The early Church in numbers: Believable or not?
Ray Sunshine
Shipmate
in Purgatory
In his book The Rise of Christianity, Rodney Stark attempts in Chapter 1 to quantify the growth of the Christian population of the Roman Empire in the earliest centuries, from the Crucifixion to the reign of Constantine. At first sight his projections seem farfetched, but on Catholic websites, in the last year or two, I have seen reputable posters assuming that Stark’s figures are probably as close to the true picture as we are ever likely to get.
I would invite shipmates, at least those who share my interest in this kind of thing, to take a look at Chapter 1 of Stark’s book, using the “Look inside" feature at Amazon .com. This short table summarises and simplifies Stark’s table on p. 7, entitled “Christian Growth Projected at 40% per Decade”:
Year … … No. of Christians … … % of population
40 … … … ... … 1,000 … … … … (negligible)
100 … … ... … … 7,500 … … … … … 0.01
200 … … ... … 218,000 ... … … … 0.36
300 … ... … 6.3 million … … … … 10.5
What do shipmates think?
I would invite shipmates, at least those who share my interest in this kind of thing, to take a look at Chapter 1 of Stark’s book, using the “Look inside" feature at Amazon .com. This short table summarises and simplifies Stark’s table on p. 7, entitled “Christian Growth Projected at 40% per Decade”:
Year … … No. of Christians … … % of population
40 … … … ... … 1,000 … … … … (negligible)
100 … … ... … … 7,500 … … … … … 0.01
200 … … ... … 218,000 ... … … … 0.36
300 … ... … 6.3 million … … … … 10.5
What do shipmates think?
Comments
To be rather less dismissive, at what point do we have a reasonably accurate figure for the number of Christians in the Roman Empire? If the figure for AD300 is sourced from elsewhere, and if it is reasonable to assume that the number of identifiable Christians immediately after the end of Christ's mission on earth was no more than a few hundred, then I can see where we get an increase of 40% per decade from - and on a broad-brush basis, it's probably easiest to assume a constant rate of growth rather than trying to identify particular points of growth over the 260 years of the table. But if the table is based on assuming growth of 40% per decade and then seeing where that takes us for numbers of Christians, then I'm not sure what light this actually casts on the growth of the Church: we're sta
Ahem. I stopped rather more abruptly than I meant to. Let's back up a bit and see where we go:
But if the table is based on assuming growth of 40% per decade and then seeing where that takes us for numbers of Christians, then I'm not sure what light this actually casts on the growth of the Church: we're starting from an unknown (but small) figure, applying an arbitrary growth rate and ending with a figure that isn't compared against a known fact. If that's the case - that none of the variables are based on otherwise known facts - then we've got an interesting exercise in arithmetic but not something that adds anything to our factual knowledge.
So I'd be happy to be told that at least some of the variables are based on fact.
(n.b. written without benefit of having seen @Ray Sunshine 's post which appears immediately before this)
I'm not sure that Stark gives a precise definition. I'll have a look and see.
I think it's perfectly possible quibble with some or all of his methodology, but don't see why what he's trying to do is completely off base.
We don't have exact figures, but we have estimates (via different methods) of total population, percentage of the Empire that might have been Christian at the point of Constantine and so on.
Stark claims (p. 6) that the numbers are based on Goodenough's (1931) estimate that ~10% of the population of the Roman Empire was Christian in 300 CE. Given an Imperial population of ~60,000,000 (a consensus figure according to Stark) that gives us ~6,000,000 Christians in 300 CE. Stark then uses this figure to estimate what kind of growth rate would be required to go from a church of ~1,000 people in 40 CE (a guess, but not an unreasonable one) to one with ~6,000,000 adherents in the span of 260 years. In other words, the 40% growth rate is a derived figure rather than an assumed one.
And I have to admit to being mildly amused by someone named "Goodenough" doing numerical analysis.
Thanks.
O no - I'm not quibbling with the methodology.
I just don't quite see how any figure can be calculated so many years later. It must surely depend, too, as @ExclamationMark says, on how *Christian* is defined.
It's interesting to reflect, though, that the churches to which St Paul wrote his letters may well have been quite small - rather like some of the churches we find around us in today's post-Christian western society!
I have carefully reread Chapter 1 and skimmed through a few more chapters after that, and my impression is that Stark never actually addresses the question that you and @Bishops Finger are asking here. In those circumstances, I can only attempt — like the Jesuit in the old joke — to answer your question with another question. When Paul says he baptised the household of Stephanas (1 Cor 1:16), or when he sends his greetings to the household of Onesiphorus (2 Tim 4:19), how many members of those two households are Christians,
(a) In the view of Stephanas and Onesiphorus respectively?
(b) Each in his or her own view?
