problem F, we construct (Section 4.6) a new elliptic
moduli problem T ∗ [−1]F that has a symplectic form of degree −1. It turns
out that many examples of field theories of interest in mathematics and physics
arise in this way.
2.4 The P0 Structure

In finite dimensions, if X is a formal moduli problem with a symplectic form
of degree −1, then the dg algebra O(X) of functions on X is equipped with a
Poisson bracket of degree 1. In other words, O(X) is a P0 algebra (i.e., has a
1-shifted Poisson bracket).
In infinite dimensions, we show that something similar happens. If F is a
classical field theory, then we show that on every open U, the commutative
algebra O(F(U)) = Obscl (U) has a P0 structure. We then show that the com-
mutative factorization algebra Obscl forms a P0 factorization algebra. This is
not quite trivial; it is at this point that we need the assumption that our Euler–
Lagrange equations are elliptic.
3
Elliptic Moduli Problems
The essential data of a classical field theory is the moduli space of solutions to
the equations of motion of the field theory. For us, it is essential that we take
not the naive moduli space of solutions but rather the derived moduli space of
solutions. In the physics literature, the procedure of taking the derived moduli
of solutions to the Euler–Lagrange equations is known as the classical Batalin–
Vilkovisky (BV) formalism.
The derived moduli space of solutions to the equations of motion of a field
theory on X is a sheaf on X. In this chapter, we will introduce a general
language for discussing sheaves of “derived spaces” on X that are cut out by
differential equations.
Our focus in this book is on perturbative field theory, so we sketch the
heuristic picture from physics before we introduce a mathematical language
that formalizes the picture. Suppose we have a field theory and we have found
a solution to the Euler–Lagrange equations φ0 . We want to find the nearby
solutions, and a time-honored approach is to consider a formal series expansion
around φ0 ,
φt = φ0 + tφ1 + t2 φ2 + · · · ,
and to solve iteratively the Euler–Lagrange equations for the higher terms
φn . Of course, such an expansion is often not convergent in any reasonable
sense, but this perturbative method has provided insights into many physical
problems. In mathematics, particularly the deformation theory of algebraic
geometry, this method has also flourished and acquired a systematic geometric
interpretation. Here, though, we work in place of t with a parameter that is
nilpotent, so that there is some integer n such that n+1 = 0. Let
φ = φ0 + φ1 + 2 φ2 + · · · + n φn .
20
Again, the Euler–Lagrange equation applied to φ becomes a system of simpler

differential equations organized by each power of . As we let the order of go
to infinity and find the nearby solutions, we describe the formal neighborhood
of φ0 in the space of all solutions to the Euler–Lagrange equations. (Although
this procedure may seem narrow in scope, its range expands considerably
by considering families of solutions, rather than a single fixed solution. Our
formalism is built to work in families.)
In this chapter, we will introduce a mathematical formalism for this proce-
dure, which includes derived perturbations (i.e., has nonzero cohomological
degree). In mathematics, this formalism is part of derived deformation theory
or formal derived geometry. Thus, before we discuss the concepts specific
to classical field theory, we will explain some general techniques from
deformation theory. A key role is played by a deep relationship between Lie
algebras and formal moduli spaces.
3.1 Formal Moduli Problems and Lie Algebras

In ordinary algebraic geometry, the fundamental objects are commutative
algebras. In derived algebraic geometry, commutative algebras are replaced by
differential-graded commutative algebras concentrated in nonpositive degrees
(or, if one prefers, simplicial commutative algebras; over Q, there is no
difference).
We are interested in formal derived geometry, which is described by
nilpotent dg commutative algebras. It is the natural mathematical framework
for discussing deformations and perturbations. In this section, we give a rapid
overview, but Appendix A.2 examines these ideas at a more leisurely pace and
supplies more extensive references.
Definition 3.1.1 An Artinian dg algebra over a field K of characteristic zero
is a differential-graded commutative K-algebra R, concentrated in degrees ≤ 0,
such that
(i) each graded component Ri is finite dimensional, and Ri = 0 for i 0;
(ii) R has a unique maximal differential ideal m such that R/m = K, and
such that mN = 0 for N 0.
Given the first condition, the second condition is equivalent to the statement
that H 0 (R) is Artinian in the classical sense.
The category of Artinian dg algebras is simplicially enriched in a natural
way. A map R → S is simply a map of dg algebras taking the maximal ideal
mR to that of mS . Equivalently, such a map is a map of nonunital dg algebras
22 Elliptic Moduli Problems
mR → mS . An n-simplex in the space Maps(R,S) of maps from R to S is

defined to be a map of nonunital dg algebras
mR → mS ⊗ ∗ (n ),
where ∗ (n ) is some commutative algebra model for the cochains on the
n-simplex. (Normally, we will work over R, and ∗ (n ) will be the usual de
Rham complex.)
We will (temporarily) let Artk denote the simplicially enriched category of
Artinian dg algebras over k.
Definition 3.1.2 A formal moduli problem over a field k is a functor (of
simplicially enriched categories)
F : Artk → sSets
from Artk to the category sSets of simplicial sets, with the following additional
properties.
(i) F(k) is contractible.

(ii) F takes surjective maps of dg Artinian algebras to fibrations of
simplicial sets.
(iii) Suppose that A,B,C are dg Artinian algebras, and that B → A, C → A
are surjective maps. Then we can form the fiber product B ×A C. We
require that the natural map
F(B ×A C) → F(B) ×F(A) F(C)
is a weak homotopy equivalence.
We remark that such a moduli problem F is pointed: F assigns to k a point,

up to homotopy, since F(k) is contractible. Since we work mostly with pointed
moduli problems in this book, we will not emphasize this issue. Whenever we
work with more general moduli problems, we will indicate it explicitly.
Note that, in light of the second property, the fiber product F(B) ×F(A) F(C)
coincides with the homotopy fiber product.
The category of formal moduli problems is itself simplicially enriched, in
an evident way. If F,G are formal moduli problems, and φ : F → G is a map,
we say that φ is a weak equivalence if for all dg Artinian algebras R, the map
φ(R) : F(R) → G(R)
is a weak homotopy equivalence of simplicial sets.

3.1.1 Formal Moduli Problems and L∞ Algebras

One very important way in which formal moduli problems arise is as the
solutions to the Maurer–Cartan equation in an L∞ algebra. As we will see
later, all formal moduli problems are equivalent to formal moduli problems of
this form.
If g is an L∞ algebra, and (R,m) is a dg Artinian algebra, we will let
MC(g ⊗ m)
denote the simplicial set of solutions to the Maurer–Cartan equation in g ⊗ m.

Thus, an n-simplex in this simplicial set is an element
α ∈ g ⊗ m ⊗ ∗ (n )
of cohomological degree 1, which satisfies the Maurer–Cartan equation

dα + n! ln (α, . . . ,α) = 0.
1
n≥2
It is a well-known result in derived deformation theory that sending R to

MC(g ⊗ m) defines a formal moduli problem (see Getzler 2009, Hinich 2001).
We will often use the notation Bg to denote this formal moduli problem.
If g is finite dimensional, then a Maurer–Cartan element of g ⊗ m is the same
thing as a map of commutative dg algebras
C∗ (g) → R,
which takes the maximal ideal of C∗ (g) to that of R.

