Recent Developments in Gauge Theories NATO ADVANCED STUDY INSTITUTES SERIES A series of edited volumes comprising multifaceted studies of contem porary scientific issues by some of the best scientific minds in the world, assembled in cooperation with NATO Scientific Affairs Division. Series B. Physics Recent Volumes in this Series Volume 52 - Physics of Nonlinear Transport in Semiconductors edited by David K. Ferry, J. R. Barker, and C. Jacoboni Volume 53 - Atomic and Molecular Processes in Controlled Thermonuclear Fusion edited by M. R. C. McDowell and A. M. Ferendeci Volume 54 - Quantum Flavordynamics, Quantum Chromodynamics, and Unified Theories edited by K. T. Mahanthappa and James Randa Volume 55 - Field Theoretical Methods in Particle Physics edited by Werner Ruhl Volume 56 - Vibrational Spectroscopy of Molecular Liquids and Solids edited by S. Bratos and R. M. Pick Volume 57-Quantum Dynamics of Molecules: The New Experimental Challenge to Theorists edited by R. G. Woolley Volume 58 - Cosmology and Gravitation: Spin, Torsion, Rotation, and Supergravity edited by Peter G. Bergmann and Venzo De Sabbata Volume 59 - Recent Developments in Gauge Theories edited by G. 't Hooft, C. Itzykson, A. Jaffe, H. Lehmann, P. K. MiUer, I. M. Singer, and R. Stora Volume 60 - Theoretical Aspects and New Developments in Magneto-optics Edited by JozefT. Devreese Volume 61 - Quarks and Leptons: Cargese 1979 edited by Maurice Levy, Jean-Louis Basdevant, David Speiser, Jacques Weyers, Raymond Gastmans, and Maurice Jacob Volume 62 - Radiationless Processes edited by Baldassare Di Bartolo This series is published by an international board of publishers in con junction with NATO Scientific Affairs Division A Life Sciences Plenum Publishing Corporation B Physics London and New York C Mathematical and D. Reidel Publishing Company Physical Sciences Dordrecht, Boston and London D Behavioral and Sijthoff & Noordhoff International Social Sciences Publishers E Applied Sciences Alphen aan den Rijn, The Netherlands, and Germantown U.S.A. Recent Developments in Gauge Theories Edited by G. 't Hooft InstitUte for Theoretical Physics Utrecht, The Netherlands c. Itzykson CENSaclay Gif-sur-Yvette, France A. Jaffe Harvard University Cambridge, Massachusetts H. Lehmann University of Hamburg Hamburg, Federal Republic of Germany P. K. Mitter University ofP aris Paris, France I. M. Singer University of California Berkeley, California and R. Stora Center of Theoretical Physics, CNRS Marseille, France PLENUM PRESS. NEW YORK AND LONDON Published in cooperation with NATO Scientific Affairs Division Library of Congress Cataloging in Publication Data Main entry under title: Recent developments in gauge theories. (NATO advanced study institutes series: Series B, Physics; v. 59) "Published in cooperation with NATO Scientific Affairs Division." Includes index. 1. Gauge fields (Physics)- Addresses, essays, lectures. I. 't Hooft, G. II. Series. QC793.3F5R42 530.1'43 80-18528 ISBN 978-1-4684-7573-9 ISBN 978-1-4684-7571-5 (eBook) DOl 10.1007/978-1-4684-7571-5 Proceedings of the NATO Advanced Study Institute on Recent Developments in Gauge Theories, held in Cargese, Corsica, August 26-September 8, 1979. © 1980 Plenum Press, New York Softcover reprint of the hardcover 1st edition 1980 A Division of Plenum Publishing Corporation 227 West 17th Street, New York, N.Y. 10011 All rights reserved No part of this book may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, microfilming, recording, or otherwise, without written permission from the Publisher PREFACE Almost all theories of fundamental interactions are nowadays based on the gauge concept. Starting with the historical example of quantum electrodynamics, we have been led to the successful unified gauge theory of weak and electromagnetic interactions, and finally to a non abelian gauge theory of strong interactions with the notion of permanently confined quarks. The.early theoretical work on gauge theories was devoted to proofs of renormalizability, investigation of short distance behaviour, the discovery of asymptotic freedom, etc .. , aspects which were accessible to tools extrapolated from renormalised perturbation theory. The second phase of the subject is concerned with the problem of quark confinement which necessitates a non-perturbative understanding of gauge theories. This phase has so far been marked by the introduc tion of ideas from geometry, topology and statistical mechanics in particular the theory of phase transitions. The 1979 Cargese Institute on "Recent Developments on Gauge Theories" was devoted to a thorough discussion of these non-perturbative, global aspects of non-abelian gauge theories. In the lectures and seminars reproduced in this volume the reader wilf find detailed reports on most of the important developments of recent times on non perturbative gauge fields by some of the leading experts and innovators in this field. Aside from lectures on gauge fields proper, there were lectures on gauge field concepts in condensed matter physics and lectures by mathematicians on global aspects of the calculus of variations, its relation to geometry and topology, and related topics. The presence of mathematicians as enthusiastic participants and masterful lecturers in this school deserves special mention. We hope this trend will continue in the future and that, in the last quarter of this century, common concerns about the fundamental interactions will bring ever closer the physical and mathematical communities as in the days of yore. We wish to express our gratitude to NATO whose generous financial contribution made it possible to organise this school. We also thank the Centre National de la Recherche Scientifique, a the Delegation la Recherche Scientifique et Technique, the v vi PREFACE C.E.N. de Saclay, as well as the University of Hamburg for financial help. We thank the University of Nice for making available to us the facilities of the Institut d'Etudes Scientifiques de Cargese. Grateful thanks are due to Marie-France Hanseler for much help with the material aspects of the organisation. Last but not least we thank the lecturers and participants for their enthusiastic involvement which contributed much to the scientific atmosphere of the school. G. 