Comment on "Observation of anticorrelation in incoherent thermal light fields"

Recently, Chen \em et al \rm.\ [Phys. Rev. A 84, 033835 (2011)] reported observation of anticorrelated photon coincidences in a Mach-Zehnder interferometer whose input light came from a mode-locked Ti:sapphire laser that had been rendered spatially incoherent by passage through a rotating ground-glass diffuser. They provided a quantum-mechanical explanation of their results, which ascribes the anticorrelation to two-photon interference. They also developed a classical-light treatment of the experiment, and showed that it was incapable of explaining the anticorrelation behavior. Here we show that semiclassical photodetection theory---i.e., classical electromagnetic fields plus photodetector shot noise---does indeed explain the anticorrelation found by Chen \em et al \rm.\ The key to our analysis is proper accounting for the disparate time scales associated with the laser's pulse duration, the speckle-correlation time, the interferometer's differential delay, and the duration of the photon-coincidence gate. Our result is consistent with the long-accepted dictum that laser light which has undergone linear-optics transformations is classical-state light, so that the quantum and semiclassical theories of photodetection yield quantitatively identical results for its measurement statistics. The interpretation provided by Chen \em et al \rm. for their observations implicitly contradicts that dictum