Thus, we can think of the Chevalley–Eilenberg cochain complex C∗ (g) as
the algebra of functions on Bg.
Under the dictionary between formal moduli problems and L∞ algebras, a
dg vector bundle on Bg is the same thing as a dg module over g. The cotangent
complex to Bg corresponds to the g-module g∨ [−1], with the shifted coadjoint
action. The tangent complex corresponds to the g-module g[1], with the shifted
adjoint action.
If M is a g-module, then sections of the corresponding vector bundle on Bg
are the Chevalley–Eilenberg cochains with coefficients in M. Thus, we can
define 1 (Bg) to be
1 (Bg) = C∗ (g,g∨ [−1]).
Similarly, the complex of vector fields on Bg is
Vect(Bg) = C∗ (g,g[1]).
Note that, if g is finite dimensional, this is the same as the cochain complex of
derivations of C∗ (g). Even if g is not finite dimensional, the complex Vect(Bg)
is, up to a shift of one, the Lie algebra controlling deformations of the L∞
structure on g.
3.1.2 The Fundamental Theorem of Deformation Theory

The following statement is at the heart of the philosophy of deformation theory:
There is an equivalence of (∞,1)-categories between the category of
differential-graded Lie algebras and the category of formal pointed moduli
problems.
In a different guise, this statement goes back to Quillen’s work on rational

homotopy theory (Quillen 1969). A precise formulation of this theorem has
been proved in Hinich (2001); more general theorems of this nature are
considered in Lurie (n.d.) and Pridham (2010). We recommend Lurie (2010)
as an excellent survey of these ideas.
It would take us too far afield to describe the language in which this state-
ment can be made precise. We will simply use this statement as motivation:
We will only consider formal moduli problems described by L∞ algebras, and
this statement asserts that we lose no information in doing so.
3.1.3 Elliptic Moduli Problems

We are interested in formal moduli problems that describe solutions to
differential equations on a manifold M. Since we can discuss solutions to
a differential equation on any open subset of M, such an object will give a
sheaf of derived moduli problems on M, described by a sheaf of homotopy Lie
algebras. Let us give a formal definition of such a sheaf.
Definition 3.1.3 Let M be a manifold. A local L∞ algebra on M consists of

the following data.
(i) A graded vector bundle L on M, whose space of smooth sections will
be denoted L.
(ii) A differential operator d : L → L, of cohomological degree 1 and
square zero.
(iii) A collection of polydifferential operators
ln : L⊗n → L
for n ≥ 2, which are alternating, are of cohomological degree 2 − n,
and endow L with the structure of L∞ algebra.
3.2 Examples of Elliptic Moduli Problems 25
Definition 3.1.4 An elliptic L∞ algebra is a local L∞ algebra L as above with

the property that (L,d) is an elliptic complex.
Remark: The reader who is not comfortable with the language of L∞ algebras
will lose little by only considering elliptic dg Lie algebras. Most of our
examples of classical field theories will be described using dg Lie algebras
rather than L∞ algebras.
If L is a local L∞ algebra on a manifold M, then it yields a presheaf BL
of formal moduli problems on M. This presheaf sends a dg Artinian algebra
(R,m) and an open subset U ⊂ M to the simplicial set
BL(U)(R) = MC(L(U) ⊗ m)
of Maurer–Cartan elements of the L∞ algebra L(U) ⊗ m (where L(U) refers

to the sections of L on U). We will think of this as the R-points of the formal
pointed moduli problem associated with L(U). One can show, using the fact
that L is a fine sheaf, that this sheaf of formal moduli problems is actually a
homotopy sheaf, that is, it satisfies Čech descent. Since this point plays no role
in our work, we will not elaborate further.
Definition 3.1.5 A formal pointed elliptic moduli problem (or elliptic moduli
problem, for brevity) is a sheaf of formal moduli problems on M that is
represented by an elliptic L∞ algebra.
The base point of the moduli problem corresponds, in the setting of field
theory, to the distinguished solution we are expanding around.

to Scalar Field Theories
We examine the free scalar field before adding interactions. In Section 4.5, we
relate this discussion to the usual formulation in terms of action functionals.
3.2.1 The Free Scalar Field Theory

Let us start with the most basic example of an elliptic moduli problem, that of
harmonic functions. Let M be a Riemannian manifold. We want to consider the
formal moduli problem describing functions φ on M that are harmonic, namely,
functions that satisfy D φ = 0, where D is the Laplacian. The base point of this
formal moduli problem is the zero function.
The elliptic L∞ algebra describing this formal moduli problem is defined by

D
L = C∞ (M)[−1] −
→ C∞ (M)[−2].
This complex is thus situated in degrees 1 and 2. The products ln in this L∞

algebra are all zero for n ≥ 2.
In order to justify this definition, let us analyze the Maurer–Cartan functor
of this L∞ algebra. Let R be an ordinary (not dg) Artinian algebra, and let m be
the maximal ideal of R. The set of 0-simplices of the simplicial set MCL (R) is
the set
{φ ∈ C∞ (M) ⊗ m | D φ = 0}.
Indeed, because the L∞ algebra L is abelian, the set of solutions to the Maurer–
Cartan equation is simply the set of closed degree 1 elements of the cochain
complex L ⊗ m. All higher simplices in the simplicial set MCL (R) are
constant. To see this, note that if φ ∈ L ⊗ m ⊗ ∗ (n ) is a closed element in
degree 1, then φ must be in C∞ (M) ⊗ m ⊗ 0 (n ). The fact that φ is closed
amounts to the statement that D φ = 0 and that ddR φ = 0, where ddR is the de
Rham differential on ∗ (n ).
Let us now consider the Maurer–Cartan simplicial set associated with a
differential-graded Artinian algebra (R,m) with differential dR . The set of
0-simplices of MCL (R) is the set
{φ ∈ C∞ (M) ⊗ m0, ψ ∈ C∞ (M) ⊗ m−1 | D φ = dR ψ}.
(The superscripts on m indicate the cohomological degree.) Thus, the

0-simplices of our simplicial set can be identified with the set R-valued smooth
functions φ on M that are harmonic up to a homotopy given by ψ and also
vanish modulo the maximal ideal m.
Next, let us identify the set of 1-simplices of the Maurer–Cartan simplicial
set MCL (R). This is the set of closed degree 1 elements of L ⊗ m ⊗ ∗ ([0,1]).
Such a closed degree 1 element has four terms:
φ0 (t) ∈ C∞ (M) ⊗ m0 ⊗ 0 ([0,1]),

φ1 (t)dt ∈ C∞ (M) ⊗ m−1 ⊗ 1 ([0,1]),
ψ0 (t) ∈ C∞ (M) ⊗ m−1 ⊗ 0 ([0,1]),
ψ1 (t)dt ∈ C∞ (M) ⊗ m−2 ⊗ 1 ([0,1]).
Being closed amounts to satisfying the three equations
D φ0 (t) = dR ψ0 (t),
3.2 Examples of Elliptic Moduli Problems 27
d
φ0 (t) = dR φ1 (t),
dt
d
D φ1 (t) + ψ0 (t) = dR ψ1 (t).
dt
These equations can be interpreted as follows. We think of φ0 (t) as providing
a family of R-valued smooth functions on M, which are harmonic up to
a homotopy specified by ψ0 (t). Further, φ0 (t) is independent of t up to a
homotopy specified by φ1 (t). Finally, we have a coherence condition among
our two homotopies.
The higher simplices of the simplicial set have a similar interpretation.
3.2.2 Interacting Scalar Field Theories

