Jesus's most prominent opponents, the Pharisees, spent a great deal of time and energy telling their contemporaries who was and who was not a true Israelite.
Working on the Sabbath? You're not in. Mixing with the unclean? You're out. Not tithing your dill and cumin? You fail. Jesus himself was repeatedly faced with the same accusation. He couldn't possibly be a real Israelite, let alone a true prophet. Just look at what he did.
There is a pleasing irony, in the light of this, that the recent story about how many real Christians there are in Britain should emanate from Professor Richard Dawkins.
According to new research, commissioned by the Richard Dawkins Foundation for Reason and Science  and conducted by Ipsos/ MORI, a lot of nominal Christians in Britain are, well, nominal. Despite the fact that 72 per centof people ticked "Christian" in the 2001 census, only 54 per cent did in the Dawkins survey, and only 28 per cent of those did so "because they believe in the teachings of Christianity".
In actual fact, even seasoned religion watchers have been surprised by some of the results of the survey - not by how low they are but how high. Is it really the case that 44 per cent of "census Christians" believe that Jesus was "the Son of God, the Saviour of Mankind"? Or that 71 per cent think he "came back to life"? Or that nearly a third believes he was physically resurrected? Or that two-thirds think that the Bible is "a perfect" or at least "the best" guide to morality we have today? I would not have imagined the figures were so high.
The precise details of the survey aside, saying who is and who is not "really" Christian is a parlous business. Church history is littered with the corpses of those who weren't really Christian, at least according to others who judged them so. Passing judgement on another's religion is hazardous.
The Dawkins survey places great weight on what people believe and practise, over and above how they self-identify. He rightly implies that people who don't know much, believe much or do much in the name of their religion aren't really very religious (although the exact number who are so totally disengaged is very small).
Yet how you choose to identify yourself does matter. Several years ago I conducted some in-depth interviews with groups of people all of whom would have fallen into Dawkins's "not real" category. They were vague about their beliefs, never went to church and knew precious little about Christianity. Half, however, were census Christians and half were not.
The difference was not only noticeable but visceral. The census Christians were generally sympathetic and supportive of Christianity, in particular its role in moral formation and in public life, whereas the others were hostile to the point of being venomous. What you called yourself clearly did make a difference.
We would all do well to remember this when we feel like making windows into people's souls. Questions of whether someone is truly Christian, or Muslim, or Hindu, or, for that matter, secularist or humanist are rarely straightforward and to categorise the world into those who are the real deal and those who are not is to do a disservice to the sheer messiness of human nature.
Nick Spencer is research director at Theos, the theology thinktank.