Il Nuovo Cimento (1955-1965)

, Volume 2, Issue 1, pp 120–134 | Cite as

Propagators of quantized field

  • P. T. Matthews
  • A. Salam


Feynman’s formulation of quantum mechanics—the sum over histories—is discussed in its field theoretic form. It expresses the propagators of the interacting fields as functional integrals over the fields. The integrals over the anti-commuting Fermi fields are carried out. The complete propagators, including all radiative corrections, are thus expressed as functional integrals, over the meson field, of integrands involving the single nucleon propagator and the vacuum to vacuum transition amplitude in an external field. The latter factor is the Fredholm determinant of the external field problem for « nucleons » satisfying Fermi statistics, but its inverse if Bose statistics are assumed. These expressions for the propagators display some interesting distinctions between « renormalizable » and « unrenormalizable » interactions which are quite independent of an expansion in the coupling constant.


Si discute l’estensione alla teoria dei campi della formulazione di Feynman della meccanica quantistica — la somma rispetto alle « storie ». Si sviluppano gli integrali sui campi di Fermi anticommutativi. I propagatori completi, comprese tutte le correzioni radiative, si esprimono così come integrali funzionali sul campo mesonico di integrandi comprendenti il propagatore del nucleone isolato e l’ampiezza di transizione vuoto-vuoto in un campo esterno. Quest’ultimo fattore è il determinante di Fredholm del problema del campo esterno per «nucleone » che soddisfano la statistica di Fermi, ma è il suo inverso se le particelle obbediscono alla statistica di Bose. Queste espressioni pei propagatori mettono in risalto alcune interessanti distinzioni tra interazioni « rinormalizzabili » e « non rinormalizzabili » del tutto indipendenti dallo sviluppo in serie della costante di accoppiamento.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. (1).
    R. P. Feynman:Rev. Mod. Phys.,20, 367 (1948).MathSciNetADSCrossRefGoogle Scholar
  2. (2).
    This is an amplification of results reported previously.P. T. Matthews andA. Salam:Nuovo Cimento,12, 563 (1954).CrossRefzbMATHGoogle Scholar
  3. (3).
    R. P. Feynman:Phys. Rev.,80, 440 (1950).MathSciNetADSCrossRefzbMATHGoogle Scholar
  4. (4).
    J. Schwinger:Phys. Rev.,82, 914 (1951);91, 713 (1953);91, 728 (1953);92, 1283 (1953);93, 615 (1954);94, 1362 (1954).MathSciNetADSCrossRefzbMATHGoogle Scholar
  5. (5).
    This was pointed out byF. J. Dyson:Advanced Quantum Mechanics, Cornell University Lectures (1951). Unpublished.Google Scholar
  6. (6).
    R. P. Feynman:Phys. Rev.,75, 749, 769 (1949).MathSciNetADSCrossRefGoogle Scholar
  7. (7).
    F. J. Dyson:Phys. Rev.,75, 486 (1949).MathSciNetADSCrossRefzbMATHGoogle Scholar
  8. (8).
    K. Nishijima:Progr. Theor. Phys. (Japan),5, 405 (1950).MathSciNetADSCrossRefGoogle Scholar
  9. (9).
    G. C. Wick:Phys. Rev.,80, 268 (1950).MathSciNetADSCrossRefzbMATHGoogle Scholar
  10. (10).
    For the functional integrals we use the notation ofS. F. Edwards andR. E. Peierls:Proc. Phys. Soc., A224, 24 (1954).MathSciNetGoogle Scholar
  11. (12).
    S. F. Edwards:Phil. Mag.,47, 758 (1954).CrossRefGoogle Scholar
  12. (13).
    J. Schwinger:Phys. Rev.,92, 1283 (1953);E. R. Caianiello:Nuovo Cimento,11, 492 (1954).MathSciNetADSCrossRefGoogle Scholar
  13. (14).
    The Lagrangian should be completely anti-symmetrized (see for exampleSchwinger, reference (4)). We use (3.1) implying the full form in any context in which the two expressions are not precisely equivalent.MathSciNetADSCrossRefzbMATHGoogle Scholar
  14. (15).
    <…>0 is the notation ofJ. Schwinger:Proc. Nat. Acad. Sci.,37, 452 (1951).MathSciNetADSCrossRefGoogle Scholar
  15. (17).
    P. T. Matthews andA. Salam:Proc. Roy. Soc., A221, 128 (1953), Theorem II.MathSciNetADSGoogle Scholar
  16. (18).
    See for exampleE. T. Whittaker andG. N. Watson:Modern Analysis (Cambridge, 1902).Google Scholar
  17. (19).
    M. Neumann:Phys. Rev.,83, 1258 (1951);85, 129 (1952);A. Salam andP. T. Matthews:Phys. Rev.,90, 690 (1953);J. Schwinger:Phys. Rev.,93, 616 (1954).ADSCrossRefGoogle Scholar
  18. (20).
    See for exampleMatthews andSalam, reference (19). By (2.12) the inverse of d(g) is obtained by replacingσ n by −σ n. Making this change in (1.6) converts determinants to permanents. We are particularly indebted to Dr.Edwards and Dr.Feldman for discussions on this point. SeeS. E. Edwards:Nucleon Green Function in pseudo-scalar meson theory, I; to be published.ADSCrossRefGoogle Scholar
  19. (21).
    Great stress has been laid byR. P. Feynman:Phys. Rev. 80, 440 (1950);84, 108 (1951)MathSciNetADSCrossRefzbMATHGoogle Scholar
  20. (22).
    S. F. Edwards andR. E. Peierls:Proc. Roy. Soc., A224, 24 (1954). A derivation of the complete formula from II has recently been given byE. C. Fradkin:Dokl. Akad. Nauk SSSR,98, 47 (1954).MathSciNetADSCrossRefGoogle Scholar
  21. (23).
    S. F. Edwards:Proc. Roy. Soc., A228, 411 (1955) andNucleon Green Function in pseudo-scalar Meson Theory I and II, to be published.ADSCrossRefGoogle Scholar
  22. (24).
    F. J. Dyson:Phys. Rev. 75, 1736 (1949).MathSciNetADSCrossRefzbMATHGoogle Scholar
  23. (25).
    It should be noted that the zeros ofN(ϕ), which are quite harmless, will appear as logarithmic infinites inL(ϕ). This remark may be of significance in connection with recent work ofR. P. Feynman in theProceedings of the Fifth Rochester Conference (January 1955).Google Scholar

Copyright information

© Società Italiana di Fisica 1955

Authors and Affiliations

  • P. T. Matthews
    • 1
  • A. Salam
    • 2
  1. 1.Department of Mathematical PhysicsThe UniversityBirminghamEngland
  2. 2.St. John’s CollegeCambridgeEngland

Personalised recommendations