Next, we will consider an elliptic moduli problem that arises as the Euler–
Lagrange equation for an interacting scalar field theory. Let φ denote a smooth
function on the Riemannian manifold M with metric g. The action functional is

S(φ) = 2 φ D φ + 4! φ dvolg .
1 1 4
M
The Euler–Lagrange equation for the action functional S is
Dφ + 1 3
3! φ = 0,
a nonlinear partial differential equation (PDE), whose space of solutions is
hard to describe.
Instead of trying to describe the actual space of solutions to this nonlinear
PDE, we will describe the formal moduli problem of solutions to this equation
where φ is infinitesimally close to zero.
The formal moduli problem of solutions to this equation can be described as
the solutions to the Maurer–Cartan equation in a certain elliptic L∞ algebra,
which we continue to call L. As a cochain complex, L is
D
L = C∞ (M)[−1] −
→ C∞ (M)[−2].
Thus, C∞ (M) is situated in degrees 1 and 2, and the differential is the
Laplacian.
The L∞ brackets ln are all zero except for l3 . The cubic bracket l3 is the map
l3 : C∞ (M)⊗3 → C∞ (M),
φ1 ⊗ φ2 ⊗ φ3 → φ1 φ2 φ3 .
Here, the copy of C∞ (M) appearing in the source of l3 is the one situated in
degree 1, whereas that appearing in the target is the one situated in degree 2.
If R is an ordinary (not dg) Artinian algebra, then the Maurer–Cartan simpli-

cial set MCL (R) associated with R has for 0-simplices the set φ ∈ C∞ (M) ⊗ m
such that D φ + 3! φ = 0. This equation may look as complicated as the full
1 3
nonlinear PDE, but it is substantially simpler than the original problem. For
example, consider R = R[]/( 2 ), the “dual numbers.” Then φ = φ1 and
the Maurer–Cartan equation becomes D φ1 = 0. For R = R[]/( 4 ), we have
φ = φ1 + 2 φ2 + 3 φ3 and the Maurer–Cartan equation becomes a triple of
simpler linear PDE:
D φ1 = 0, D φ2 = 0, and Dφ3 + 12 φ13 = 0.
We are simply reading off the k components of the Maurer–Cartan equation.

The higher simplices of this simplicial set are constant.
If R is a dg Artinian algebra, then the simplicial set MCL (R) has for
0-simplices the set of pairs φ ∈ C∞ (M)⊗m0 and ψ ∈ C∞ (M)⊗m−1 such that
Dφ + 1 3
3! φ = dR ψ.
We should interpret this as saying that φ satisfies the Euler–Lagrange equations

up to a homotopy given by ψ.
The higher simplices of this simplicial set have an interpretation similar to
that described for the free theory.

to Gauge Theories
We discuss three natural moduli problems: the moduli of flat connections on a
smooth manifold, the moduli of self-dual connections on a smooth 4-manifold,
and the moduli of holomorphic bundles on a complex manifold. In Section 4.6
we relate these directly to gauge theories.
3.3.1 Flat Bundles

Next, let us discuss a more geometric example of an elliptic moduli problem:
the moduli problem describing flat bundles on a manifold M. In this case,
because flat bundles have automorphisms, it is more difficult to give a direct
definition of the formal moduli problem.
Thus, let G be a Lie group, and let P → M be a principal G-bundle equipped
with a flat connection ∇0 . Let gP be the adjoint bundle (associated with P by
3.3 Examples of Elliptic Moduli Problems Related to Gauge Theories 29
the adjoint action of G on its Lie algebra g). Then, gP is a bundle of Lie algebras
on M, equipped with a flat connection that we will also denote ∇0 .
For each Artinian dg algebra R, we want to define the simplicial set DefP (R)
of R-families of flat G-bundles on M that deform P. The question is, “What
local L∞ algebra yields this elliptic moduli problem?”
The answer is L = ∗ (M,gP ), where the differential is d∇0 , the de Rham
differential coupled to our connection ∇0 . But we need to explain how to find
this answer so we will provide the reasoning behind our answer. This reasoning
is a model for finding the local L∞ algebras associated with field theories.
Let us start by being more precise about the formal moduli problem that
we are studying. We will begin by considering only the deformations before
we examine the issue of gauge equivalence. In other words, we start by just
discussing the 0-simplices of our formal moduli problem.
As the underlying topological bundle of P is rigid, we can only deform the
flat connection on P. Let’s consider deformations over a dg Artinian ring R
with maximal ideal m. A deformation of the connection ∇0 on P is given by an
element
A ∈ 1 (M,gP ) ⊗ m0,
since the difference ∇ − ∇0 between any connection and our initial connection
is a gP -valued 1-form. The curvature of the deformed connection ∇0 + A is
F(A) = d∇0 A + 12 [A,A] ∈ 2 (M,gP ) ⊗ m.
Note that, by the Bianchi identity, d∇0 F(A) + [A,F(A)] = 0.
Our first attempt to define the formal moduli functor DefP might be that
our moduli problem only returns deformations A such that F(A) = 0. From a
homotopical perspective, it is more natural to loosen up this strict condition by
requiring instead that F(A) be exact in the cochain complex 2 (M,gP ) ⊗ m of
m-valued 2-forms on M. In other words, we ask for A to be flat up to homotopy.
However, we should also ask that F(A) is exact in a way compatible with the
Bianchi identity, because a curvature always satisfies this condition.
Thus, as a preliminary, tentative version of the formal moduli functor DefP ,
prelim
we will define the 0-simplices DefP (R)[0] by

F(A) = dR B

A ∈ 1 (M,gP ) ⊗ m0,B ∈ 2 (M,gP ) ⊗ m−1 .
d∇0 B + [A,B] = 0
These equations say precisely that there exists a term B making F(A) exact and
that B satisfies a condition that enforces the Bianchi identity on F(A).
prelim
This functor DefP [0] does not behave the way that we want, though.
Consider fixing our Artinian algebra to be R = R[n ]/(n2 ), where n has degree
−n; this is a shifted version of the “dual numbers.” As a presheaf of sets on M,
prelim
the functor DefP [0](R) assigns to each open U the set
{a ∈ 1 (U,gP ),b ∈ 2 (U,gP ) | d∇0 a = 0, d∇0 b = 0}.
In other words, we obtain the sheaf of sets 1cl (−,gP ) × 2cl (−,gP ), which
returns closed 1-forms and closed 2-forms. This sheaf is not, however, a
homotopy sheaf, because these sheaves are not fine; hence, we have higher
cohomology groups.
How do we ensure that we obtain a homotopy sheaf of formal moduli
problems? We will ask that B satisfy the Bianchi constraint up a sequence
of higher homotopies, rather than satisfy the constraint strictly. Thus, the
0-simplices DefP (R)[0] of our simplicial set of deformations are defined by
⎧ ⎫
⎪ ⎪
⎨ A ∈ (M,gP ) ⊗ m
1 0
⎬

