Second Edition

Gauge Fields
Introduction to Quantum
Theory
L. D. Faddeev
Academy of Sciences U.S.S.R.
Steclov Mathematical Institute, Leningrad
and
A. A. Slavnov
Academy of Sciences U.S.S.R.
Steclov Mathematical Institute, Moscow

Translated from the Russion Edition

by G. B. Pontecorvo
Joint Institute For Nuclear Research, Dubna

CRC Press
T a ylo r & Francis G ro u p
Boca Raton London New York

CRC Press is an im p rin t o f the

Taylor & Francis G roup, an in fo rm a business
First published 1991 by Westview Press

Published 2018 by CRC Press

Taylor & Francis Group
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742

CRC Press is an imprint of the Taylor & Francis Group, an informa business

Copyright © 1991 Taylor & Francis Group LLC

No claim to original U.S. Government works

This book contains information obtained from authentic and highly regarded sources.
Reasonable efforts have been made to publish reliable data and information, but the
author and publisher cannot assume responsibility for the validity of all materials or the
consequences of their use. The authors and publishers have attempted to trace the
copyright holders of all material reproduced in this publication and apologize to
copyright holders if permission to publish in this form has not been obtained. If any
copyright material has not been acknowledged please write and let us know so we may
rectify in any future reprint.

Except as permitted under U.S. Copyright Law, no part of this book may be reprinted,
reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other
means, now known or hereafter invented, including photocopying, microfilming, and
recording, or in any information storage or retrieval system, without written permission
from the publishers.

For permission to photocopy or use material electronically from this work, please access
w w w .copyright.com (http://www.copyright.com/) or contact the Copyright Clearance
Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. CCC is a
not-for-profit organization that provides licenses and registration for a variety of users.
For organizations that have been granted a photocopy license by the CCC, a separate
system of payment has been arranged.

Trademark Notice: Product or corporate names may be trademarks or registered

trademarks, and are used only for identification and explanation without intent to
infringe.

Visit the Taylor & Francis Web site at

http://www.taylorandfrancis.com

and the CRC Press Web site at

http://www.crcpress.com

Library of Congress Cataloging-in-Publication Data

Slavnov, A. A. (Andrei Alekseevich)

[Vvedenie v kvantovuiu teoriiu kalibrovochnykh polei. English]
Gauge fields, introduction to quantum theory /L.D. Faddeev,
A.A. Slavnov; translated from the second Russian edition by
G.B. Pontecorvo.
p. cm.
Translation of: Vvendenie v kvantovuiu teoriiu kalibrovochnykh
polei / A.A. Slavnov, L.D. Faddeev. Izd. 2., perer. i dop. 1988.
1. Gauge fields (Physics) 2. Quantum field theory. I. Faddeev,
L.D. II. Tide.
QC793.3E5S5313 1991 91-846
530.1'435-dc20
ISBN 0-201-52472-4

ISBN 13: 978-0-201-40634-4 (pbk)

Gauge Fields
Introduction to Quantum
Theory
Frontiers in Physics
DAVID PINES/Editor
Volumes of the Series published from 1961 to 1973 are not officially numbered. The
parenthetical numbers shown are designed to aid librarians and bibliographers to check
the completeness of their holdings.
Titles published in this series prior to 1987 appear under either the W. A. Benjamin or
the Benjamin/Cummings imprint; titles published since 1986 appear under the Addison-
Wesley imprint.

(1) N. Bloembergen Nuclear Magnetic Relaxation: A Reprint Volume, 1961

(2) G. F. Chew S-Matrix Theory of Strong Interactions: A Lecture Note and
Reprint Volume, 1961
(3) R. P. Feynman Quantum Electrodynamics: A Lecture Note and Reprint
Volume, 1961
(4) R. P. Feynman The Theory of Fundamental Processes: A Lecture Note
Volume, 1961
(5) L. Van Hove, Problem in Quantum Theory of Many-Particle Systems:
N. M. Hugenholtz A Lecture Note and Reprint Volume, 1961
and L. P. Howland
(6) D. Pines The Many-Body Problem: A Lecture Note and Reprint
Volume, 1961
(7) H. Frauenfelder The Mossbauer Effect: A Review— With a Collection of
Reprints, 1962
(8) L. P. Kadanoff Quantum Statistical Mechanics: Green’s Function Methods
G. Baym in Equilibrium and Nonequilibrium Problems, 1962
(9) G. E. Pake Paramagnetic Resonance: An Introductory Monograph, 1962
[cr. (42)— 2nd edition]
( 10) P. W. Anderson Concepts in Solids: Lectures on the Theory of Solids, 1963
( 11) S. C. Frautschi Regge Poles and S-Matrix Theory, 1963
(12) R. Hofstadter Electron Scattering and Nuclear and Nucleon Structure:
A Collection of Reprints with an Introduction, 1963
(13) A. M. Lane Nuclear Theory: Pairing Force Correlations to Collective
Motion, 1964
(14) R. Omnes Mandelstam Theory and Regge Poles: An Introduction for
M. Froissart Experimentalists, 1963
(15) E. J. Squires Complex Angular Momenta and Particle Physics: A Lecture
Note and Reprint Volume, 1963
(16) H. L. Frisch The Equilibrium Theory of Classical Fluids: A Lecture Note
J. L. Lebowitz and Reprint Volume, 1964
(17) M. Gell-Mann The Eightfold Way (A Review— With a Collection of
Y. Ne’eman Reprints), 1964
(18) M. Jacob Strong-Interaction Physics: A Lecture Note Volume, 1964
G. F. Chew
(19) P. Nozieres Theory of Interacting Fermi Systems, 1964
(20) J. R. Schrieffer Theory of Superconductivity, 1964 (revised 3rd printing, 1983)
(21) N. Bloembergen Nonlinear Optics: A Lecture Note and Reprint Volume, 1965
(22) R. Brout Phase Transitions, 1965
(23) I. M. Khalatnikov An Introduction to the Theory of Superfluidity, 1965
(24) P. G. deGennes Superconductivity of Metals and Alloys, 1966
(25) W. A. Harrison Pseudopotentials in the Theory of Metals, 1966
(26) V. Barger Phenomenological Theories of High Energy Scattering:
D. Cline An Experimental Evaluation, 1967
(27) P. Choquard The Anharmonic Crystal, 1967
 (28) T. Loucks Augmented Plane Wave Method: A Guide to Performing
Electronic Structure Calculations— A Lecture Note and
Reprint Volume, 1967
(29) Y. Ne’eman Algebraic Theory of Particle Physics: Hadron Dynamics in
Terms of Unitary Spin Current, 1967
(30) S. L. Adler Current Algebras and Applications to Particle Physics, 1968
R. F. Dashen
(31) A. B. Migdal Nuclear Theory: The Quasiparticle Method, 1968
(32) J. J. J. Kokkedee The Quark Model, 1969
(33) A. B. Migdal Approximation Methods in Quantum Mechanics, 1969
V. Krainov
(34) R. Z. Sagdeev and Nonlinear Plasma Theory, 1969
A. A. Galeev
(35) J. Schwinger Quantum Kinematics and Dynamics, 1970
(36) R. P. Feynman Statistical Mechanics: A Set of Lectures, 1972
(37) R. P. Feynman Photon-Hadron Interactions, 1972
(38) E. R. Caianiello Combinatorics and Renormalization in Quantum Field
Theory, 1973
(39) G. B. Field, H. Arp, The Redshift Controversy, 1973
and J. N. Bahcall
(40) D. Horn Hadron Physics at Very High Energies, 1973
F. Zachariasen
(41) S. Ichimaru Basic Principles of Plasma Physics: A Statistical Approach,
1973 (2nd printing, with revisions, 1980)
(42) G. E. Pake The Physical Principles of Electron Paramagnetic Resonance,
T. L. Estle 2nd Edition, completely revised, enlarged, and reset, 1973
[cf. (9)— 1st edition]

