18.3. Numbers: What the Experts Say

A quick survey of important considerations and scholarship regarding the actual rate of growth of Christianity in its first century is presented by Rodney Stark. He notes that the highest estimate of Christian numbers ever in bona fide scholarship is 15 million believers... in the year 300 A.D. Scholarly consensus, however, trends quite strongly toward half that figure, or even less. Given the best estimate of the total population of the Roman Empire as 60 million, this means that even by the most favorable scholarly estimate on record, Christians comprised only 25% of the population even after nearly three centuries of evangelism. And most scholars agree the ratio was probably closer to 10%.[9] Even so, all scholars agree a ratio higher than 25% is completely unsupportable. As Stark rightly points out--and he is a sociologist by profession--a final number like this allows us to predict the average rate of growth over the previous centuries from known historical precedents and scientific models. Stark does the math, and demonstrates that only a rate of growth of around 40% per decade fits the actual data we have as well as known precedents (it roughly matches the rate of growth of the Mormon Church, for example).[10]

Stark surveys the evidence from antiquity that corroborates this estimate, and he is probably right--for there is no evidence to contradict him, and what little evidence we have supports him. Indeed, as Stark explains, the strongest evidence we have--Roger Bagnall's hard data from Egyptian papyrological documents--essentially agrees with Stark's growth curve. Moreover, most experts agree with Stark's conclusions--we shall mention Hopkins, Fox, and Finn below. So there is no plausible case to be made against Stark's estimate. No evidence counters it. All relevant evidence supports it. One could still "insist" the numbers were higher, and that somehow no real evidence of this survives, but any argument based on a blind conjecture is itself a blind conjecture, and that won't suit Holding's argument at all. The fact is, the evidence we have agrees only with Stark--so if we reject Stark, we still have no evidence the numbers were higher.

Yet Stark's conclusion entails there could not have been more than 8,000 Christians in the Church by the end of the first century, which fits the above picture of 100 Christians in each of 70 towns--more than enough to be a visible problem, but nowhere near enough to make Holding's case. However, we must not confuse this number with the number of converts in the first century--for almost all converts made in the 40s would be dead by the year 100, and there is also the inevitable question of apostasy. In both cases Stark is assuming their members are replaced. So if, for example, 75% of those converted throughout the first century were no longer alive by 100 (a reasonable assumption), then by Stark's own estimate the Christians actually won 32,000 converts over its first hundred years. If we add the hypothesis that 1 in 4 converts eventually left the faith voluntarily (out of disillusionment, disagreement, persuasion by outsiders, or simple fear), and that should be a fair assumption even if Christianity was 100% true (roughly 1 in 4 Americans is not a serious Christian today, and Holding would probably argue that they have access to sufficient facts to know they are mistaken), then according to Stark the Christians could claim to have appealed to as many as 40,000 members of the population, which over that same period of time would have included at least 120 million adults overall. That's it. Only 1 out of 3000 people--only 1/30th of 1 percent--were ever impressed enough to join. That's a trivial scale of success. Indeed, even if we exaggerate beyond all proportion and imagine Stark's math is off by a factor of ten, and Christianity appealed to 400,000 people in the first century, we still are at far less than 1% of the population, which we can never claim to be more than a tiny fringe minority.

A more thorough survey of the evidence and scholarship pertaining to Christian numbers was provided in a landmark paper by Keith Hopkins.[11] Hopkins rightly explains that no one can claim anything definite on this subject, at least for the first two centuries. Anyone who says anything about Christian numbers is speculating, and not asserting a fact. This is a fatal problem for Holding, whose argument requires factual premises, not speculative ones. The best we can hope for is to arrive at conclusions that do not contradict any relevant evidence, while conforming to that evidence better than any alternative in the light of known historical precedents and scientific models--exactly the standards employed by Stark. And, in fact, Hopkins demonstrates the accuracy and plausibility of Stark's conclusions. Thomas Finn also agrees with Hopkins and Stark, and adds further corroborating evidence, while Robin Lane Fox surveys every kind of evidence of Christian numbers one could expect to find (especially archaeological), and finds that Christians were practically invisible until the 3rd century.[12] We can apparently trust the eyewitness testimony of the Christian scholar Origen that by the dawn of the 3rd century "only a very few" had joined the Christian movement.[13]

In addition to all this, especially the direct numerical corroboration of Stark's model from Bagnall's papyrological survey, we have one other statistic that is probably exact and accurate: Bishop Cornelius of Rome tells us the exact size of the Church at Rome in a letter he wrote around 251 A.D., which Eusebius quotes at length.[14] In passing, Cornelius gives a list of the personnel which is so exact it surely derives from financial record books, and altogether the total comes to 60 priests of various grades, an additional staff of 94, "over" 1500 beggars and widows on the Church dole, and other members "too many to count." The fact that only dependents and staff were counted means, even at this advanced stage in the Church's development, no effort was being made to count the size of its membership--so all earlier counts surely can't be any more than optimistic boasts or guesses. But from this hard data different scholars have variously estimated the Roman Church at between 14,000 and 30,000, or even 50,000 members, in the year 251. No one argues for anything more than that, and even those numbers are probably too high. With only, at most, a hundred men qualified to lead or supervise services each week, and given that the largest meeting spaces available to Christians at the time could accommodate no more than 100 people, the Church at Rome probably could not have claimed more than 11,000 believers--which is pretty close to Stark's prediction of 14,000. There is absolutely no evidence it was larger than that. Of course, there may have been heretical churches in Rome at the time. Though Holding does not regard alternative sects as "true" Christians, we have no evidence how many "false" Christians were in Rome anyway.