1−k F(A) + dB + [A,B] + 2 [B,B] = 0 .
1
⎪ B∈ (M,gP ) ⊗ m
k
⎪
⎩ ⎭
k≥2
Here, d refers to the total differential d∇0 + dR on the tensor product

≥2 (M,gP ) ⊗ m of cochain complexes.
If we let Bi ∈ i (M,gP ) ⊗ m1−i , then the first few constraints on Bi can be
written as
d∇0 B2 + [A,B2 ] + dR B3 = 0,
d∇0 B3 + [A,B3 ] + 12 [B2,B2 ] + dR B4 = 0.
Thus, B2 satisfies the Bianchi constraint up to a homotopy defined by B3 and
so on.
The higher simplices of this simplicial set must relate gauge-equivalent
solutions. If we restricted our attention to ordinary Artinian algebras (i.e., to
dg algebras R concentrated in degree 0, and so with zero differential), then
we could define the simplicial set DefP (R) to be the homotopy quotient of
DefP (R)[0] by the nilpotent group associated with the nilpotent Lie algebra
0 (M,gP )⊗m, which acts on DefP (R)[0] in the standard way (see, for instance,
Kontsevich and Soibelman n.d. or Manetti 2009).
This approach, however, does not extend well to the dg Artinian algebras.
When the algebra R is not concentrated in degree 0, the higher simplices
of DefP (R) must also involve elements of R of negative cohomological
degree. Indeed, degree 0 elements of R should be thought of as homotopies
between degree 1 elements of R, and so should contribute 1-simplices to our
simplicial set.
Another random document with
no related content on Scribd:
Cercopithecidae
[Inhalt]
1. Macacus maurus F. Cuv.
Tafel I und II
mas a.sen., Balg mit Schädel, Loka, c 1300 m am Pik von

Bonthain, Süd Celébes, 21. X 95 (Tafel I und II).
mas b.juv., Balg mit Schädel, Barabatuwa bei Pankadjene,
Süd Celébes, 30. VI 95.
mas ad.,
c. Haut mit Schädel, in Spiritus, Süd Celébes.
mas d.ad., Skelet, Gowa bei Makassar, Süd Celébes.
juv., Skelet,
e. Makassar, Süd Celébes.
Exemplar a ist, der Grösse und Färbung (Tafel I c. ⅓ n. Gr.), sowie

dem Schädel (Tafel II ¾ n. Gr.) nach, ein Greis. Bräunlich, Kopf
heller, Rücken dunkler (R i d g w a y III 1, vorn mehr III 3; R a d d e
29b, am Rücken und an den Extremitäten dunkler 1. Haare des
Nackens und Oberrückens sehr lang, bis 10 cm und darüber, stark
mit Weiss gemischt, Unterrücken mit nur wenig weissen Haaren, am
Gesässe weisslich, viel heller als das von mir Abh. Ber. 1896/7 Nr. 6
Tafel I, 5 abgebildete Exemplar von Tonkean, Hinter- und Innenseite
der Oberschenkel grau, Vorderhals und Oberbrust hellgrau. Die
oberen Extremitäten aussen bräunlich, ebenfalls stark mit Grau
melirt, Hand grau und weiss, Vorderarm innen weiss; am
Unterschenkel wenig Weiss, Hinterhand stark grau und weiss
behaart, Augenbrauen mit viel Weiss, Wangen, Bart, Kinn, Kehle
stark melirt. Im Ganzen liegt ein bräunliches Exemplar mit weisser
und grauer Zeichnung vor gegenüber dem erwähnten schwarzen,
mit Grau gezeichneten von Tonkean. — Die M a a s s e sind am
Balge nicht mit Sicherheit zu nehmen, da das Skelet fehlt: Vom
Vertex zum Anus c 520 mm, von der Lippe zum Anus (alle
Krümmungen mitgemessen) 720, Hinterhand 165.
S c h ä d e l m a a s s e : Grösste Länge 143 mm, Jochbogenbreite 92,
Breite am proc. zygom. os. front. 71.5, geringste Breite zwischen
den Augenhöhlen 6.5, Breite an den Alveolen der Caninen 35.6,
geringste Breite am Pterion 44.4, grösste Breite am Pterion 55.
Exemplar b ist durchweg bräunlich, an den Extremitäten zum Theil

heller, am Gesäss, an der Innenseite des Oberschenkels und an der
Kehle graubräunlich. Maasse am Balge nicht gut zu nehmen.
Hinterhand c 150 mm.
Exemplar c. Färbung wie b.
Exemplar d. Maasse: Grösste Länge des Schädels 131 mm,

Jochbogenbreite 87, Femur 196, Tibia 171, Humerus 169, Radius
175. 12 Caudalwirbel (Abh. Ber. 1896/7 Nr. 6 p. 3 habe ich
irrthümlich 9 Caudalwirbel statt 12 für das Tonkean-Exemplar, 7 statt
10 für das Parepare-Exemplar angegeben). [3]
Es ist nicht ohne Interesse durch den genauen Fundort von a zu

erfahren, dass dieser Affe auch in einer Höhe von 1300 m lebt.
E v e r e t t beobachtete ihn, wenn ich ihn recht verstehe, ebenda
zwischen 6000 und 7000 Fuss (NZ. III, 150 1896), und die Herren
S a r a s i n bemerken, dass er bis auf die höchsten Grate des
Berges, c 9800 Fuss, gehe, wo es schon empfindlich kalt sein kann.
Die starke und dicke Behaarung mag ihn genügend schützen, und
es bleibt zu konstatiren, ob so alte Exemplare aus der Ebene, wo die
Art auch lebt — denn S a r a s i n s erhielten sie in Makassar recht
häufig aus dem Gowaschen — einen ebenso dicken Pelz haben. Bis
jetzt sind mir nur jüngere Individuen aus der Ebene untergekommen,
vielleicht dass Ex. g des Leidener Museums (Cat. MPB. VI, 118 1876
und XI, 32 1892) alt genug wäre, um den Vergleich zu gestatten,
allein es ist ohne Fundort.
Das grosse alte Männchen vom Pik von Bonthain, das die Herren
S a r a s i n als seltenes Jagdstück erbeuteten, erweitert unsere
Kenntniss dieser immer noch ungenügend bekannten Art von
Celébes in, wenigstens für mich, unerwarteter Weise. Ich hatte das
grosse adulte Männchen von Tonkean, Nordost Celébes, das ich
Abh. Ber. 1896/7 Nr. 6 p. 3 beschrieb und Tafel I 5, II 1–2, III 1–2
abbildete, trotz seiner schwarzen Extremitäten (die auch das junge
Weibchen von da — Tafel I 4 — aufwies) für einen alten M. maurus
angesehen, wenn ich auch (S. 4) hervorhob, dass weiteres Material
nöthig sei, um klar zu erkennen, ob die graue Phase an Unterarm
und Unterschenkel auch übersprungen werden könne, und wenn ich
auch ferner hervorhob, dass adulte schwarzgliedrige Exemplare bis
jetzt überhaupt noch nicht bekannt geworden seien (S. 3). Ich nahm
in Folge dessen an, dass M. maurus über ganz Celébes mit
Ausschluss der nördlichen Halbinsel, wo Cynopithecus niger
(Desm.) und nigrescens (Temm.) hausen, vorkomme. Das grosse
alte Exemplar vom Pik von Bonthain aber zeigt nun, dass das,
wenigstens bezüglich der nordöstlichen Halbinsel, nicht zutrifft. Hier
lebt eine andere Art. Sie ist schlank und schwarz, die vom Süden
dagegen gedrungen und bräunlich, mit Grau an den Extremitäten.
Selbst ein sehr altes Individuum von Nordost Celébes mit weissen
Altershaaren würde nie so braun sein können wie das Bonthain-
Exemplar, während die Differenz in der Farbe des Gesässes und der
Parthien darunter vielleicht als Altersdifferenz angesehen werden
kann. Ebenso differiren die Schädel und die Bezahnung. Bei fast
gleicher Schädellänge: 143 gegen 144 (Tonkean), sind alle Maasse
kleiner bei dem älteren Bonthain-Exemplare (vgl. obige Maasse
gegen die l. c. Seite 3 gegebenen), die knöcherne Nasenöffnung ist
breiter, die fossa canina viel flacher, das os zygomaticum lange nicht
so weit ausladend, der ganze Schädel graziler, was neben weiteren
anderen Differenzen auch aus der Abbildung erhellt. Dem hohen
Alter entsprechend sind alle Schädelnähte geschlossen, die crista
sagittalis und occipitalis sehr stark entwickelt, die Zähne hochgradig
abgenutzt, aber viel graziler, die Länge der Zahnreihen kürzer: p 2
bis m 3 sup. 2 33.4 mm gegen 37.2 bei dem Tonkean-Exemplar. In
der Abbildung erscheint die norma facialis (Tafel II Fig. 1) kürzer als
bei dem Tonkean-Exemplare (l. c. Tafel II Fig. 1), was aus der
überhaupt anderen Schädelconfiguration resultirt, besonders aber
steigt das Schädeldach weniger an (siehe norma lateralis, Fig. 2 in
beiden Fällen), wodurch sich die norma facialis verkürzt 3. Die
grössere Schlankheit des Tonkean-Affen kommt deutlich in den
Maassen der Extremitätenknochen zum Ausdrucke gegenüber den
Maassen des vielleicht älteren Exemplares von Parepare, die ich l. c.
Seite 3 gab, und gegenüber denen des ziemlich gleichaltrigen Ex. d
(s. oben). Von dem alten Bonthain-Männchen liegen die Knochen
nicht vor.
Nach alledem ist eine Identificirung beider Formen ausgeschlossen,