Volumes published from 1974 onward are being numbered as an integral part of the
bibliography.
43 R. C. Davidson Theory of Nonneutral Plasmas, 1974
44 S. Doniach Green’s Functions for Solid State Physicists, 1974
E. H. Sondheimer
45 P. H. Frampton Dual Resonance Models, 1974
46 S. K. Ma Modern Theory of Critical Phenomena, 1976
47 D. Forster Hydrodynamic Fluctuations, Broken Symmetry, and
Correlation Functions, 1975
A. B. Migdal Qualitative Methods in Quantum Theory, 1977
49 S. W. Lovesey Condensed Matter Physics: Dynamic Correlations, 1980
50 L. D. Faddeev Gauge Fields: Introduction to Quantum Theory, 1980
A. A. Slavnov
51 P. Ramond Field Theory: A Modern Primer, 1981 [cf. 74— 2nd ed.]
52 R. A. Broglia Heavy Ion Reactions: Lecture Notes Vol. I, Elastic and
A. Winther Inelastic Reactions, 1981
53 R. A. Broglia Heavy Ion Reactions: Lecture Notes Vol. II, 1990
A. Winther
54 H. Georgi Lie Algebras in Particle Physics: From Isospin to Unified
Theories, 1982
55 P. W. Anderson Basic Notions of Condensed Matter Physics, 1983
56 C. Quigg Gauge Theories of the Strong, Weak, and Electromagnetic
Interactions, 1983
57 S. I. Pekar Crystal Optics and Additional Light Waves, 1983
 58 S. J. Gates Superspace or One Thousand and One Lessons in
M. T. Grisaru Supersymmetry, 1983
M. Rocek
W. Siegel
59 R. N. Cahn Semi-Simple Lie Algebras and Their Representations, 1984
60 G. G. Ross Grand Unified Theories, 1984
61 S. W. Lovesey Condensed Matter Physics: Dynamic Correlations, 2nd
Edition, 1986
62 P. H. Frampton Gauge Field Theories, 1986
63 J. I. Katz High Energy Astrophysics, 1987
64 T. J. Ferbel Experimental Techniques in High Energy Physics, 1987
65 T. Applequist Modern Kaluza-Klein Theories, 1987
A. Chodos
P. G. O. Freund
66 G. Parisi Statistical Field Theory, 1988
67 R. C. Richardson Techniques in Low-Temperature Condensed Matter
E. N. Smith Physics, 1988
68 J.W.Negele Quantum Many-Particle Systems, 1987
H. Orland
69 E. W. Kolb The Early Universe, 1990
M. S. Turner
70 E. W. Kolb The Early Universe: Reprints, 1988
M. S. Turner
71 V. Barger Collider Physics, 1987
R .J.N . Phillips
72 T. Tajima Computational Plasma Physics, 1989
73 W. Kruer The Physics of Laser Plasma Interactions, 1988
74 P. Ramond Field Theory: A Modern Primer 2nd edition, 1989
jcf. 51— 1st edition]
75 B. F. Hatfield Quantum Field Theory of Point Particles and Strings, 1989
76 P. Sokolsky Introduction to Ultrahigh Energy Cosmic Ray Physics, 1989
77 R. Field Applications of Perturbative QCD, 1989
80 J.F. Gunion The Higgs Hunter’s Guide, 1990
H. E. Haber
G. Kane
S. Dawson
81 R. C. Davidson Physics of Nonneutral Plasmas, 1990
82 E. Fradkin Field Theories of Condensed Matter Systems, 1991
83 L. D. Faddeev Gauge Fields, 1990
A. A. Slavnov
84 R. Broglia Heavy Ion Reactions, Parts I and II, 1990
A. Winther
85 N.Goldenfeld Lectures on Phase Transitions and the Renormalization
Group, 1991
86 R. D. Hazeltine Plasma Confinement, 1991
J. D. Meiss
 EDITOR’S FOREWORD

The problem of communicating in a coherent fashion recent developments in the most

exciting and active fields of physics continues to be with us. The enormous growth in
the number of physicists has tended to make the familiar channels of communication
considerably less effective. It has become increasingly difficult for experts in a given
field to keep up with the current literature; the novice can only be confused. What is
needed is both a consistent account of a field and the presentation of a definite ‘‘point
of view” concerning it. Formal monographs cannot meet such a need in a rapidly
developing field, while the review article seems to have fallen into disfavor. Indeed, it
would seem that the people most actively engaged in developing a given field are the
people least likely to write at length about it.
FRONTIERS IN PHYSICS was conceived in 1961 in an effort to improve the
situation in several ways. Leading physicists frequently give a series of lectures, a
graduate seminar, or a graduate course in their special fields of interest. Such lectures
serve to summarize the present status of a rapidly developing field and may well
constitute the only coherent account available at the time. Often, notes on lectures
exist (prepared by the lecturer himself, by graduate students, or by postdoctoral
fellows) and are distributed on a limited basis. One of the principal purposes of the
FRONTIERS IN PHYSICS Series is to make such notes available to a wider audience
of physicists.
It should be emphasized that lecture notes are necessarily rough and informal,
both in style and content; and those in the series will prove no exception. This is as it
should be. The point of the series is to offer new, rapid, more informal, and, it is hoped,
more effective ways for physicists to teach one another. The point is lost if only elegant
notes qualify.
The informal monograph, representing an intermediate step between lecture notes
and formal monographs, offers an author the opportunity to present his views of a field
which has developed to the point where a summation might prove extraordinarily
fruitful but a formal monograph might be feasible or desirable.
x Editor’s Foreword
During the past decade, the informal text monograph, Gauge Fields, has pro­
vided the reader with a lucid introduction to the role played by gauge fields in
quantum field theory. As its eminent authors note, over the same period gauge
invariant models have evolved from providing an attractive physical hypothesis into a
working theory which describes accurately the physics of elementary particles at
moderate energies. A second edition which contains both supplementary and
improved material is therefore both timely and highly useful, and it gives me pleasure
to welcome once more Drs. Faddeev and Slavnov to FRONTIERS IN PHYSICS.

DAVID PINES
Urbana, Illinois
September, 1990
 Preface to the Second
Revised (Russian) Edition

During the past ten years, since the first edition of this book, gauge invariant models
of elementary particle interactions were transformed from an attractive plausible
hypothesis into a generally accepted theory confirmed by experiments. It was
therefore natural that the development of the methods of gauge fields attracted the
attention of the great majority of specialists in quantum field theory. The new
interesting lines of activity that arose in this period included the formulation of
gauge theories on a lattice, the investigation of non-trivial classical solutions of
the Yang-Mills equations and quantization in their neighborhood, the application
of methods of algebraic topology in gauge field theory. In preparing the second
edition of our book we were confronted with a difficult dilemma: either we were
to extend the book significantly by including a serious discussion of the novel
fields of research, or we would, in the main, adopt the same plan as for the first
edition. We decided in favour of the latter version, since, in our opinion, the
most promising issues mentioned above have not as yet attained a completed form.
Besides, an exposition of these issues would require a significant extension of
the mathematical apparatus utilized. Therefore, in the second edition we limited
ourselves to presenting such supplements that are related in a natural way to the
main content of the first edition, and we also introduced a number of improvements
which, as we hope, should facilitate reading of the book and render it more self-
consistent.
This Preface is being written just at a time, when hopes are arising that a more
fundamental basis is to be developed for elementary particle theory, the theory of
superstrings.

xi
xii Preface to the Second Revised (Russian) Edition

However, independently of whether these hopes come true, gauge field the­
ory, clearly, describes the physics of elementary particles adequately at moderate
energies. Besides, the methods applied in the field theory of relativistic strings
represent a direct generalization of the methods of gauge field theory, to which
this book is devoted. For this reason we consider a new edition of it to be useful,
both for direct applications of the already developed gauge theory and for search
of new ways.
Moscow - Leningrad, 1986 L. D. Faddeev
A. A. Slavnov
 Preface to the Original
(Russian) Edition

