Ethnic variation in sentencing

Depressing article in the Guardian yesterday. Note the academic's caution in assigning a causal link with prejudice within the courts system, though. For example, Coretta Phillips from the LSE:

"[But] on the basis of the data published yesterday, it would be hard to conclude anything about prejudice and discrimination from the magistracy or judiciary or that there is greater criminality among minority groups."


I think this shows admirable restraint! She is of course correct that the statistics on their own cannot prove discrimination or prejudice. They are only the raw numbers, and don't therefore take into account things like offending histories, potential mitigating factors and so on. The figures certainly do not prove discrimination. But the data from certain areas, in particular, certainly indicate it. For some prejudice not to be present, at least in some of the local areas mentioned, we would have to believe that the black offenders had very different records and applicable mitigating factors to the white offenders. I wonder how plausible this idea really is......?