Has Superluminal Light Propagation Been Observed? *

It says in the report by Wang et al. that a negative group velocity u = −c/310 is obtained and that a pulse advancement shift 62-ns is measured. The authors claim that the negative group velocity is associated with superluminal light propagation and that the pulse advancement is not at odds with causality or special relativity. However, it is shown here that their conclusions above are not true. Furthermore, I give some suggestion concerning a re-definition of group-velocity and a new explanation in special relativity of causality. The velocity of u = −(c/310)k̂0 is subluminal but not superluminal (the term “superluminal” is usually understood as such a light propagation with phase, group, and energy velocities all exceeding the vacuum speed of light). It is well-known that the 4-dimensional interval for a signal in special relativity is given by ds = cdt − ( dx + dy + dz ) = dt ( c − u ) , (1)