Progress in quantum field theory, during the last ten years, is to a great extent
due to the development of the theory of Yang-Mills fields, sometimes called gauge
fields. These fields open up new possibilities for the description of interactions of
elementary particles in the framework of quantum field theory. Gauge fields are
involved in most modern models of strong and also of weak and electromagnetic
interactions. There also arise the extremely attractive prospects of unification of
all the interactions into a single universal interaction.
At the same time the Yang-Mills fields have surely not been sufficiently con­
sidered in modem monographical literature. Although the Yang-Mills theory seems
to be a rather special model from the of view of general quantum field theory, it
is extremely specific and the models used in this theory are quite far from being
traditional. The existing monograph of Konoplyova and Popov, “Gauge Fields”,
deals mainly with the geometrical aspects of the gauge field theory and illuminates
the quantum theory of the Yang-Mills fields insufficiently. We hope that the present
book to some extent will close this gap.
The main technical method, used in the quantum theory of gauge fields, is
the path-integral method. Therefore, much attention is paid in this book to the
description of this alternative approach to the quantum field theory. We have
made an attempt to expound this method in a sufficiently self-consistent manner,
proceeding from the fundamentals of quantum theory. Nevertheless, for a deeper
understanding of the book it is desirable for the reader to be familiar with the
traditional methods of quantum theory, for example, in the volume of the first
four chapters of the book by N. N. Bogolubov and D. V. Shirkov, “Introduction
to the Theory of Quantized Fields”. In particular, we shall not go into details of

xiii
xiv Preface to the Original (Russian) Edition

comparing the Feynman diagrams to the terms of the perturbation-theory expansion,

and of the rigorous substantiation of the renormalization procedure, based on the
R-operation. These problems are not specific for the Yang-Mills theory and are
presented in detail in the quoted monograph.
There are many publications on the Yang-Mills fields, and we shall not go into a
detailed review of this literature to any extent. Our aim is to introduce the methods
of the quantum Yang-Mills theory to the reader. We shall not discuss alternative
approaches to this theory, but shall present in detail that approach, which seems
to us the most simple and natural one. The applications dealt with in the book
are illustrative in character and are not the last work to be said about applications
of the Yang-Mills field to elementary-particle models. We do this consciously,
since the phenomenological aspects of gauge theories are developing and changing
rapidly. At the same time the technique of quantization and renormalization of
the Yang-Mills fields has already become well established. Our book is mainly
dedicated to these specific problems.
We are grateful to our colleagues of the V. A. Steclov Mathematical Institute
in Moscow and Leningrad for numerous helpful discussions of the problems dealt
with in this book.
We would especially like to thank D. V. Shirkov and 0.1. Zav’ylov who read
the manuscript and made many useful comments and E. Sh. Yegoryan for help in
checking the formulas.
Moscow-Leningrad-Kirovsk L. D. Faddeev, A. A. Slavnov
 Contents

PREFACE TO THE SECOND REVISED (RUSSIAN)

EDITION iv

PREFACE TO THE ORIGINAL (RUSSIAN) EDITION vi

1 INTRODUCTION: FUNDAMENTALS OF CLASSICAL GAUGE

FIELD THEORY 1
1.1 Basic Concepts and Notation 1
1.2 Geometrical Interpretation of the Yang-Mills Field 8
1.3 Dynamical Models With Gauge Fields 12

2 QUANTUM THEORY IN TERMS OF PATH INTEGRALS 18

2.1 The Path Integral Over Phase Space 18
2.2 The Path Integral in the Holomorphic Representation 25
2.3 The Generating functional for the 5-matrix in field theory 30
2.4 The 5-Matrix as a Functional on Classical Solutions 39

xv
xvi Contents

2.5 The Path Integral Over Fermi Fields 43

2.6 The Properties of the Path Integral in Perturbation Theory 52

QUANTIZATION OF THE YANG-MILLS FIELD 61

3.1 The Lagrangian of the Yang-Mills Field and the Specific Properties of Its
Quantization 61
3.2 The Hamiltonian Formulation of the Yang-Mills Field and Its Quantiza­
tion 63
3.3 Covariant Quantization Rules and the Feynman Diagram Technique 77
3.4 Interaction with Fields of Matter 87

RENORMALIZATION OF GAUGE THEORIES 97

4.1 Examples of the Simplest Diagrams 97

4.2 The R-Operation and Counterterms 103
4.3 Invariant Regularizations: The Pauli-Villars Procedure 108
4.4 The Method of Higher Covariant Derivatives 113
4.5 Dimensional Regularization 119
4.6 Gauge Fields in Lattice Space-Time 127
4.7 Generalized Ward Identities 134
4.8 The Structure of the Renormalized Action 144
4.9 The Renormalized 5-Matrix 158
4.10 The 5-Matrix in the Covariant Formalism 164
4.11 Anomalous Ward Identities 170

SOME APPLICATIONS AND CONCLUSION 181

5.1 Unified Models of Weak and Electromagnetic Interactions 181

5.2 Asymptotic Freedom. Gauge Theories of Strong Interactions 189
 Contents xvil

BIBLIOGRAPHY NOTES 200

SUPPLEMENT IN PROOF: ANOMALOUS COMMUTATOR OF THE

GAUSS LAW 206

REFERENCES 212

NOTATION 217
 Introduction:
Fundamentals of Classical
Gauge Field Theory

1.1 Basic Concepts and Notation

The theory of gauge fields at present represents the widely accepted theoretical basis
of elementary particle physics. Indeed, the most elaborate model of field theory,
quantum electrodynamics, is a particular case of the gauge theory. Further, models
of weak interactions have acquired an elegant and self-consistent formulation in the
framework of gauge theories. The phenomenological four-fermion interaction has
been replaced by the interaction with an intermediate vector particle, the quantum
of the Yang-Mills field. Existing experimental data along with the requirement of
gauge invariance led to the prediction of weak neutral currents and of new quantum
numbers for hadrons.
Phenomenological quark models of strong interactions also have their most
natural foundation in the framwork of a gauge theory known as quantum chromo­
dynamics. This theory provides a unique possibility of describing, in the framework
of quantum field theory, the phenomenon of asymptotic freedom. This theory also
affords hopes of explaining quark confinement, although this question is not quite
clear.
Finally, the extension of the gauge principle may lead to the gravitational
interaction also being placed in the general scheme of Yang-Mills fields.
So the possibility arises of explaining, on the basis of one principle, all the
hierarchy of interactions existing in nature. The term unified field theory, discred­
2 Introduction: Fundamentals of Classical Gauge Field Theory Chapter 1

ited sometime ago, now acquires a new reality in the framework of gauge field
theories. In the formation of this picture a number of scientists took part. Let us
mention some of the key dates.
In 1953 C. N. Yang and R. L. Mills, for the first time, generalized the principle
of gauge invariance of the interaction of electric charges to the case of interacting
isospins. In their paper, they introduced a vector field, which later became known
as the Yang-Mills field, and within the framework of the classical field theory its
dynamics was developed.
In 1967 L. D. Faddeev, V. N. Popov, and B. De Witt constructed a self-
consistent scheme for the quantization of massless Yang-Mills fields. In the same
year, S. Weinberg and A. Salam independently proposed a unified gauge model
ofweak and electromagnetic interactions, in which the electromagnetic field and
the field of the intermediate vector boson were combined into a multiplet of Yang-
Mills fields. This model was based on the mechanism of mass generation for
vector bosons as a result of a spontaneous symmetry breaking, proposed earlier by
P. Higgs and T. Kibble.
In 1971 G. t’Hooft showed that the general methods of quantization of massless
Yang-Mills fields may be applied, practically without any change, to the case of
spontaneously broken symmetry. Thus the possibility was discovered of construct­
ing a self-consistent quantum theory of massive vector fields, which are necessary
for the theory of weak interactions and, in particular, for the Salam-Weinberg
model.
By 1972 the construction of the quantum theory of gauge fields in the frame­
work of perturbation theory was largely completed. In papers by A.A. Slavnov, by
J. Taylor, by B. Lee and J. Zinn-Justin, and by G. t’Hooft and M. Veltman, various
methods of invariant regularization were developed, the generalized Ward identities
were obtained, and a renormalization procedure was constructed in the framework
of perturbation theory. This led to the construction of a finite and unitary scattering
matrix for the Yang-Mills field.
Since then, the theory of gauge fields has developed rapidly, both theoretically
and phenomenologically. Such development led to the construction of a self-
consistent theory of weak and electromagnetic interactions based on the Weinberg-
Salam model, as well as to a successful description of hadron processes in the
region of asymptotic freedom, where one can apply perturbation theory. From
a purely theoretical point of view, profound relations were established of gauge
theories with differential geometry and topology.
At present the main efforts are directed at the creation of computational meth­
ods not related to the expansion in the coupling constant. Along this way promising
lines of activity are coming into being that raise great hopes. These hopes, how­
ever, have not been fully implemented yet. These include quantization in the
neighborhood of nontrivial classical solutions (instantons), computations on large
computers in the framework of the lattice approximation, application of methods
of the theory of phase transitions, expansion in inverse powers of the number of
colors, and a number of other methods.
 1.1 Basic Concepts and Notation 3

Approaches are also being developed which combine utilization of the quantum
theory of gauge fields and the dispersion technique (sum rules). In brief, hard work
aimed at development of the theory of gauge fields is well under way.
From the above short historical survey we shall pass on to the description of
the Yang-Mills field itself. For this, we must first recall some notation from the
theory of compact Lie groups. More specifically, we shall be interested mainly
in the Lie algebras of these groups. Let Q be a compact semisimple Lie group,
that is a compact group which has no invariant commutative (Abelian) subgroups.
The number of independent parameters that characterize an arbitrary element of
the group (thatis,the dimension is equal to n. Amongthe representations of this
group and its Liealgebra, there exists the representation o f n x n matrices (adjoint
representation). It is generated by the natural action of the group on itself by the
similarity transformations

h —>whu~l; h,w £ Q. ( 1. 1)