und ich trenne daher die von Tonkean in Nordost Celébes ab als
Macacus tonkeanus n. sp.
Früher hatte ich mich dazu nicht berechtigt gefühlt, da M. maurus

sich als sehr variabel in der Färbung erwies, und mir ein notorisch
altes Exemplar gegenüber dem von Tonkean nicht vorlag, überhaupt
wohl nicht bekannt war. Durch weitere Ausbeute wird sich erst
feststellen lassen, wo sich diese beiden Arten gegeneinander
abgrenzen, eventuell ob sie ineinander übergehen. Aus Central
Celébes kennt man, wie ich (l. c. p. 2) bereits hervorhob, bis jetzt nur
Affenfellstücke an Mützen der Eingebornen, die aber die
Extremitäten gerade dazu nicht verwenden; man kann daran nur
constatiren, dass es Macacus-, und [4]nicht Cynopithecus-Felle sind.
Es ist daher auch nicht richtig, wenn A d r i a n i (TTLV. XL, 343 1898)
den Affen von Central Celébes Papio niger nennt 4.
Gewiss wird es noch lange dauern bis eine erschöpfende Kenntniss

der Macacus-Formen von Celébes vorhanden ist, wenn auch seit
W e b e r s Untersuchungen (Zool. Erg. I, 103 1890) Licht auf die bis
dahin herrschende Dunkelheit fiel.
[Inhalt]
2. Cynopithecus niger (Desm.)
mas a.
ad., Balg und Schädel, Lilang bei Kema, Minahassa,
Nord Celébes, 8. II 96.
mas b.
juv., Skelet, Tomohon, Minahassa, Nord Celébes,
17. X 94.
In Bezug auf die Verbreitung von C. niger über Batjan (l. c. p. 7)

möchte ich noch bemerken, dass der Umstand seines
Nichtvorkommens auf dem g a n z n a h e n Halmahéra, ein
genügender Beweis dafür sein dürfte, dass er nach Batjan vom
Menschen gebracht worden ist. Wenn P. L. S c l a t e r (Geogr. Mam.
1899, 228) es noch ganz neuerdings für „möglich“ hält, dass die Art
ebenso nach den Philippinen gekommen sei, so kann ich dem
gegenüber nur wiederholt betonen (s. auch Abh. Ber. 1896/7 Nr. 6 p.
8), dass sie sicher dort nicht zu Hause ist.
[Inhalt]
3. Cynopithecus niger nigrescens (Temm.)
mas a.
ad., Balg mit Schädel, Buol, Nord Celébes, VIII 94.
mas b.
ad., Skelet, zwischen Malibagu und Duluduo, Nord
Celébes.
mas ad.,
c. Schädel, Negeri lama, östlich von Gorontalo,
Nord Celébes.
fem. d.
juv., Schädel, Bone Thal bei Gorontalo, Nord
Celébes, c 500 m.
Ex. a ist ausgesprochen nigrescens (im Gegensatze zu niger) durch

die braune Färbung, besonders an den hinteren Extremitäten, sowie
durch die ungetheilten Gesässchwielen.
Ex. b–d. Hier erschliesse ich die Bestimmung nigrescens nur aus
dem Fundorte. Te m m i n c k (Coup-d’oeil III, 111 1849, s. auch
S c h l e g e l Cat. MPB. VII, 121 1876) hatte nigrescens von
Gorontalo, Tulabello und Tomini von niger aus der Minahassa
abgetrennt; die Fundorte dieser 3 Exemplare sind alle westlich von
Bolang Mongondo. Als ich kürzlich glaubte (Abh. Ber. 1896/7 Nr. 6 p.
7) auf die braunschwarze Färbung (nach Te m m i n c k ) besonders
auf den Schultern und dem Rücken, kein Gewicht legen zu müssen,
weil Exemplare der Minahassa dies auch mehr oder weniger zeigen,
hatte ich noch kein Fell vor Augen; Ex. a aber überzeugt mich, dass
die braune Färbung, besonders an den hinteren Extremitäten, so
ausgesprochen ist, dass sie gar nicht übersehen werden kann, und
ich muss es ausschliessen, dass damit ein individueller Charakter
vorliegt, ebensowenig wie in den ungetheilten Gesässchwielen; hier
handelt es sich gewiss um Charaktere, die an die Localität gebunden
sind, was weitere Exemplare zu bestätigen hätten. Ob die anderen
von Te m m i n c k angegebenen Unterschiede von niger: „face plus
comprimée et queue fort peu apparente“ stichhalten, kann ich
vorläufig nicht entscheiden. Ich muss meine Beobachtung an
lebenden Boliohuto Exemplaren im Walde, dass sie von unten
gräulich waren (l. c. p. 6), nun auch so deuten, dass diese
Farbenwirkung von ihrem bräunlichen Fell herrührte, um so mehr als
gerade die Unterseite der hinteren Extremitäten von Exemplar a
heller braun ist.
Vom Gorontaloschen bis Tomini ist also nur niger nigrescens

bekannt, von der Minahassa niger. Welche Form in Bolang
Mongondo lebt, muss noch constatirt werden, ebenso ob C.
nigrescens oder Macacus, resp. welche Art Macacus, zwischen
Tomini und Parigi vorkommt (bezüglich der Localitäten verweise ich
auf Karte II der „Birds of Celebes“ 1898).
1 Die Deutsche Zoologische Gesellschaft empfahl (Verh. 1894, 103) zur

