The solution of the Dirac equation is of the form ue^(i(px-Et)) with u the vector part. If the probability density is the complex conjugate of the solution multiplied by the solution, and e^i*e^-i always equals one, so won't the probability density come out to be a scalar constant?