Any matrix T in the adjoint representation of the Lie algebra can be represented
by a linear combination of n generators,
JF = T a a a . ( 1 .2 )

For us it is essential that the generators T a can be normalized by the condition

tr(T“T 4) = - 2 6ah. (1.3)

In this case the structure constants tahc that take part in the condition

[Ta,T h] - tabcT c, (1.4)

are completely antisymmetric. The reader unfamiliar with the theory of Lie groups
may keep in mind just these two relationships, which are actually a characterizing
property of the compact semisimple Lie group.
A compact semisimple group is called simple if it has no invariant Lie sub­
groups. A general semisimple group is a product of simple groups. This means that
the matrices of the Lie algebra in the adjoint representation have a blocked-diagram
form, where each block corresponds to one of the simple factors. The generators
of the group can be chosen so that each one has nonzero matrix elements only
within one of the blocks. We shall always have in mind exactly such a choice of
generators, in correspondence with the structure of the direct product.
The simplest example of such a group is the simple group SU(2). The di­
mension of this group equals 3, and the Lie algebra in the adjoint representation is
given by the antisymmetric 3 x 3 matrices; as generators the matrices
0 0 / 0
0 1\ /° -1 0
°\
T1= 0 0 -1 ; T2 = 0 0 0 ; T 3 = f 11 0 00 ; (1.5)
_1 )■’
0 1 0/ \ -l 0 0/ \0 0 0
can be chosen; the structure constants tabc in this base coincide with the completely
antisymmetric tensor eabc.
4 Introduction: Fundamentals of Classical Gauge Field Theory Chapter 1

Besides semisimple compact groups, we shall also deal with the commutative
(Abelian) group U(l). The elements of this group are complex numbers, with
absolute values equal to unity. The Lie algebra of this group is one-dimensional
and consists of imaginary numbers or of real antisymmetric 2 x 2 matrices.
The Yang-Mills field can be associated with any compact semisimple Lie
group. It is given by the vector field A^ix), with values in the Lie algebra of
this group. It is convenient to consider A^(x) to be a matrix in the adjoint repre­
sentation of this algebra. In this case it is defined by its coefficients A*(x):
A»{x) = A l{x )T a ( 1 .6)
with respect to the base of the generators T a.
In the case of the group U( 1) the electromagnetic field A^ix) = iA^(x) is an
analogous object.
We shall now pass on to the definition of the gauge group and its action
on Yang-Mills fields. In the case of electrodynamics the gauge transformation is
actually the well known gradient transformation
A^ix) -> Af,(x) + idn\(z). (1.7)
Let us recall its origin in the framework of the classical field theory. The elec­
tromagnetic field interacts with charged fields, which are described by complex
functions In the equations of motion the field A^{x) always appears in the
following combination:

V ^ = (6^ - A^)tp = (3,, - iA Jilt. (1.8)

The above gradient transformation provides the covariance of this combination with
respect to the phase transformation of the fields ip. If ip transforms according to
the rule
rp(x) — eiXM^(x),
(1.9)
i>{x) — e~iMx)i>{x),
then transforms in the same way. Indeed,
(0„ - A J * - - id„A(*) - ,4„(*)]e<Atrty (*) = eiMx)[8 , - A » (x )M x ).
1 10
( - )

As a result, the equations of motion are also covariant with respect to the trans­
formations ( 1.7 ) and ( 1.9); if the pair 2p (x ),A ^ x) is a solution, then e*A(rty(jc)»
Afi(x) -f idpXix) is also a solution.
In other words, a local change in phase of the field rp(x)t which can be con­
sidered to be the coordinate in the charge space, is equivalent to the appearance
of an additional electromagnetic field. We see here a complete analogy with the
weak equivalence principle in Einstein’s theory of gravity, where a change of the
coordinate system leads to the appearance of an additional gravitational field.
Extending this analogy further, one may formulate the relativity principle in the
charge space. This principle was first introduced by H. Weyl in 1919: The field
 1.1 Basic Concepts and Notation 5

configurations ^(aOM^x) and \()(x)etX^x\ A fi(x) 4 idp A(x) described the same
physical situation. If the construction of theory is based on this principle, then the
above-described way of constructing the equations of motion in terms of covariant
derivatives is the only possible one.
The generalization of this principle to the case of the more complicated charge
space leads to the Yang-Mills theory. Examples of such charge (or internal, as they
are often called) spaces are the isotopic space, the unitary-spin space in the theory
of hadrons, and so on. In all these examples we deal with fields rp(x) that acquire
values in the charge space, which itself is a representation space for some compact
semisimple groups Q(SU(2),SU(3), etc.). The equations of motion for the fields
V>(x) contain the covariant derivative

V„ = 0 , - 1X 4,), (1-11)
where r(. 4 „) is the representation of the matrix A? corresponding to the given
representation of the group Q. For example, if Q = 577(2) and the charge space
corresponds to the two dimensional representation, then the above-mentioned gen­
erators T ° are represented by the Pauli matrices

rcra) = L r4, (i.i2)

2i
where

t1 = (? o)' r2 = 0 o)> 7-3= (o -?)• (U 3 )

and in this case
r C4 „ ) = 5 ^ r«. (1.14)

The transformation of the fields x) analogous to the local phase transforma­

tion in electrodynamics has the following form:
iKx) — ^ ( x ) = r M a i M * ) , (1.15)
where u (x) is a function of x which has its values in the group Q. It is convenient
to consider u(x) to be a matrix in the adjoint representation of the group Q. The
derivative ( 1. 11) will be covariant with respect to this transformation if the field
Afi(x) transforms according to the rule
A^(x) -+ A^(x) = w(x)Ap(x)v~l(x) 4* dftvix)u~l(x). (1.16)
It is not difficult to see that this transformation obeys the group law.The set of
these transformations composes a group that may formally be denoted by

n = J| n . (i.i7 )
x
This group is called the group of gauge transformations.
Often it is convenient to deal with the infinitesimal form of the gauge trans­
formation. Let the matrices u(x) differ infinitesimally from the unit matrix
u(x) = 1 4 a(x) = 1 4 a a(x)Ta, (1.18)
6 Introduction: Fundamentals of Classical Gauge Field Theory Chapter 1

where a(x) belongs to the Lie algebra of the group Q. Then the change of
under such a transformation will be

SA p = d^ a — a] = V ^a, (L 1 9 )

or for the components,

6AI = d^aa - ( 1.20)
The corresponding transformation for xp takes the form
6xp = T(a)xp. ( 1.21)
It is obvious that the group of gradient transformations in electrodynamics is a
particular case of the gauge group.
The existence of covariant derivatives makes it possible to dynamically realize
the relativity principle in the internal space: The field configurations xp(x), A^(x)
and T[u(x)]xp(x),AW p (x) describe the same physical situation. If we take this
principle as a basis for constructing the dynamics, we then automatically come to
the Yang-Mills theory.
The relativity principle means that not only one set of fields but also a whole
class of gauge-equivalent configurations corresponds to the true physical configu­
ration. To be clearer, this principle implies that in the internal charge space there
is no special fixed basis with respect to which the physical fields of matter Q are
represented in terms of components: xp = (xp\, . . . , xpm). Such a basis can be in­
troduced localy at each space-time point; however, there is no physical reason for
fixing its position. The local change of basis is interpreted asa change of the gauge
field, which plays a role analogous to the role of gravitationalor electromagnetic
fields.
The relativity principle leads to a significant formal difference in the description
of the dynamics of gauge fields in comparison with more customary fields such
as, for example, the self-interacting scalar field. In order to work in practice with
classes of equivalent configurations, they must somehow be parameterized; that is,
in each class unique representatives must be chosen. Usually, this is achieved by
imposing a subsidiary condition that eliminates the gauge freedom. This subsidiary
condition is called the gauge condition, or simply gauge. The most frequently used
gauges are the following conditions:
$ l = dpAp = 0 (Lorentz gauge),
= dkA k = 0 (Coulomb gauge),
= Ao = 0 (Hamilton gauge),
$ A = A3 = 0 (axial gauge).
For a general system including both fields A^ and fields xp, the latter may enter into
the gauge condition. Examples of such conditions will be presented in Section 1.3.
In general the gauge condition <£(A,t/>;x) is a family of functionals of A^
and xp, one for each x. For fixed xp; x) is an element of the Lie algebra of
 1.1 Basic Concepts and Notation 7