Farbenbestimmung S a c c a r d o s Chromotaxia, allein dessen 50 Töne
genügen zu einer auch nur etwas feineren Bestimmung nicht, während
R i d g w a y s (Nomenclature of colors 1886) 186 Töne viel weiter führen;
vollständig dienen kann jedoch nur R a d d e s Farbenscala mit ihren c. 900 Tönen,
die aber ihres Preises wegen (60 M) keine grössere Verbreitung gefunden hat,
während R i d g w a y in Vieler Hände ist, so dass es einen Nutzen haben kann, ihn
zu citiren. Es wäre wünschenswerth, dass man von den noch allgemein üblichen
vagen Farbenbezeichnungen, die eine Verständigung erschweren, abginge. ↑
2 H e n s e l sche Bezeichnung. ↑
3 Der Unterkiefer hat links einen überzähligen Backzahn, m 4, und von p 2 ist nur
noch ein kleiner Rest vorhanden, die Alveole zum grössten Theile
verwachsen. ↑
4 A d r i a n i erwähnt daselbst (p. 344) u. a., dass die To Radja fest glauben, die
Affen seien Menschen. ↑
[Inhalt]
Tarsiidae
[Inhalt]
4. Tarsius fuscus Fisch.-Waldh.
Tafel III, Fig. 1–2
mares,
a, b. Bälge mit Schädel, Tomohon, Minahassa, Nord
Celébes, 26. V und 12. VI 94.
mas, c.
Skelet, Tomohon, V 94.
8 Exemplare
d–l. in Spiritus aus der Minahassa, IV 94, und
Tomohon, II und IV 94.
[5]
Je nach dem Alter verschieden gefärbt, jüngere gelber, ältere grauer.

Das kleine Exemplar oben links auf Tafel II Fig. 2 stellt das jüngere
Stadium in c ⅔ nat. Grösse dar. Das andere kleine, Fig. 3, ist T.
sangirensis von Siao in c ½ nat. Grösse. Ich erwähnte schon früher
(Abh. Ber. 1896/7 Nr. 6 p. 8), dass es keine genügende Abbildung
von T. fuscus gäbe, besonders da er früher meist mit T. spectrum
zusammengeworfen worden ist, welche letztere Art vielleicht in
mehrere Rassen zerfällt; dies zu beurtheilen genügt das vorhandene
Material von den verschiedenen Fundorten noch nicht.
O. T h o m a s (TZS. XIV, 381 1898) monirt, dass ich die genaueren

Unterschiede zwischen T. sangirensis und T. philippensis nicht
angegeben habe, während ich die zwischen fuscus und philippensis
wohl aufführe; allein da sangirensis sich fuscus, und nicht
philippensis anschliesst, so wäre es überflüssig gewesen, diese
Unterschiede nochmals ausführlich zu wiederholen. Auch meint er,
dass ich die Tarsen von philippensis als „vollkommen nackt“
bezeichne, übersieht aber, dass ich sie (Abh. Ber. 1896/7 Nr. 6 p. 9
Zeile 7) „so gut wie nackt“ und (Zeile 27) „fast nackt“ nenne, also
genau so wie er sie bezeichnet: „tarsis fere nudis“.
Das Thier spielt eine Hauptrolle in den Erzählungen der Eingebornen

von Central Celébes, die die Baree-Sprache reden (A d r i a n i :
Étude sur la litt. des To Radja, TTLV. XL, 342–53 1898). Es heisst da
nggasi oder tangkasi (Minahassa: tangkasi, Sangi: tenggahĕ, Dajak
ngadju, Bórneo: ingkir).
[Inhalt]
Chiroptera
Megachiroptera
Pteropidae
[Inhalt]
5. Pteropus wallacei Gr.
Tafel IV, Fig. 1
Bälge,
a–c. 2 mares, 1 fem., Tomohon, Minahassa, Nord
Celébes, XI 94 (94, 89, 89 mm) 1.
mas,d.
in Spiritus, Tomohon, 6. IV 94 (87 mm).
Nord und Süd Celébes: Amurang (Mus. Leid.), Lotta, Masarang 3500
Fuss hoch (Mus. Dresd.), Tomohon (Sarasins), Makassar (Brit. Mus.).
H i c k s o n (Nat. N. Cel. 1889, 85) glaubt die Art auch auf der kleinen
Insel Talisse im Norden von Celébes gesehen zu haben.
[Inhalt]
6. Pteropus alecto Temm.
mas,a. Balg, Buol, Nord Celébes, VIII 94 (155 mm).

mas,b. Balg, Bonthain, Süd Celébes, 4. X 95 (160 mm).
fem.,c.Balg, Sokoija, Matanna See, Südost Central Celébes,
6. III 96 (115 mm).
fem.,d.in Spiritus, Tomohon, Minahassa, Nord Celébes, IV 94
(143 mm).
mas,e. in Spiritus, Umgegend von Makassar, Süd Celébes
(170 mm).
[6]
Diese Art, die die Herren S a r a s i n von Nord, Central und Süd
Celébes mitbrachten, soll von Celébes nach Osten bis Neu Guinea
vorkommen. Wie weit sie nach Westen geht, ist noch unsicher; bis jetzt
ist sie westlich von Celébes nur von Bawean, zwischen Java und
Bórneo, genannt. Te m m i n c k beschrieb sie nach einem Exemplar
aus der Minahassa (Mon. Mam. II, 76 1835–41), dieses Exemplar fehlt
aber in J e n t i n k s Catalog des Leidner Museums (XII, 147–8 1888);
es war, der Beschreibung nach, sehr dunkel gefärbt und ebenso sind
die 5 Exemplare der Herren S a r a s i n und die 3 von Celébes im
Dresdner Museum, die aus der Minahassa, Gorontalo und Makassar
stammen. Es fragt sich, ob, bei genügend grossem Materiale von allen
Fundorten, nicht Localrassen zu unterscheiden sein werden. Keinenfalls
genügt D o b s o n s Diagnose von alecto (Cat. Chir. Brit. Mus. 1878,
56).
[Inhalt]
7. Pteropus hypomelanus Temm.
Bälge,
a, b. mas, fem., Makassar, Süd Celébes, IX 95 (123, 124
mm).
4 c–e
fem.,
1. in Spiritus, Makassar, VIII, IX 95 (120, 128, 125, 100
mm).
mares,
f–h. in Spiritus, Insel Bonerate, im Süden von Celébes,
30. XII 94 (115, 118, 113 mm).
Eine über den ganzen Ostindischen Archipel verbreitete Art. Im

Dresdner Museum ist sie auch von Sulu und Talaut vertreten. Ob nicht
Localrassen zu unterscheiden sein werden innerhalb des grossen
Verbreitungsbezirkes der Art, kann nur an der Hand eines grossen
Materiales von überall her beurtheilt werden.
[Inhalt]
8. Pteropus mackloti Temm.
(Pteropus celebensis Schl.)
Balg,a.Tomohon, Minahassa, Nord Celébes, III 94 (127 mm).

