Bart Ehrman argues that since there is much diversity in early religious texts, no version of Christianity is the right one (so he talks about “lost Christianities”). Michael Kruger and Andreas Kostenberger counter Ehrman’s logic quite well in The Heresy of Orthodoxy. Here’s one section of this book that tackles Ehrman’s (over)emphasis on diversity.
“In the midst of discussions about canonical diversity within early Christianity, rarely is consideration given to what we should expect early Christianity to be like. Modern scholars eager to trumpet the vast diversity within early Christianity often present their findings as if they are scandalous, unexpected, and sure to shake the foundations of faith.”
“However, the mere existence of diversity would only produce such a reaction if one had reasons to expect there to be very little diversity within early Christianity. Indeed, it seems that Ehrman has presented the existence of diversity as if it were contrary to what we would expect if an original, apostolic version of Christianity really existed.”
“But is this a reasonable assumption to make? Ehrman simply slips this assumption into the debate, expecting everyone would agree that high levels of diversity must mean that no version of Christianity is the apostolic and original one. Thus his argument succeeds only if he sets the bar artificially high for the traditional view – it is only if there are very few (if any) dissenters, and virtually immediate and universal agreement an all twenty-seven canonical books, that we can believe we have found the original and true version of Christianity.”
“But such an artificial standard decides the debate from the outset, before any evidence is even considered. After all, no historical religion could ever meet such an unhistorical standard. Ehrman never bothers to tell us what amount of diversity is ‘too much’ or what amount is ‘reasonable.’ One gets the impression that he has challenged Christianity to vault over a bar where he gets to control (and can quickly change) the height” (p. 158-159).