the group Q, so the number of independent gauge conditions coincides with the
dimension of the gauge group. In the example (1.22) all the conditions are exactly
of such a form. Furthermore, in these examples the gauge conditions are local, that
is $ 04 , ip; x) depends on the values of A^ and ip in the neighborhood of point x.
Let us discuss the requirements to be satisfied by the gauge conditions. The
most important one implies that the system of equations
* (A u , i r ; x ) = 0 (1.23)
has a unique solution u(x) for fixed A^ and ip. This requirement means that in
eachset of equivalent fields there actually exists a unique set of fields A which
satisfies the condition (1.23). This set, considered as a representative of the class,
characterizes uniquely the true physical configuration. Another requirement that is
less fundamental, although important practically, is that Equation (1.23) not be too
complicated and should give a sufficiently explicit solution cj (z), at least in the
framework of perturbation theory.
Equation (1.23) is a system of nonlinear equations for u(x). For local gauge
conditions it is a nonlinear system of partial differential equations. For instance,
for the Lorentz gauge this system of equations takes the following form:
tiT*It — &u7'u, f-4nj J— .An r
(1.24)
L n = uj dpLij)
and for small A^ and a(x) it is rewritten as
□or - f . . . = -d^Ap, (1-25)
where the dots stand for terms of higher order in a. Equation (1.25) can be
uniquely solved with respect to a in the framework perturbation theory if the
operator □ = is supplied with suitable boundary conditions. Such boundary
conditions arise in the description of the dynamics of Yang-Mills fields and will be
discussed in Chapter 3. Nevertheless, beyond the domain of perturbation theory
for large fields A the uniqueness of the solution of Equation (1.24) may fail.
Discussion of this possibility is not within the scope of this book.
A necessary condition for the solvability of the equations (1.23) is the non-
degeneracy of the corresponding Jacobian. Variation of the gauge condition under
an infinitesimal gauge transformation of a defines the linear operator that acts
on a:

M* a ~ I [ ^ 6 A ^ idtia{y) ~ ^ r(a (y » ^ ( 2/)j dy,

(1.26)
This operator plays the role of the Jacobian matrix for the condition (1.23). Non-
degeneracy of the operator M$,
detM *^0 (1.27)
is a necessary condition for the existence of a unique solution for the system (1.23).
8 Introduction: Fundamentals of Classical Gauge Field Theory Chapter 1

For local gauge conditions M$ is a differential operator obtained while lin­

earizing the system (1.23). For example, in the case of the Lorentz gauge condition,
= M l has the form

M loi —D a —dfilAfXfOt]. (1-28)

This operator is uniquely reversible within the framework of perturbation theory
provided boundary conditions are introduced. As was noted above, these conditions
will be discussed in Chapter 3.
It seems appropriate here to explain the meaning of the determinant of the
differential operator M l - We shall represent ML in the form

M l = D (1 (1.29)

where □ ~ 1 is an integral operator, the Green function, defined uniquely upon

introduction of some boundary conditions. Thus, Ml is represented as the product
of the operator □ , which is independent of and an operator of the form I -f K ,
where I\ is the Fredholm operator. As the determinant of the operator M l we
shall take det(I-f K ), which may formally be introduced with the aid of the formula
In det(I + I<) = Trln(I -f K). (1.30)
We shall call the condition (1.27) the admisibility condition for the gauge
condition; it will be frequently discussed further on.

1.2 Geometrical Interpretation of the

Yang-Miils Field ________________

The construction described in the previous section allows an elegant geometrical

interpretation when the Yang-Mills fields play the same role as the Christoffel
symbols in gravitation gheory. Analogously to the latter, the Yang-Mills fields
describe parallel translation in the charge space and determine the curvature of this
space, the fields rp(x) being analogs of tensor fields.
A natural geometrical language for the description of this analogy is yielded
by the fiber-bundle theory. In this theory the concept of connection in the princi­
ple bundle corresponds to the Yang-Mills field. Although the fiber-bundle theory
produces the most adequate language for axiomatization of the classical field the­
ory, in this book, which is addressed mainly to physicists, we shall not use it.
We just point out that the general concept of connection, equivalent to the Yang-
Miils field, appeared in the mathematical literature only in 1950, that is, practically
simultaneously with the work of Yang and Mills.
Let us now explain in what sense the Yang-Mills fields determine parallel
translation. Let j(s ) be a contour in space-time defined by the equation
Xp — £/*($)• (2 .1)
 1.2 Geometrical Interpretation of the Yang-Mills Fie

The vector field 7 ( 5) with components

= (2 .2)
as
is tangential to the curve 7 (s) at each of its points. We shall say that the field tp(x)
undergoes parallel translations along y(s) if at each point of the contour,
V ^ ( x ) | c=l(, ) ^ = 0 , (2.3)
that is, the covariant derivative in the tangential direction is equal to zero.
Generally speaking, parallel translation along a closed contour changes the
field Let us calculate this change for an infinitesimal contour. We shall
consider a contour that has the form of a parallelogram with vertices
(x ,x -f Ajar,a: + A \x + A t x .x -f A2X).
It may be readily verified that if the covariant derivative along this contour is
equal to zero, then the total change in ^(x) corresponding to a whole turn round
the closed contour is equal to
A\2^(x) = T(J7fit/)^(A\x^A2Xv - A\xuA2Xp), (2.4)
where
Fpi, — dvAp — d^Av ■+■lA^, Au]. (2.5)
Indeed, since the covariant derivative along the side (x ,x -f Aix) is equal to
zero, the change in \p(x) corresponding to the change of x along the first contour
equals
AitKz) = V>(z 4- Aix) - ip(x) = dpipAiip = T(An)\p{x)A\x^. (2.6)
Performing analogous calculations for the remaining sides of the parallelogram,
and taking into account the linear dependence of I X ^ ) upon A^ and the fact that
[r(w4M),r(.4„)] = TdApyAA) we obtain the formula (2.4) for the total change in
%l){x). This formula shows that it is natural to call the curvature of the charge
space.
Under gauge transformations, A^(x) changes in the same way as ip(x). This is
because for the construction of Aip(x) we have used only the covariant derivative.
Then from (2.4) it follows that T{T^v(x)) transforms according to the law
r(*>„(x)) -> r(w (*))rcF M„ (z » r(w -1(* » . (2 .7 )
Therefore T^v{x) itself under gauge transformations transforms as
Ffxv(x) w («)^,y(x)w“ l (x). (2 .8)
If we adopt the convention that ip(x) is a vector with respect to gauge transfor­
mations, the rXJF^Oc)) is a tensor of rank two. And T ^ ix ) itself is sometimes
conveniently considered a vector in the adjoint representation.
Our indirect derivation of (2.8) is verified by a straightforward check if one
takes advantage of the explicit expression (2.5) for T^u{x) in terms of A^{x) and
10 Introduction: Fundamentals of Classical Gauge Field Theory Chapter 1

of the transformation law (1.16) for A ^ x ). Thus we conclude our short description
of the geometrical interpretation of the Yang-Mills fields: They describe parallel
translation of vectors in the charge space, and the tensor is the curvature
tensor of this space. The reader familiar with the theory of gravity must surely have
already noticed the complete analogy between A^(x) and Christoffei symbols, and
between T ^ ix ) and the curvature tensor of the gravitational field. To conclude
this analogy, we point out that the tensor F ^ x ) is the commutator of the covariant
derivatives

^ ( * ) = [V|I>V J (2.9)
and the Jacobi identity

[[V„, V * ] , V J 4- cyclic permutations = 0 (2 .10)

leads to the identity

^a^fMu(x) 4- cyclic permutations = 0, (2.11)

where = da^>y(®)—Wtr(ar), T^v{x)\ which is the analog of the Bianchi

identity in the theory of gravity. A similar consideration can be carried out in the
case of the Abelian group U (l). In this case

Tpv(x) = dvAn(x) - dpAvix) = i(duA^(x) — dMAu(x)), (2.12)

which obviously coincides with the tensor of the elecltromagnetic field strength.
The interpretation of F ^ i x ) as the curvature of the charge space, originating with
V. Fock and H. Weyl, is the most natural approach to the geometrization of the
electromagnetic field. Numerous attempts to relate this field to the geometrical
properties of space-time itself have never had any success.
In concluding this section we shall say a few words about the classical dynam­
ics of the Yang-Mills field. Our task is to construct a gauge-invariant Lagrange
function coinciding, in the case of the Abelian group Z7(l), with the Lagrangian of
the electromagnetic field

c = + C m W, (2.13)

where C m describes the gauge-invariant interaction of the fields A^(x) and tp(x)
and is deduced from the free Lagrangian of the fields tp by replacing ordinary
derivatives with covariant ones, and where e plays the role of the electric charge.
This formula may be easily rewritten in a more familiar form if one changes the
normalization of the fields:
A ^x) - eA ^ x ). (2.14)