't Hooft C. Itzykson A. Jaffe H. Lehmann P.K. Mitter I.M. Singer R. Stora CONTENTS Remarks on Morse Theory . . . . . . . . . . . . . . . . . 1 M.F. Atiyah Morse Theoretic Aspects of Yang-Mills Theory . . • . . . 7 R. Bott A Semiclassical Approach to the Strong Coupling Physics of QCD . . . . . . . . 29 C.G. Callan, Jr. Lattice Gauge Theories 45 J. Glimm Some Results and Comments on Quantized Gauge Fields .............. . S3 J. Frohlich String States in Q.C.D. 83 J.-L. Gervais and A. Neveu Why Do We Need Local Gauge Invariance in Theories with Vector Particles? An Introduction. (Lecture I) 101 G. 't Hooft Which Topological Features of a Gauge Theory can be Responsible for Permanent Confinement? (Lecture II) ...... . 117 G.'t Hooft Naturalness, Chiral Symmetry, and Spontaneous Chiral Symmetry Breaking. (Lecture III) 135 G. 't Hooft Introduction to Lattice Gauge Theories 159 C. Itzykson vii viii CONTENTS Classical Gauge Theories and Their Quantum Role . • • . . • . . • • • • • • 189 A. Jaffe The Coupling Constant in a ~4 Field Theory 201 J. GI~ and A. Jaffe Exact Instanton Gases • • 205 M. Luscher Properties of Lattice Gauge Theory Models at Low Temperatures . . . . . . . 217 G. Mack Geometry of the Space of Gauge Orbits and the Yang-Mills Dynamical System . . 265 P .K. Mitter On the Str~cture of the Phases in Lattice Gauge Theories •........ 293 G. Parisi Superalgebras and Confinement in Condensed Matter Physics . . .. •....• 303 v. Poenaru Cutoff Dependence in Lattice ~4 Theory 313 K. Symanzik 4 Gauge Concepts in Condensed Matter Physics 331 G. Toulouse Monte-Carlo Calculations for the Lattice Gauge Theory . . . • . . . . . . . 363 K.G. Wilson The lIN Expansion in Atomic and Particle Physics ..•.. .• . . . 403 E. Witten Generalized Non-Linear a-Models with. Gauge Invariance 421 J. Zinn-Justin Index 437 REMARKS ON MORSE THEORY M. F. Atiyah University of Oxford 'Mathematical Institute 24-29 St. Giles, oxford §l. INTRODUCTION Morse theory is a topological approach to the Calculus of Variations. It aims to relate the critical points of a functional to the topology of the function space on which the functional is defined. It is only directly applicable in special rather res trictive conditions, notably for problems involving one independent variable. However I will discuss a number of special examples, in some of which the Morse theory really works, and others in which it clearly fails but where nevertheless some aspects appear still to survive. These examples include those of physical interest and it would be interesting to investigate these further. One can make a number of speculations in this direction. §2. GEODESICS ON A GROUP The classical example for which Morse developed his theory is that of geodesics. This can be formulated as follows'. Given a compact Riemannian manifold M we consider closed paths in M described by a periodic function f(t) with values in M or more formally a map where Sl is the unit circle. As our functional we take E(f) where f' (t) is a tangent vector to M and If' (t) I denotes its length in the Riemannian metric of M. The critical points of 2 M. F. ATIYAH E , that is the solutions of the corresponding Euler-Lagrange equatioris, correspond to closed geodesics on M parametrized by arc length. As a special example we may take M = G a compact Lie group with bi-invariant metric. The closed geodesics through the identity of G are just the closed I-parameter subgroups, and so are completely known in terms of the group structure of G. Using the Morse theory one can then derive information about the topology of the loop space n (G) , i. e. the space of closed paths based at the· identity. This programme was carried through many years ago by Bott [3J. As a simple example if G = SU(2) , the 3-sphere, the closed geodesics are the great circles (including their n-fold iterates and for n = 0 the degenerate point-map) . From this one finds that the homology of nG is infinite cyclic in every even dimension and zero in odd<dimensions. §3. YANG-MILLS IN TWO DIMENSIONS We consider Yang-Mills theory over a compact two-dimensional surface, for example the 2-sphere S2. The Yang-Mills equations are rather trivial in 2-dimensions since they assert that the field Fuv (or curvature) is covariant constant. From this one can easily describe all solutions and one finds that they are precisely the homogeneous connections, i.e. connections which admit the action of SU(2) (the double cover of the rotation group SO(3)). By a general principle homogeneous connections with group G are given by homomorphisms of the isotropy group, in this case the circle subgroup of SU(2), into G. Thus we see that the critical points of the Yang-Mills action correspond to the critical.points of §2. If we look at the function space situation, the space of all connections is contractible but since our action is invariant under gauge transformations it is appropriate to factor out by these (actually for technical reasons it is best to use only gauge transformations which are fixed to be the identity at some base point of S2). The resulting space turns out to have the same homotopy type as nG. Thus we see that our example is, in its essential features, like that of §2. This should not be too surprising to physiCists since, if we put the Yang-Mills theory in Hamiltonian form (treating I-dimensional space as compact) we get the motion of a free particle on G , which is precisely the Hamiltonian version of §2. Mathematically there is interest in replacing S2 by a general surface of genus g and this case has been extensively studied by Bott and myself. For a preliminary account see [lJ. It is perhaps worth pointing out at this stage that solutions
Description: