Where to get it

*If you have a fast connection, PDF is the best version of it (and other preprints from lanl; set your cookie there appropriately so it is an automatic option). View with Adobe Acrobat Reader 3.01 or later. (Ghostscript/view seems to have some problems.) If you have a slow connection from outside the US or Europe, you can try lanl's mirrors, but they don't have the PDF version. Sorry, we aren't mailing printed copies. If you still have computer-related problems, please contact your System Manager before e-mailing me.

What makes it different

• It's free.
• It's fast. You can download it from the Los Alamos archive or its mirrors, just like preprints, without a trip to the library or bookstore or waiting for an order from the publisher.
• It's electronic. You can print it, but the PDF version has many advantages, like:
  • Download it at work, home, etc. (or carry it on a Zip ``floppy"), rather than carrying a book or printing multiple copies.
  • Get updates just as quickly, rather than printing yet again.
  • It has the usual Web links, so you can get the referenced papers just as easily.
  • It has a separate "outline" window containing a table of contents on which you can click to take the main window to that item.
  • You can electronically search (do a "find" on) the text.
  • Saving trees (and ink).
• It covers many recent topics at an introductory yet nontrivial level, such as:
  • supersymmetry
  • general relativity
  • supergravity
  • strings
• It introduces many topics not appearing in other textbooks, including:
  • 1/N expansion (color ordering) in QCD, including relation to random worldsheets
  • spacecone (spinor helicity), including explicit calculations of 4- and 5-point S-matrices in Yang-Mills
  • many useful gauges, such as Gervais-Neveu, Nielsen-Kallosh, unitary lightcone, and even string gauges in gravity
  • finite N=1 supersymmetric theories
• It is NOT:
  • a history book. All the other recent, comprehensive field theory texts take the "traditional" approach of covering topics in chronological (rather than logical) order, in storybook fashion. (This is strongly reminiscent of introductory classical mechanics courses that still teach Newton's laws before energy-momentum conservation.) This book takes advantage of hindsight, using what we now know to be the most efficient and general approaches. (For example, these other texts still quantize QED canonically, even though they know that method is inadequate for QCD. Some even claim path integrals are less rigorous, even though constructive quantum field theory has shown the opposite to be true.) Whenever I have questioned anyone who prefers the traditional approach, after eliminating all the spurious clichés, it all boils down to nostalgia. (I have even heard the excuse that it is useful to learn the less useful approaches simply because they ultimately failed --- certainly an excellent reason to relegate these topics to true history courses, for those who have the time and interest.)
  • an art book. It covers topics that have proven useful, not those that have appealed to certain tastes.
  • a concept book. All the recent texts that use a modern approach, although giving the appearance of being comprehensive except for conciseness, are curiously deficient in explicit S-matrix calculations, especially for QCD. This book both includes modern concepts and calculates with them.
  • a cookbook. Some books race to Feynman diagrams as quickly as possible, because they either consider them the only useful part of field theory, or they think such an approach is an introductory one. (One consequence is that the Higgs effect must take a back seat, and thus weak interactions are underemphasized or explained more phenomenologically.) This book contains both concepts and calculations, since the dualistic approach of concept book plus calculation book has always proved deficient for lack of two good books that work well together.
  • a survey. With few exceptions, theories are described in this book at a level that allows explicit calculations.

See also the APS News article.
An interesting article on open source books.

Corrections, additions, etc.

1. "excercise" -> "exercise" throughout the entire book!
2. p. 399: d G-hat / d mu^2 = 0, not partial, not plain G
3. p. 425 (1st word): "satisfies" -> "satisfy"
4. p. 469: the indices in the diagram aren't identical to those in the equation, and should not be confused with vector indices
5. p. 633: "mu.a" -> "i mu.a"
6. p. 647: new ref. 16 ---
   "C. Destri and H.J. de Vega, Nucl. Phys. B290 (1987) 363, Phys. Lett. 201B (1988) 261, J. Phys. A 22 (1987) 1329:
   regular (square) 2D lightcone lattice"
7. p. 701: "c_{F + F-bar}" -> "c_{D + D-bar}"

Jan. 2000

1. p. 2: "intrerest" -> "interest"
2. p. 143: ref. 3, "mechanics" -> "mechanics action"
3. p. 237: "everyting" -> "everything"
4. p. 531: some useful distances ---
   1. 1 AU (Sun to Earth): 1.5 x 10^8 km
   2. Solar radius: 7 x 10^5 km
   3. Earth radius: 6000 km
   4. Solar gravitational (Schwarzschild) radius (2GM/R): 3 km
   5. Earth gravitational radius: 0.9 cm

Feb.

1. p. xviii: Weinberg v. 3 appeared in 2000, 419 pp.
2. p. 43, ex. IB5.3a: poor notation --- "(ijk)" -> "(abc)", etc., where a,b,c = 1,2,3.
3. p. 257, 1st eq.: "Psi-tilde" -> "Psi-hat", no "-i" multiplying "d/d phi"
4. p. 528: "denisty" -> "density"
5. p. 553, ref. 6: Scans of my unpublished papers are available on the Web from KEK ---
   1. Supergravity superfields without a supermetric
   2. The superfield supergravity action
   3. A polynomial action for a massive, self-interacting chiral superfield coupled to supergravity
   4. A derivation of the supercurrent superfield
6. p. 625, between the two exercises: Missing a comment on the fact that both coupling constants in string theory can be associated with vacuum values:
   1. The string coupling appears as the vacuum value of the dilaton, since it counts loops.
   2. alpha' comes from the vacuum value of the (spacetime) metric, as can be seen from the worldsheet action. This is the string-gauge equivalent of the fact that the gravitational constant naturally arises as the vacuum (or asymptotic) value of the metric in ordinary gravity (see subsection IXA5).
   In string theory, the fact that the gravitational constant is a combination of alpha' and the string coupling is equivalent to the field redefinition from the string gauge to the particular Weyl gauge where the Einstein term in the action appears in the usual way.

Mar.

1. p. 376 (end of subsection VIIB2): "Note that the (one-loop) vacuum bubble, with no background fields of any kind, must always be dropped, as it is totally meaningless (although how it is subtracted may be regularization dependent): In terms of the graphs summed here, which have equal numbers n of propagators and vertices (P-V=L-1 by the usual hbar counting), it is the term n=0. Thus, in a coordinate space calculation, where there are also n integrations d^D x, this term would have no propagators, no vertices, and no integrals (contrary to some statements in the literature, where this graph is misidentified as a one-propagator graph with one integration). All that remains is the permutation factor, 1/n, but in this case that is an undefined 1/0."
   The Index entry for "Vacuum bubble" should also be modified to "VC7,VIB8,10,C5,VIIB1-2".
2. p. 655 (end of subsection XIIA2): "U_2 is nonunitary, which makes S-tilde^{plus-or-minus alpha} hermitian (rather than antihermitian) after the transformation, requiring a modification of the usual representation for S-tilde. The usual representation can also be used by introducing an i into the transformation, which gives S-tilde^{plus-or-minus alpha} a factor of plus-or-minus i in caligraphic-Q^alpha. However, this i can be removed by the same method used in subsection IIB4 to remove i's associated with the index -1, only now it is applied to both the + and - indices."
3. pp. 682-3 (all of subsection XIIB9): To be perfectly general, we should replace S^{alpha beta} -> S-hat^{alpha beta} (all components). (It and caligraphic-Q^alpha are given in subsection XIIA2.)
4. p. 683, near bottom: "The transformation is then" -> "The result of the transformation is then"

Apr.

1. p. 173 (end of first paragraph): "Lagrangr" -> "Lagrange"
2. p. 198: "The identification of C is somewhat semantic in a nonabelian gauge theory (except for unbroken U(1) subgroups), since it is defined by changes in sign of unobserved charges: The C appearing above at an intermediate stage of breaking of the SO(10) GUT originates as a global symmetry of only the Higgs sector, leaving all ``fundamental" particles with spin invariant. After breaking to SU(4)xSU(2)xSU(2), the vectors and the spinors are composites of the original ones and the Higgs responsible for the breaking, so they pick up this symmetry. (In the same way, the spinning particles of the Standard Model pick up the broken global SU(2) of its Higgs.)"
3. p. 334 (last sentence): "canceling" -> "Canceling"
4. p. 349, ref. 12: "(VB, ref. 14)" -> "(VIB, ref. 17)"
5. p. 439: ref. 17 should be replaced with ---
   "I.Jack, D.R.T.Jones and A.Pickering, hep-ph/9712542, Phys Lett 426B (1998) 73, I. Jack and D.R.T.Jones, hep-ph/9907255, Phys Lett 465B (1999) 148:"
6. p. 537: ref. 3 --- drop the parentheses around "super"

May

1. p. 573: ref. 6 --- "421" -> "421B"
2. p. 677 (2nd paragraph): "contraints" -> "constraints"

July

1. p. 312, ex. VIB5.4b: The easiest way is to use d^4 = int d^4 zeta exp(zeta^alpha d_alpha).
2. p. 389: The remark here about nonlocal divergences is incorrect, as can be seen easily from the fact that the subdivergences are canceled by tadpoles, which are local. However, such nonlocal divergences do appear in the following example.

Aug.

1. Some updates from Review of Particle Physics, D.E. Groom et al., The European Physical Journal C15 (2000) 1 (same URL, p. 60):
   1. p. 54: electron-neutrino mass < 3 x 10^-9, Higgs mass > 95.3
   2. p. 57: B_c^± (6.4), B_s⁰ (5.370) observed
   3. p. 58: D_s^*± (2.1124) observed
   ... and many last-digit changes, too numerous to repeat here (and similarly for the list of authors of the above).
2. p. 148, ex. IIIC1.2e: To "Expand the left-hand side" add "of the field equation".

Oct. 14

p. 188: P switches L,R -> R,L (not L,R).