In this case the factor e ~2 vanishes from the first term but appears instead in the
expression for the covariant derivative,
 1.2 Geometrical Interpretation of the Yang-Mills Field 11

In the following we shall use both methods of normalizing the fields A^(x) without
specially mentioning this.
A natural (and the only possible) generalization of the formula (2.13) to the
case of the simple non-Abelian gauge group is the following expression:

C= ^ "t” (2-15)

The first term may be rewritten also as

( 2 i 6)

where T *v(x) are the components of the matrix T ^ {x ) with respect to the base T a.
Obviously, this Lagrangian is invariant with respect to the gauge transformations
(1.15), and (1.16). In the case of the semisimple group of general form, the
Lagrangian contains r arbitrary constants gi, i = 1 , . . . , r, where r is the number
of invariant simple factors. Then the formula analogous to (2.16) takes the form

( = (2 -17>
*
where i is the index number of a simple factor.
Contrary to electrodynamics, the Lagrangian (2.16) of the Yang-Mills field in
vacuum (that is, in the absence of the fields ^), in addition to the second-order
terms in the fields, contains higher-order terms. This means that Yang-Mills fields
have nontrivial self-interaction. In other words, quanta of the Yang-Mills field
themselves have charges, the interaction of which they transfer. The main specific
feature of the Yang-Mills field dynamics is related to this self-action; therefore we
shall often confine ourselves to the model of the Yang-Mills field in vacuum when
dealing with general problems.
The equations of motion arising from the Lagrangian (2.16) for the Yang-Mills
field in vacuum have the foim

VpTpu = = 0 (2.18)

and written in terms of the A ^

□ A , - dydpAr + lA ^ (d vA„ - + W „ ,A ])] - d ^ A ^ A A = 0, (2.19)

represent a system of second-order equations. These equations are gauge-invariant

in the following sense: If A „ is a solution of (2.19), then A” is also a solution for
any arbitrary u(x). This means that the standard parameterization of the solutions
in terms of the initial conditions M ^(x,<),3<>w4/i(x ,t) at a fixed t) is unfit for
the system (2.19). If we impose the gauge conditions, this obstacle is eliminated;
however, the initial conditions are then not arbitrary but are restricted by the gauge
conditions.
12 Introduction: Fundamentals of Classical Gauge Field Theory Chapter 1

Models of interaction of the Yang-Mills field with fields of matter will be

considered in the next section.

1.3 Dynamical Models with Gauge Fields

The Lagrangian describing the interaction of the Yang-Mills field with spinor fields
is the simplest one. Let a multiplet of spinor fields V’fc(z) realize the representation
T(u) of a simple, compact gauge group Q. Then the Lagrangian has the form

£ = C ym + i^ (x> )> V ^ (x) - m^(x)^(x). (3.1)

We have used here the following notation: C ym is the already familiar Lagrangian
of the Yang-Mills field in vacuum,

&YM -= t f piu• (3.2)

In the scalar product of two spinors the sum is over the indices corresponding
to internal degrees of freedom; for instance, the mass term may be written down
as

mtl>(x)ip(x) = n#jb(x)^jfe(x). (3.3)

Furthermore,
( V ^ ( s ) ) fc = d ^ k(x) - (TOA»{x) ) ) M x), (3.4)

where (T(A^))ki = A *(T (T a))ki and the matrix (T(Ta))ki which in the following
will be denoted simply by Tfa is the matrix of the generator T a in the representation
realized by the fields ^(x). Then
pi/six) = ^jk(x)7M(3MV,Jb(ic) - A^(x)Thxpi(x)). (3.5)
For example, let the gauge group be Q = S U (2), and let the fields tp(x) realize
the fundamental representation of this group. Then

(r(^))jt/ = —^A“(ra)jfci, (3.6)

where r a are the Pauli matrices, and the complete Lagrangian has the form

C = - ^ p d vA l - d p A l + e aicA iliAlf+iii>'ftl ^A “ r a4?J-mij>rp. (3.7)

In the case when the gauge group is the group SU (3) and the spinors rp(x) realize
its fundamental (spinor) representation, the analogous Lagrangian takes the form

C- (d^ + ^ “7“^ -rntpxp, (3.8)

 1.3 Dynamical Models with Gauge fields 13

where f abc are the structure constants of the group St/(3) and the matrices A° are
the well-known Gell-Mann matrices:
/0 1 0 \ /0 -i 0\ 100
Ai = ( 10 0 j ; A2 = * 0 0 ; A3 = 0 1 0
\0 0 0/ Vo 0 0/ 0 00
'0 0 1\ /0 0 - i ' 000 \
A4 = | 0 0 0 ; A5 = 0 0 0) ; A6 = | 000 0 1 |; (3.9)
.1 0 0/ \i 0 Oy 0 10 J
0 0 0\ j /I 0 0
A7 = 0 0 -t ; Ag = 0 1 0
VO i 0/ V3 Vo 0 -2
Renormalization of the fields
^ (* ) - g A l(x ) (3.10)
changes the form of Lagrangians (3.7) and (3.8) to a more familiar one, where g
is involved only in the interaction term.
The latter Lagrangian is used, for example, in the theory of strong interactions.
In this case the spinors ip are identified with the quark fields, the Yang-Mills fields
are called gluons, and the internal space is called the space of colors.
In terms of the variables (3.10) the equations of motion in quantum chromo­
dynamics have the form

d ^ -g lA ^ W = g b * yV>, (3.11)

h n (dni> + A " - mrp = 0, (3.12)

-i (d ^ - A“) - mip = 0 . (3.13)

In the above examples, when the gauge group is simple, all interactions are char­
acterized by a single coupling constant. Such universality of the interactions is a
specific feature of the Yang-Mills theory.
The next useful example is the interaction of the Yang-Mills field with a scalar
field. Let the multiplet of scalar fields <pk(x) realize a real representation T(u) of
the simple compact group Q. Then the gauge-invariant Lagrangian has the form
1 m2
£ = £ ym + - — tpip - V(<p), (3.14)

where the covariant derivative is constructed as above

V ^ = ^ - T ( > 4 <4)9, (3.15)

where <p<p, as before, is a scalar product in the charge space, and V((p) is a form of
third and fourth degree in the fields <p that is invariant with respect to the group Q.
14 Introduction: Fundamentals of Classical Gauge Field Theory Chapter 1

In the case when Q = SU(2) and fields p realize the adjoint representation
p = <pa,a = 1,2,3, the corresponding formula becomes

c = C ym + \(d,<pa - geaheAb
^ f - - A2(v > V )2, (3-16)

where parameters m and A2 play the role of the masses and of the contact-
interaction coupling constants of the scalar fields. The Lagrangian (3.13) itself is,
evidently, of little interest from the viewpoint of physical applications; however,
an insignificant-looking modification leads to the extremely interesting possibility
of describing massive vector fields within the framework of the Yang-Mills theory.
This mechanism for the mass generation of the vector field is called the Higgs
effect. We now proceed to discuss this effect.
We shall continue to deal with the gauge group SU (2) as an example. We first
cnsider the case when the scalar field belongs to the adjoint representation. We
shall use the following Lagrangian:

C - C y m A j ( V ^ a)2 - \2(<pa<pa - p f . (3.17)

This Lagrangian differs from the Lagrangian (3.13), which we examined earlier
by the constant term -A 2//4 and the sign of the term with ip squared. At first
sight, the Lagrangian (3.14) appears to describe particles with imaginary masses
and therefore seems to have no physical meaning. Such a conclusion, however,
would be too hasty. The term with p 1 plays the role of the mass only if <p = 0
is the stable equilibrium point and is, therefore, the potential-energy minimum. In
our case the potential energy is