fem.,b.in Spiritus, Tomohon (133 mm).
Ich folge J e n t i n k (Webers Zool. Erg. I, 126 1891), der die

Berechtigung von Pt. celebensis Schl. von Celébes als Art oder
Unterart, auf Grund des ihm vorliegenden Materiales von Nord, Central
und Süd Celébes, sowie von Sula, Flores und Timor, nicht anerkannte,
ohne aber dass ich ein gegründetes eigenes Urtheil darüber hätte. Die
Art ist auch von Batjan registrirt und dürfte sich wohl noch anderswo
finden. Das Dresdner Museum besitzt sie ausserdem von der Insel
Saleyer im Süden von Celébes.
[Inhalt]
9. Xantharpyia 2 minor (Dobs.)
D 1873
o b s o n J. As. Soc. Beng. XLII pt. II 203 pl. XIV, 9 (Ohr),
Java, Cynonycteris minor
id.1876
Mon. As. Chir. 32 (Ohr abgeb.), Java, Cynonycteris minor
id.1878
Cat. Chir. Br. M. 73, Java, Cynonycteris minor
Hickson
1889 Nat. N. Cel. 84, Talisse, Cynonycteris minor.
fem.,a.in Spiritus, Tomohon, Minahassa, Nord Celébes III 94
(74 mm).
fem.,
b–d.in Spiritus, Minahassa (72, 67, 67 mm).
Während sonst Xantharpyia amplexicaudata (Geoffr.) mit weiter

Verbreitung über Südasien bis zu den Philippinen und Aru auch von
Celébes registrirt ist (z. B. D o b s o n Cat. Chir. Br. M. 1878, 73,
J e n t i n k Cat. MPB. 1887 IX, 263 und 1888 XII, 151) — das
Originalexemplar kam nach G e o f f r o y aus Timor —, war H i c k s o n
der erste und einzige, der X. minor von der kleinen Insel Talisse im
Norden von Celébes aufführte, und zwar als „very common“. Die von
den Herren S a r a s i n in Nord Celébes gesammelten vier Exemplare
können ihrer geringen Grösse wegen nicht zu amplexicaudata gestellt
werden, und auch desshalb nicht, weil der kleine pm 1 sup. zwischen c
und pm 2 eng eingekeilt ist, statt durch Zwischenräume getrennt
(D o b s o n l. c.). Eine Revision der in Sammlungen vorhandenen
Exemplare von amplexicaudata ist daher angezeigt, zumal alle, die das
Dresdner Museum von Nord Celébes und den Sangi Inseln besitzt, im
Ganzen 35, zu minor gehören, welche Art D o b s o n nach e i n e m
Weibchen von Java beschrieb, von wo sie aber sonst nicht wieder
registrirt worden zu sein scheint. Hingegen hat J e n t i n k (NLM. V, 173
1883) Xantharpyia brachyotis (Dobs.) von Amurang, Minahassa, Nord
Celébes, aufgeführt. [7]Diese Art wurde von D o b s o n von Neu Irland
beschrieben (PZS. 1877, 116 und Cat. 1878, 74) und ist später von
Shortland und Fauro (Salomo Inseln) und von Duke of York
nachgewiesen worden (PZS. 1887, 323 und 1888, 483 und Cat. MPB.
XII, 151 1888). J e n t i n k sagt, dass die zwei Exemplare von Celébes
„in allen Punkten“ mit D o b s o n s Beschreibung übereinstimmen. Die
Unterschiede zwischen X. minor und brachyotis bestehen nach
D o b s o n bei letzterer in viel kürzeren Ohren, längerer Schnauze und
darin, dass pm 1 sup. nur bei jungen Exemplaren vorhanden ist;
D o b s o n erwähnt noch, dass die Schulterdrüse der Männchen durch
dicke gelbe Haarbüschel, wie bei Pteropus, verdeckt seien. Letzteres
zeigen auch die Exemplare von minor von Nord Celébes und den Sangi
Inseln, und zwar nicht nur die alten Männchen, sondern auch die alten
Weibchen; die Haare sind zum Theile lebhaft rostroth.
Aus alle dem dürfte hervorgehen, dass unsere Kenntniss dieser Formen
noch sehr ungenügend ist. Einerseits wäre zu untersuchen, ob X. minor
(von Java und Celébes) nicht identisch ist mit X. brachyotis (vom
Bismarck Archipel und Celébes), oder ob und eventuell wie sich beide
Formen subspecifisch von einander abgrenzen, und andrerseits, wie
sich diese beiden zu X. amplexicaudata verhalten, sowohl artlich, als
auch geographisch. Dazu aber ist ein weit umfangreicheres Material
von den verschiedensten Fundorten nöthig als bis jetzt die besten
Museen enthalten.
[Inhalt]
10. Cynopterus latidens Dobs.
fem.,a.in Spiritus, Tomohon, Minahassa, Nord Celébes, 11. IV

