Solution to the ghost problem in fourth order derivative theories

TL;DR

This paper offers a novel solution to the ghost problem in fourth order derivative theories, demonstrating that ghost states vanish in the pure fourth order limit due to a hidden symmetry, thus resolving a longstanding issue.

Contribution

The authors develop a Dirac constraint quantization approach and reveal a hidden symmetry that eliminates ghost states in pure fourth order theories.

Findings

Ghost states move off shell when second order actions are switched off.

Pure fourth order theories are free of negative norm states in the asymptotic spectrum.

A hidden symmetry explains the disappearance of ghosts in the pure fourth order limit.

Abstract

We present a solution to the ghost problem in fourth order derivative theories. In particular we study the Pais-Uhlenbeck fourth order oscillator model, a model which serves as a prototype for theories which are based on second plus fourth order derivative actions. Via a Dirac constraint method quantization we construct the appropriate quantum-mechanical Hamiltonian and Hilbert space for the system. We find that while the second-quantized Fock space of the general Pais-Uhlenbeck model does indeed contain the negative norm energy eigenstates which are characteristic of higher derivative theories, in the limit in which we switch off the second order action, such ghost states are found to move off shell, with the spectrum of asymptotic in and out S-matrix states of the pure fourth order theory which results being found to be completely devoid of states with either negative energy or negative norm. We confirm these results by quantizing the Pais-Uhlenbeck theory via path integration and by constructing the associated first-quantized wave mechanics, and show that the disappearance of the would-be ghosts from the energy eigenspectrum in the pure fourth order limit is required by a hidden symmetry that the pure fourth order theory is unexpectedly found to possess. The occurrence of on-shell ghosts is thus seen not to be a shortcoming of pure fourth order theories per se, but rather to be one which only arises when fourth and second order theories are coupled to each other.