Even so, I'll be freakishly optimistic and run with the largest estimate on the scholarly record (that of Edward Gibbon, over 200 years obsolete and pretty much universally rejected by modern experts). Let's just "assume" this same data suggests a Christian population in Rome of 50,000 in 251 A.D. All scholars agree the population of Rome at this time exceeded 700,000. Christians, therefore, could claim barely 7% of the population of Rome even by the mid-3rd century--even by the most flawed and exaggerated estimate on record--which mathematically entails the Church was far smaller in the 2nd century, and even smaller in the 1st. In order for there to be only a 7% penetration of the population of Rome after more than 200 years, this mathematically requires an average rate of growth no greater than 50% faster than Stark's--any faster would require implausible phases of zero or even negative growth over several decades in order to fit the facts. Yet increasing Stark's rate of growth by 50% still leaves us with no more than 17,000 Christians throughout the entire Roman Empire by 100 A.D., which entails a total tally of "converts" in the first century of roughly 63,000 (using the same assumptions stated earlier). In other words, barely 1 in 2000 people knew about and found Christianity attractive even assuming the most inflated interpretation possible of the best data we have.

Stark begins his progression from an initial base of 1,000. But what if there really were 5,000 in 40 A.D., as Luke claims? The number is dubious. But Stark's model would still predict no more than 38,000 members by 100 A.D., which means fewer than 200,000 conversions throughout the whole of the 1st century--little more than 1/10th of 1%. Fewer than 1 in 600 conversions after several generations of preaching is still fringe, not a popular success. And this isn't plausible anyway, since to match the hard data we have for the 3rd century (from Bagnall and Cornelius), the rate of growth of the Church would have to be lower than Stark's estimate. So even starting with 5,000 members in 40 A.D., the total number of conversions in the 1st century was probably somewhere near 100,000 rather than 200,000. Thus the most credible estimates we have, from what little evidence we have, still get us nowhere beyond a tiny fringe minority, and nowhere near a miraculous success.

Of course, one could dink the rate of growth around in some voodoo seesaw, with huge losses and zero growth over numerous decades, just to get higher numbers in the first century. But there is no evidence the rate fluctuated so wildly, or at all. Holding cannot say "Christianity was miraculously successful in the first century because I said so." It seems the only way to turn is either to accept the Stark model, or a model with even slower net growth than his--or abandon any assertions at all about how many Christians there were in the first century. No one can claim to know, and since Holding's argument requires knowing, his argument fails for lack of data. Any conclusion that actually has evidential support, even if we start with 5,000 Christians in the year 40, must still fit projections for the 3rd and 4th century, and when we do that--when we use the evidence we have--we never even approach 1% of the population by 100 A.D. In fact, we can barely pass 1/10th of 1%. The evidence simply does not exist to push the numbers higher.

The fact that larger numbers have no support does not entail the numbers weren't larger, only that we cannot claim to know they were. And this still means Holding can't claim to "know" the scale of Christianity's success was miraculous. Even in the realm of pure speculation, we find little help for his argument. Earlier we could estimate 400,000 total converts in the 1st century only by multiplying Starks' prediction by ten--for no reason whatsoever. This would allow for at most 100,000 members in 100 A.D., but again we're just making this up. We're not arguing from any evidence. But even if by chance we're right, that's still only 1 in 300 people converting over the course of sixty years of active recruitment. That means the largest estimate for the whole Empire by the end of the 1st century could never be greater than half of 1%. And again, such a size by 100 A.D. would entail a subsequent rate of growth far less than Stark's, even to meet the wildly inaccurate estimate of Gibbon for the size of the Roman church in mid-3rd century, much less Bagnall's data. And that's already stretching the evidence too far. In truth, the numbers must be less than 1/10th of 1%, probably far less, because that is the only estimate that actually fits the data well. So no matter how we try to tweak our growth model, the actual evidence permits only one conclusion: we cannot prove Christianity was attractive to any more than one out of every thousand people in the first century. That's simply not miraculous, or even surprising.

Reception[edit]

"Stark has produced a provocative, insightful, challenging account of the rise of Christianity. The thesis—that Christianity was a success because it provided those who joined it with a more appealing, more assuring, happier, and perhaps longer life—may anger many readers and force all readers to stop and think. It is a marvelous exercise in the sociological imagination and a warning to those who like simple explanations--such as that Constantine was ultimately responsible for the success of Christianity when he made it the official religion of the Roman Empire" (Andrew M. Greeley, National Opinion Research Center, University of Chicago).^[3]

"For years, biblical scholars and church historians have used sociological jargon to promote ideological views. Now an established sociologist has entered the fray with devastating results. This brilliant and highly provocative book will revolutionize the way people think about both biblical studies and church history. Love it or hate it, Rodney Stark's The Rise of Christianity is a book nobody interested in the study of religion can ignore" (Irving Hexham, University of Calgary).^[3]

In the media[edit]

This book prominently featured within the storyline of Hidden Empire by Orson Scott Card, according to the book's afterword, and The Rise of Christianity even inspired the book's plot.^[4]