94 (70 mm).
fem.,
b–d.in Spiritus, Minahassa (76, 71, 72 mm).
Diese Art wurde von D o b s o n nach einem Weibchen von der Insel
Morotai bei Halmahéra beschrieben (Cat. Chir. 1878, 86 pl. V, 3,
Zähne), allein schon J e n t i n k (Cat. MPB. XII, 155 1888) führte ein
Männchen von „Menado“ (Celébes) auf, von v. F a b e r gesammelt, das
allerdings in dem Verzeichnisse der F a b e r schen Sammlung (NLM. V,
173 1883) nicht vorkommt (diese Sammlung stammte von Amurang,
siehe p. 170, nicht von Manado). Die 4 von den Herren S a r a s i n aus
Nord Celébes gebrachten Exemplare stimmen nur in sofern nicht mit
D o b s o n s Beschreibung überein, als der Kopf vor und über den
Augen nicht fast schwarz, sondern mit dem Hinterkopfe gleich gefärbt
ist; da alle 4 aber in der Kopffarbe überhaupt etwas untereinander
differiren, indem einige heller sind als andere, und D o b s o n nur e i n
Exemplar von Morotai vorlag, so lässt sich nicht beurtheilen, ob der
hellere Vorderkopf der Celébes-Exemplare ein constanter Charakter ist;
die Kopffarbe mancher Flederhunde variirt bedeutend, und das könnte
daher bei Cynopterus auch statthaben. Keinenfalls fühle ich mich
vorläufig berechtigt, die Celébesform desshalb subspecifisch
abzutrennen; erst weiteres Material wird darüber entscheiden können.
Es ist das Material fast aller Flederhunde in den Museen noch viel zu
unzulänglich, um bei weiter verbreiteten Arten Localrassen mit
Sicherheit unterscheiden zu können; diese Erkenntniss ist der Zukunft
vorbehalten. Wenn wir bei Arten mit grösserem Verbreitungsbezirk oft
stillschweigend annehmen, dass sie fortdauernd von Insel zu Insel
fliegen, so ist dies doch keineswegs bewiesen. Bei der Nähe von Nord-
Celébes und Morotai könnte man a priori ja vielleicht geneigt sein, ein
Überfliegen des Meeres für möglich zu halten; sieht man doch von der
Höhe des Klabat unter Umständen den Vulkan Ternate (M e y e r &
W i g l e s w o r t h : Birds of Celebes I Intr. 52 1898). So kommt z. B.
Pteropus mackloti in Nord Celébes und Batjan vor. Allein nicht jede Art
muss infolge von Isolirung abändern. Auf der anderen Seite sind
Pteropus personatus von Ternate und Pt. wallacei von Nord Celébes
zwar nahe verwandt, aber verschieden, ein Beweis, dass der
Meeresarm sehr wohl auch Fledermäuse trennen kann, so gut wie
Vögel ein selbst viel schmälerer (l. c. 125). Ausnahmsweise wird die
See überflogen, nach der Isolirung aber ist die Abänderung vor sich
gegangen, und die jetzige Constanz der Formen beweist eben, dass ein
weiteres regelmässiges Überfliegen nicht statt findet.
[Inhalt]
Anmerkung
Cynopterus brachyotis (S. Müll.)
ist noch nicht von Celébes registrirt, und wenn auch in Sammlungen
wohl vorhanden, doch mit C. marginatus (Geoffr.) verwechselt worden.
Das Dresdner Museum erhielt sie in den J. 1877 und 1894 aus der
Minahassa, [8]sowie 1893 von Sangi und 1897 von Talaut, im Ganzen
13 Exemplare. J e n t i n k wies in einem lehrreichen Artikel (NLM. XIII,
201 1891) diese von Bórneo, den Andamanen und Nepal bekannte,
aber von D o b s o n in seinem Catalog (1878) vergessene Art von Java
und Sumátra nach; von Sumátra und „Indien“ ist sie auch im Dresdner
Museum. Die folgende Synonymie giebt in Kürze ihre Geschichte:
S 1839
a l . M ü l l e r Tijdschr. Natuur. Gesch. en Phys. V, 146
Pachysoma brachyotis (Bórneo)
1835–1841
Te m m i n c k Mon. Mam. II, 362 Pachysoma brachyotum
(Bórneo)
J .1870
E . G r a y Cat. Monkeys etc. 123 Cynopterus
marginatus var. brachyotis (Bórneo)
D 1873
o b s o n J . As. Soc. Beng. XLI pt. II, 201 pl. XIV, 5 (Ohr)
C. m. var. andamanensis (Andamanen)
id.1876
Mon. As. Chir. 26 Cynopterus brachyotus subsp., Ohr
abgeb. („Andaman Island“, Bórneo)
id.1878
Cat. Chir. Br. Mus. vacat!
S 1887
c u l l y J . As. Soc. Beng. LVI pt. II, 239 Cynopterus
brachyotus (Nepal)
B 1888
l a n f o r d Fauna Br. Ind. Mam. 264 Cynopterus
brachyotus (Andamanen, Bórneo, Nepal)
J e1888
n t i n k Cat. MPB. XII, 154 Cynopterus brachyotis
(Bórneo)
id.1891
NLM. XIII, 202 Cynopterus brachyotis (Bórneo, Java,
Sumátra).

Factorization Algebras in Quantum Field Theory 1St Edition Kevin Costello Online Ebook Texxtbook Full Chapter PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Factorization Algebras in Quantum Field Theory 1St Edition Kevin Costello Online Ebook Texxtbook Full Chapter PDF

Uploaded by

Copyright:

Available Formats

Factorization Algebras in Quantum

Field Theory 1st Edition Kevin Costello

Advanced Topics in Quantum Field Theory 2nd Edition

Relativistic Quantum Field Theory Michael T. Strickland

Quantum Field Theory 2nd Edition Michael V. Sadovskii

Wilson Lines in Quantum Field Theory 2nd Edition Igor

Statistical Approach to Quantum Field Theory Andreas

Nonequilibrium Quantum Field Theory 2nd Edition Esteban

Quantum Field Theory A Diagrammatic Approach 1st

Quantum Field Theory An Introduction 1st Edition Gordon

K e v i n C o s t e l l o is Krembil William Rowan Hamilton Chair in Theoretical

O w e n G w i l l i a m is Assistant Professor in the Department of Mathematics and

1. M. Cabanes and M. Enguehard Representation Theory of Finite Reductive Groups

Cambridge University Press is part of the University of Cambridge.

Contents of Volume 1 page xi

PART I CLASSICAL FIELD THEORY 15

4 The Classical Batalin–Vilkovisky Formalism 43

PART II QUANTUM FIELD THEORY 73

9 Further Aspects of Quantum Observables 144

PART III A FACTORIZATION ENHANCEMENT

13.6 Trivializing the Action on Factorization Homology 313

PART I PREFACTORIZATION ALGEBRAS 21

3.6 The Factorization Envelope of a Sheaf of Lie Algebras 73

PART II FIRST EXAMPLES OF FIELD THEORIES

PART III FACTORIZATION ALGEBRAS 205

8 Factorization Algebras: Examples 243

This is the second volume of our two-volume book on factorization algebras as

1.1 The Factorization Algebra of Classical Observables

means we “impose” the Euler–Lagrange equations by a Koszul complex. In

1.2 The Factorization Algebra of Quantum Observables

1.3 The Physical Importance of Factorization Algebras

As a consequence, each theory has a factorization algebra on R4 that

1.3.1 Correlation Functions from Factorization Algebras

1.3.2 Factorization Algebras and Renormalization Group Flow

In Costello (2011b), a scaling action of R>0 on the collection of theories

factorization algebras is the RG flow.

1.3.3 Factorization Algebras and the Operator

observables is defined to be the limit limr→0 Obsq (D(0,r)) of the space of

O1 (0) · O2 (x) ∈ Obsq (Rn ).

The operator product is defined by expanding the product O1 (0) · O2 (x) as a

1.4 Poisson Structures and Deformation Quantization

the failure of the algebra of quantum operators to be commutative is given, to

0 → Obscl → Obsq mod 2 → Obscl → 0.

This operator is a cochain map of cohomological degree 1. Because Obsq is not

D(ab) − (−1)|a| aDb − (Da)b.

1.5 The Noether Theorem

1.5.1 Symmetries in Classical Mechanics

then it acts on X by symplectic vector fields preserving the Hamiltonian

sending the central element to 1 ∈ A.

1.5.2 The Noether Theorem in the Language

on R. We have simply reformulated the Noether theorem in factorization-

1.5.3 The Noether Theorem in Quantum Field Theory

Theorem 1.5.1 In this situation, there is a canonical -dependent (shifted)

We have seen that specializing to observables of cohomological degree 0 and

1.5.4 The Noether Theorem Applied to Two-Dimensional

1.6 Brief Orienting Remarks toward the Literature

• The role of shifted symplectic structures in derived geometry and enlarged

Damien Calaque, Alberto Cattaneo, Lee Cohn, Ivan Contreras, Vivek

Classical Field Theory

2.1 The Euler–Lagrange Equations

systems of partial differential equations on a manifold M. Since it is essential

Obscl (U) = O(EL(U)),

EL(U) → EL(V1 ) × · · · × EL(Vk ).

1 (Bg) = C∗ (g,g∨ [−1]).

φ0 (t) ∈ C∞ (M) ⊗ m0 ⊗ 0 ([0,1]),

We are simply reading off the k components of the Maurer–Cartan equation.