(c) In Paul’s view?
... says the person speculating down below on the Synoptic problem ...
Hm. I will admit I have no knowledge of this question. However ... part of my undergraduate dissertation touched on a similar question, viz., when did Islamic Spain become majority-Muslim. Before the Arab conquest it was presumably 0% Muslim, and there is a certain point (can't remember when) when we can say 'Yes, by this point it was pretty definitely majority-Muslim'. However, plotting the line between those points is based on a heavy degree of speculation, and - crucially - the estimate has changed over the past few decades.
(For example, one method of estimating the relative sizes of the Christian and Muslim population was to look at the relative proportions of distinctively Christian and Muslim names on gravestones. This sounded reasonable until someone found evidence of Christian men using Muslim names for whatever reason.)
I would be very surprised if an estimate from 1931 for the Christian population of the Roman Empire a.) was watertight and b.) had not found something to challenge it over the intervening 90 years.
Which is relevant to interpreting scripture. What I'm asking is what is this relevant to? It may well matter to some question, but I'm just not seeing it.
I fail to see how the existence or non-existence of Q makes any significant difference to how I interpret the scriptures we've got, or how it makes any practical difference to the way I should respond to the scriptures. It seems to me a purely historical question.
You could say that about most of history as a discipline. That is, there are a few bits of history that have a utilitarian value in the sense of 'Those who do not learn from history are doomed to repeat it', but mostly ISTM it is knowledge for its own sake, i.e., people find it interesting and therefore worth knowing - rather like the existence of Q. If you don't find a particular historical subject interesting in itself, then in most cases I don't think there is any utilitarian benefit that will persuade you otherwise.
That's not his only source. On that page he also references the work of Michael Grant (1978) and Ramsay MacMullen (1984), his initial aim being to arrive at a probable range for the number of Christians in the Roman Empire in 300CE.
The book was written in 1996, so I assume it is likely additional work has been written on it since, and as long as we don't get hung up on the numbers I think there's probably a useful discussion to be had about that and the other points in his book.
If anything his argument has a somewhat 'anti-apologetic' flavour to it, as he attempts to show that there was nothing particularly special about the rate at which Christianity spread in the ancient world.
Who knows how Christianity might have developed (or not) if Constantine had (say) lost the civil war against Maxentius and Licinius?
I can appreciate the interest and appeal that the subject has...
I realize that a small minority can still be scapegoated(see Jews and anti-semitism), but still, I find it surprising that a group so tiny and also so NEW could have been so prominent on Nero's radar to begin with.
Mind you, I don't know what the population of the city of Rome was at that time. Was it small enough that a few thousand Christians would have been a significant percentage of the population?
Ah, fair enough. Like others I'd not been able to get the 'See inside' function.
Christians were new (Romans respected religions that were old), they were weird (didn't respect anyone else's gods), and worst of all many of them acted happy about the Great Fire (they thought it was a sign of the end times and that Jesus would soon return). They were practically a tailor-made scapegoat, something Nero needed desperately when rumors started that he himself had set the fire to clear land for his new palace.
The population of Rome at the time was somewhere between one million and one and a half million. Still, from what we know of early Christians' in-your-face style of proselytizing it's highly plausible that they'd come to the negative attention of the Roman state by this point.
We can publicly profess faith as much as we like, but do our actions back that up? But of course you knew that...
Of course, I understand and accept the point you and Bishop's Finger are making here, but the more I read the Gospels (and I don't read them nearly enough), the more it strikes me that Christ wasn't so concerned about who was 'in' and who was 'out' so much as whether their behaviour was consistent with their religious profession .
Any how, on sociological terms, however we cut it, people meeting together whether as a purely 'cultural' thing or because they publicly profess faith, IS a cultural thing. Your church is a cultural thing. It has its own flavour and culture. The church down the road of another denomination is a 'cultural thing' - so is the mosque, the Hindu temple or any other religious gathering. We are all of us part of some culture or other. We don't transcend and float above our culture on a cloud of sanctity because we publicly profess faith.
The Pharisees publicly professed faith. Look what Jesus had to say about them.
On the issue of how many Christians (or 'cultural Christians' if you will) there were around at the time of Constantine, I was always told that it wasn't so much the numbers as their ubiquity throughout the Empire that inclined Constantine to sanction Christianity as a potentially unifying force. That does imply a certain critical mass, of course.
