If the statistics of happiness depend on personal reporting, how can we be sure that anyone is as happy as they claim to be? What if they aren't telling the truth? No, we have to assume that they are, or at least that the testing system allows for lying. So the real question lay beneath: assuming that those canvassed by anthropologists and sociologists are reliable witnesses, then surely 'being happy' is the same as 'reporting yourself happy'? Whereupon any subsequent objective analysis - of brain activity, for instance - becomes irrelevant. To say sincerely that you are happy is to be happy. At which point, the question disappears.