Fourier phases contain a vast amount of information about structure in direct space, that most statistical tools never tap into. We address ALMAs ability to detect and recover this information, using the probability distribution function (PDF) of pha
se increments, and the related concepts of phase entropy and phase structure quantity. We show that ALMA, with its high dynamical range, is definitely needed to achieve significant detection of phase structure, and that it will do so even in the presence of a fair amount of atmospheric phase noise. We also show that ALMA should be able to recover the actual amount of phase structure in the noise-free case, if multiple configurations are used.