U (A „ v ) = J

+ A V v * - M2)2] d3z, *,4 = 1,2,3, (3.18)

and the configuration <pa = 0, A* = 0 is a saddle point. The corresponding

equilibrium is unstable. However, stable equilibrium points also exist; they are
the configurations corresponding to zero A* and to constant <p with a fixed length
p 1 = p2. Such A m, <p nullify all the three positive terms that make up the potential
energy. (It should be pointed out that besides these configurations themselves,
gauge transformation of these configurations obviously yields configurations that
are also minima. However, owing to the relativity principle, these configurations
present no new physical information, and we shall not take them into consideration.)
Besides these translation-invariant minima the potential energy has other ones,
for example, minima corresponding to t h e ’t Hooft-Polyakov monopoles. The
energy values for these configurations, however, are higher, so they are only local
minima.
In order to determine the real masses, it is necessary to expand the potential
energy in Taylor’s series around the true minimum. In our case the equilibrium
 1.3 Dynamical Models with Gauge fields 15

point is degenerate. The minimal configurations form a two-dimensional sphere

S 2 with points corresponding to directions of the constant vector <p. We shall note
these directions by n and write the corresponding <p with the index n, so that
<pn = fin. The degeneracy is eliminated if we reduce the configuration space and
take into consideration only fields <p which coincide asymptotically with one of
the (pn at high |a?|. Such a choice, naturally, destroys the invariance under SU(2)
transformations with constant parameters (isotopic invariance). It may be shown
that this condition does not contradict the dynamics and that theories corresponding
to different choices of <pn are physically equivalent The reader familiar with solid-
state physics will, of course, see here an analogy to the theory of ferromagnetics,
in which a choice of the direction of the spontaneous magnetization must be made
in order to formulate the theory itself.
For definiteness, let us choose the vector n to be directed along the third axis:
n = (0,0,1). The corresponding vector <pn is (0,0, fi).
Transition of fields <p(x) with a zero asymptotics at infinity,

<p(x)-> <pn + <p(x) (3.19)

makes the isotopic-symmetry breaking explicit, and the Lagrangian takes the form

c = Cym + | (V ^ 0)2 + + C42)2] + m ,{A \d ^ 2 - A \ d ^ x)

+ g m il^ K A 1^ + C42)]2 - A\WA\ + <p2A\]]

mi = fig; m2 = 2\/2\fi. (3.20)

Although we have explicity broken isotopic invariance, the Lagrangian and the
boundary conditions are invariant under local gauge transformations with func­
tions u(x) tending to unity at infinity. We shall give the explicit form of the gauge
transformations in new variables, confining ourselves to infinitesimal transforma­
tions:

6<pa(x) = - g£abc<pb(x)ac(x) - mi£a3ca c(x). (3 .21)

In order to analyze the spectrum of masses generated by the Lagrangian (3.20),

one must choose representatives in gauge-equivalent classes of fields (z), <p(x)
that is, one must fix the gauge. It is convenient to choose the folowing gauge
condition:

<p\x)=0\ <p2(x) = 0; d„A l(x) = 0. (3.22)

It can be verified that for sufficiently small <p*(.x) the admissibility condition is
fulfilled. Indeed,

= C " 3 " 9 ^ bcd M y i (3-23)

16 Introduction: Fundamentals of Classical Gauge Field Theory Chapter 1

and 6(pl>2 are determined by the formula (3.18). As a result, the operator M
corresponding to our gauge has the form
/«‘ \ ( °> ~ 9^ -m i, g<p2 \ / <*i \
M I a2 I = -gtp3 - mi 0, -g<pl (0 2 . (3.24)
\a3 / \dftA2 + A2 dp, -A^d^ - d^A^, □ / \a3/
At small the determinant of the operator M is
det M = raj detD + 0(<p). (3.25)
Since the first term is not zero, in the framework of perturbation theory
det M ^ O , and the admissibility condition is fulfilled.
We shall now write explicitly the quadratic form determining the mass spec­
trum:

u? + ^ ( U j ,)2+ (A2)2+ ~dMp 3dM<p3 - ^ (< p 3)2. (3.26)

Co = -i(<9„A * - d»Aa

As is seen, our theory in the classical approximation describes two massive

vector fields, one massless vector field, and one massive scalar particle. Therefore,
indeed, two vector fields have acquired masses; however, quanta of two scalar
fields have disappeared from the list of particles.
It is not difficult to construct an Sf7(2)-gauge-invariant model in which all
three vector fields acquire a nonzero mass. For this it is necessary to examine the
complex scalar field multiplet in the two-dimensional (spinor) representation.

^ = t e ) ’ (3-2?)
The gauge-invariant Lagrangian has the form
C = CyM + ( V ^ ) + V ^ - A V * V - (?)2, (3.28)
where
= ^ + (3.29)

and the gauge transformation of the fields p is given by the formula

6p(x) = ^-,graa a(x)p(x). (3.30)

As in the previous case, a stable extremum conesponds to a constant p such that

<p+ p — p2. (3.31)
We see that in this case the set of stable extrema formsa three-dimensional sphere
S 3. Inorder to removethe degeneracy, we choose as a minimum

*>(*)= ( ° ) . (3.32)

