The VAN group have been publicizing claims to be predicting earthquakes since 1981. Their predictions of two basic parameters of earthquakes, magnitude and location, are very loosely constrained. Most VAN magnitude predictions are below 5.7 Ms, with a tolerance of ¿0.7 Ms. The uncertainty range spanned by a predicted magnitude is so enveloping that such forecasts cannot be viewed as plainly successful. In terms of VAN's claims to spatial resolution, these, if successful, might constrain the linear dimension of the source region of a magnitude 8.0 Ms earthquake but do not adequately constrain a magnitude as large as 7.5 Ms. Not only do VAN need to refine the tolerances in their magnitude and location predictions before careful judgements can be made, but it is also necessary to provide an unambiguous statement of the prediction hypothesis used in relation to a pre-published and communally archived dataset of complete seismic electric signal occurrences similar to Seismological Station Bulletins. ¿ American Geophysical Union 1996