The numbers question is relevant to the question of how and why the Church became the official religion of the Roman Empire. Frances Young, one of the two editors of the Cambridge History of Christianity for this period, is a Christian. More recent books often reflect an agnostic background, and so you get titles like 'The Jesus Wars' by John Philip Jenkins which looks at the seamy side of the Creed as it were. (I prefer Frances Lamb's short book on the development of the Creed myself.) You also get more books which try to get away from the idea that Providence was at work: Robert Knapp wrote 'The dawn of Christianity - People and gods in a time of magic and miracles' (2018), in which he argues that Christianity had many rivals in the competition to replace the pagan gods: Neo-Platonists, Stoics, Epicureans, Cynics, Traditional Roman religion, Isis, Cybele, Astrology, Judaism, Manichaeism, Mithras,
Gnosticism, Local gods, and Household gods. Knapp argues that the adoption of Christianity by Constantine was the decisive factor.
Good points.
And yeah, Scientology today has probably fewer than 50 000 members worldwide, but is overrepresented in terms of media attention, partly for exhibiting the same characteristics you mention about early Christians, eg. in-your-face promotional style.
I don't think he's making the claim that it always grows at 40%.
You began this by referencing my post on another thread suggesting that Luke was written in response to Matthew. It should not take a massive amount of thought to recognize that I would have to rethink my interpretation if, say, it could be established that Luke was written before Matthew.
"Constantinian Christianity" has been a touchy subject here in the past, which is a shame because I think it's an interesting discussion.
In a nutshell, I think one could say that Constantine marked the institutionalisation of Christianity. That has had both beneficial and highly negative effects over time. What is for certain is that its legacy is the timeline we are in now. We can't go back and live an alternate timeline in which we pick up at the end of Acts 28 as if nothing had happened in between. We have to move forward from where we are now, not from where some might have liked us to be. It is what it is.
Sure. I meant that I don't think he makes the claim that it grows like that forever (to your "anyway it certainly couldn't have remained at 40% per decade for many more decades after 300")
This may be my arts graduate grasp of maths, but isn't this the sort of situation where the earliest iterations are subject to the problem that small changes to your parameters have large effects on your end result?
i.e. Taking AD40 as Y(0) and AD 300 as Y(26) ...
He has Y(0) = 1000 and Y(26) = 1000 * 1.4^26 = 6.3m
But if you take Y(0) = 2000, then that is a numerically small difference (in the sense that the evidence for 2,000 Christians probably looks the same as the evidence for 1,000 Christians), but your Y(26) figure now becomes 12.6m. OTOH, you only have to revise the growth factor down to ~36% to get back to the original Y(26) figure (2000 * 1.36^26 = 5.9m).
IOW, I'm not sure how much this model reliably tells us about the size of the earliest generations.
It does seem to be an interesting study in the cumulative effect of small numbers ... thinking about @Dave W 's comments, my trusty spreadsheet has worked out that if you assume a much lower growth rate of 25% per decade, which translates to, er, something even less dramatic than 3.1% per year, then, starting from Christians as a 10.5% minority in AD 300, they become a majority by AD 370, and make up 78% of the Empire's population by AD 390, in time for the decrees of Theodosius.
How far the above is anything more than an exercise in maths I leave to the reader with better knowledge of the evidence ...
I wonder if we aren't getting tangled up in a relatively tangential bit of the book. We know that there were a very small number of Christians in the early part of the 1st Century. There are a range of historical approximations for the number of Christians in the 4th. Stark is just trying to show that the numbers can be explained by comparisons with other movements.
Surely the only people with a dog in this fight are those who assume the historical estimates for the 4th century are a magnitude too high or people who believe that there was something particularly anomalous about the spread of Christianity for whatever reason (be it spiritual or otherwise).
The other big change after the beginning of the fourth century is that adherence to Christianity becomes a plausible source of overt patronage.
I'm not sure why that casts doubt on the plausibility of the general assumption, or on what the wider point you think you're making. As I understand it: based on archaeological and literary evidence ancient historians estimate about 10% of the population of the Roman Empire was Christian at the start of the fourth century, just before Constantine came to power. (Estimates differ: but 10% represents the consensus highest likelihood.) The question is how to explain that: you either assume steady growth, or you have to assume some burst of mass conversions. Stark AIUI is arguing that there is no need to resort to the burst of mass conversion explanation, and that the numbers one generates that way fit what archaeological and literary evidence we have and are more compatible with plausible sociological explanations. Unless you're wanting to defend the burst of mass conversion explanation, I'm not sure why one would disagree.
Exponential growth is the type of growth we expect to see in situations like religious conversion or epidemic spread, where each person converted/infected can then serve as a vector for further conversions/infections. In other words, situations where the number affected is directly proportional to the ability to spread further. (Eventually this bumps up against a lack of further people to convert/infect, but that's a different question.)
The basis for assuming smooth growth is the lack of "any particularly good data for this population over the period in question". In the absence of more finely tuned data the simplest model that fits what information is available is preferred.
Although, to be fair, it does sound like social science.