It can be verified that the condition

(pi(x) = 0; Im ¥>2(2) = 0 (3.33)
 1.3 Dynamical Models with Gauge fields 17

is an admissible gauge. In this gauge there remains only one scalar field

Passing to fields with zero asymptotics at infinity,

ct( x ) - > y /l f i + cr(x), (3.34)
we obtain the Lagrangian

mi = —p ; m2 = 2A/i, (3.35)
V2
which describes the interaction of three massive vector fields and one massive
scalar field.
The above mechanism will further be used for construction of unified gauge-
invariant models of weak and electromagnetic interactions. We have finished the
discussion of the classical Yang-Mills theory, and we shall now proceed to its
quantization.
References
Bogolubov, N. N. , Shirkov, D. V. , An Introduction to the Theory of Quantized Fields.
Moscow, Nauka, 1976.
Feynman, R. P. , Hibbs, A. R. , Quantum Mechanics and Path Integrals. New York, McGraw-
Hill, 1965.
Vasilev, A. N. , Functional Methods in Quantum Field Theory and Statistics. Leningrad,
Leningrad University Press, 1976.
Popov, V. N. , Path Integrals in Quantum Field Theory and Statistical Physics. Moscow,
Atomizdat, 1976.
Konoplyova, N. P. , Popov, V. N. , Gauge Fields. Moscow, Atomizdat, 1972.
Taylor, J. C. , Gauge Theories of Weak Interactions. Cambridge, Cambridge University
Press, 1976,
Ramond, P. , Field Theory: A Modern Primer. London, Benjamin/Cummings, 1981.
Itzykson, C. , Zuber J.-B. , Quantum Field Theory. New York, McGraw-Hill, 1980.
Yang, C. N. , Mills, R. L. , Phys. Rev., 96, 191, 1954.
Utiyama, R. , Phys. Rev., 101, 1597, 1956.
Salam, A. , Ward, J. C. , Nuovo Cimento, XI, 568, 1959.
213 Glashow, S. L. Gell-Mann, M. , Ann. of Phys., 15, 437, 1961.
Feynman, R. , Acta Phys. Polonica, 24, 697, 1963.
De Witt, B. , Phys. Rev. Lett., 12, 742, 1964.
De Witt, B. , Phys. Rev., 160, 1113; 1195, 1967.
Faddeev, L. D. , Popov, V. N. , Phys. Lett., B25, 30, 1967.
Faddeev, L. D. , Popov, V. N. , Preprint, Institute for Theoretical Physics, Ukrainian Acad.
Sci., Kiev, 1967.
Mandelstam, S. , Phys. Rev., 175, 1580, 1968.
Fradkin, E. S. , Tyutin, I. V. , Phys. Lett., B30, 562, 1969.
Vainstein, A. I. , Khriplovich, I. B. , Yadernaya Fizika, 13, 198, 1971.
Boulware, A. , Ann. Phys., 56, 140, 1970.
Slavnov, A. A. , Faddeev, L. D. , TMP,1 3, 18, 1970.
Higgs, P. W. , Phys. Lett., 12, 132, 1964.
Englert, F. , Brout, R. , Phys. Rev. Lett., 13, 321, 1964.
Kibble, T. W. B. , Phys. Rev., 155, 1554, 1967.
Weinberg, S. , Phys. Rev. Lett., 19, 1264, 1967.
Salam, A. , Elementary Particle Theory. N. Svartholm (ed.). Stockholm, Almquist, Forlag AB,
1968.
tHooft, G. , Nucl. Phys. B35, 167, 1971.
Slavnov, A. A. , TMP, 10, 99, 1972.
Slavnov, A. A. , TMP, 13, 174, 1972.
Taylor, J. C. , Nucl. Phys., B33, 436, 1971.
Lee, B. W. , Zinn-Justin, J. , Phys. Rev., D, 5, 3137, 1972.
tHooft, G. Veltman, M. , Nucl. Phys., B44, 189; B50, 318, 1972.
Ehresmann , Coll.top. Bruxelles, p. 29, 1950.
Lischnerovich, A. , Theory of Connections as a Whole and Holonomy Groups. Moscow,
Inostrannaya Literatura, 1960 (Russian translation).
Weyl, H. , Z.f. Phys., 56, 330, 1929.
Fock, V. , Journ. de Physique, 10, 392, 1929.
Polyakov, A. M. , Lett, to JETP, 20, 430, 1974.
214 tHooft, G. , Nucl. Phys., B79, 2761, 1974.
Faddeev, L. D. Preprint MPI-RAE/Pth Mnchen, 1974.
Belavin, A. A. , Polyakov, A. N. , Tyupkin, Y. , Schwarz, A. S. , Phys. Lett. B59, 85, 1975.
Feynman, R. P. , Rev. Mod. Phys., 20, 367, 1948.
Feynman, R. P. , Phys. Rev., 80, 440, 1950.
Faddeev, L. D. , Les Houches Lecture, Session 20. North-Holland, 1976.
Tobocman, W. , Nuovo Cimento, 3, 1213, 1956.
Berezin, F. A. , The Method of Secondary Quantization. Moscow, Nauka, 1965.
Zavyalov, O. I. , Thesis, Math. Inst Acad. Sci., Moscow, 1970.
Schwinger, J. , Phys. Rev., 75, 651, 1949.
Schwinger, J. , Proc. Nat. Acad. Sci., 37, 452, 1951.
Arefeva, I. A. , Faddeev, L. D. , Slavnov, A. A. , TMP, 21, 311, 1974.
Slavnov, A. A. , TMP, 22, 177, 1975.
Zinn-Justin, J. , Lecture Notes in Physics, Vol. 37, Berlin, Springer-Verlag.
Dirac, P.A.M. , Proc. Roy. Sci., A246, 326, 1958.
Dirac, P.A.M. , Lectures on Quantum Mechanics, New York, Yeshiva University, 1964.
Schwinger, J. , Phys. Rev., 125, 1043, 1962; 127, 324, 1962.
Faddeev, L. D. , TMP, I, 3, 1969.
Slavnov, A. A. , Frolov, S. A. , TMP, 68, 360, 1986.
Popov, V. N. , Faddeev, L. D. , Uspekhi Fiz. Nauk, 111, 427, 1973.
Kramers, H. A. , Rapports du 8e Conseil Solvay, Bruxelles, 1950.
Bethe, H. A. , Phys. Rev., 72, 339, 1947.
Bogolubov, N. N. , Parasyuk, O. S. , Doklady Akad. Nauk USSR, 55, 149, 1955; 100, 429,
1955, Acta Math. 97, 227, 1957.
Zavyalov, O. I. , Renormalized Feynman Diagrams. Moscow, Nauka, 1979 (in Russian).
Slavnov, A. A. , Nucl. Phys., B31, 301, 1971.
Slavnov, A. A. , TMP, 33, 210, 1979.
Bollini, C. G. , Giabidgi, J. T. , Phys. Lett., B40, 566, 1972.
215 Ashmore, J. F. , Nuovo Cimento Lett., 4, 289, 1972.
Wilson, K. , Phys. Rev., D14, 2445 1974.
Creutz, M. , Quarks, Gluons and Lattices, Cambridge, Cambridge University Press, 1985.
Ward, J. C. , Phys. Rev., 77, 2931, 1950.
Fradkin, E. S. , JETP, 29, 288, 1955.
Takahashi, Y. , Nuovo Cimento, 6, 370, 1957.
Slavnov, A. A. , Nucl. Phys., B97, 155, 1975.
Becchi, C. , Rouet, A. , Stora, R. , Comm. Math. Phys., 42, 127, 1975.
Lee, B. W. , Phys. Lett., B46, 214, 1974; Phys. Rev., 9, 933, 1974.
Slavnov, A. A. , Sov. J. of Particles and Nuclei, 5, 303 1975.
Becchi, C. , Rouet, A. , Stora, R. , Renormalization Theory ( G. Velo ,; A. S. Wightman ,
eds.). D. Reidel Publ. Co., 1976; Ann Phys., 98, 287, 1976.
Kallosh, R. , Tyutin, I. , Yadernaya Fizika, 17, 190, 1973.
Kugo, T. , Ojima I. , Suppl. Progr. Theor. Phys., N66, 1979.
Nishijima, K. , Nucl. Phys., B238, 601, 1984.
Adler, S. L. , Phys. Rev., 177, 2426, 1969.
Bell, S. , Jackiw, R. , Nuovo Cimento, A60, 47, 1969.
Slavnov, A. A. , TMP, 7, 13, 1971.
Gross, D. J. , Jackiw, R. , Phys. Rev. D, 6, 477, 1972.
Bouchiat, C. , Illiopoulos, J. , Meyer, P. , Phys. Lett., B38, 519, 1972.
Georgi H. , Glashow, S. L. , Phys. Rev. D, 6, 429, 1977.
Wess, J. , Zumino, B. , Phys. Lett., B37, 95, 1971.
Bardeen, W. A. , Phys. Rev., 184, 1848, 1969.
Zumino, B. In: Relativity, Groups and Topology. Amsterdam, North-Holland, 1983, p. 86.
Stora, R. Ibid., p. 34.
Reyman, A. G. , Semyenov-Tyanshanskij, M. A. , Faddeev, L. D. , Functional Analysis, 18,
64, 1984.
Faddeev, L. D. , Phys. Lett., B145, 81, 1984.
Glashow, S. L. , Nucl. Phys., 22, 579, 1961.
216 Glashow, S. L. , Illiopoulos, J. , Maiani, L. , Phys. Rev., D, 2, 185, 1970.
Bogolubov, N. N. , Struminsky, B. V. , Tavkhelidze, A. N. , JINR Preprint D-1968, 1965.
Hann, M.Y. , Nambu, Y. , Phys. Rev., 139, B1006, 1965.
Pati, J. , Salam A. , Phys. Rev. D, 8, 1240, 1973.
Fritzsch, H. , Gell-Mann, M. , Leutwyller, H. , Phys. Lett., B47, 365, 1973.
Weinberg, S. , Phys. Rev. Lett., 31, 494, 1973.
Stueckelberg, E.C.G. , Peterman, A. , Helv. Phys. Acta, 26, 499, 1953.
Gell-Mann, M. , Low F. , Phys. Rev., 95, 1300, 1954.
Bogolubov, N. N. , Shirkov, D. V. , Doklady Akad. Nauk USSR, 103, 203, 391, 1955.
Bogolubov, N. N. , Shirkov, D. V. , Nuovo Cimento, 3, 845, 1956.
Ovsyannikov, L. V. , Doklady Akad. Nauk, 109, 112, 1956.
Callan, C. , Phys. Rev. D, 2, 2, 1542, 1970.
Symanzik, K. , Comm. Math. Phys., 18, 227, 1970.
tHooft, G. , Report at the Conference on Lagrangian Field Theories, Marseille, 1972.
Gross, D. , Wilczek, F. , Phys. Rev. D, 8, 3633, 1973.
Politzer, H. D. , Phys. Reports, C14, 129, 1974.
Polyakov, A. M. , Phys. Lett., B59, 82, 1975.
Wilson, K. , Erice Lectures. CNLS-321, 1975.
Okun, L. B. , Leptons and Quarks. Moscow, Nauka, 1980.
Huang, K. , Quarks, Leptons, and Gauge Fields, Singapore, World Scientific, 1982.
Nambu, I. , Quarks, Moscow, Mir, 1985. (Russian translation from Japanese).
Faddeev, L. D. , Shatashvili, S. L. , Phys. Lett., 167B, 225, 1986.
Alekseev, Yu. A. , Madajchik, Ya. , Faddeev, L. D. , Shatashvili, S. L. , Preprint R-787,
Leningrad, LOMI, 1987.
Fujikawa, K. , Phys. Rev., D21, 2848, 1980.
Fujikawa, K. , Phys. Lett., 171B, 424, 1986.
Treiman, S. , Jackiw, R. , Witten, E. , Current Algebra and Anomalies. Singapore, World
Scientific, 1985.
Faddeev, L. D. , Phys. Lett., 145B, 82, 1984.
Jo, S.-G. , Phys. Lett., 163B, 353, 1985.
Kobayashi, M. , Seo, K. , Sugamoto, A. , Nucl. Phys., B273, 607, 1986.