I have found the following data points: the Catholic Church required banns and a record kept of marriages, and recording of baptisms after the Council of Trent. The Church of England started keeping baptism records unevenly a little earlier, and more consistently a little later. Ditto the Dutch Reformed. From this, I concluded that:
(1) Mid 1500s is when record keeping on everyone started.
(2) Probably somehow an artifact of the Reformation.
Step 2: Can we blame this on Martin Luther? It occurred to me that I knew some of the much repeated high points (consubstantiation vs. transubstantiation, church officials can marry, etc.), but I was betting there was a ton of detail that I was unaware of.
First, however, a side jaunt to an online public domain Catholic Encyclopedia:
In which there is this hilarious pair of sentences:
"In the same year  the Fourth Lateran Council made it a general ecclesiastical law (c. 3, x, De clandest, desponsat., iv, 3). The Council of Trent confirmed this law, and specified to a certain extent the manner of its execution." Let's just pretend the intervening, uh, three and one third centuries didn't matter. Funny.
Back to Luther:
Okay, seriously? That is one seriously fucked up guy. Honestly, I'd take scandalous popes over this guy in a _heartbeat_. I don't like the pedophilia at all; that's probably worse than what Luther is laying out here. In any event, Luther picks and chooses as much as anyone else when it comes to interpreting scripture, and his idea of acceptable consanguinity is way more dangerous in a number of ways.
The fact that Luther is pretty concerned about people becoming engaged and then marrying someone else, and that he repeatedly recommends that people deal with impossible situations by running of to a distant country and marrying there is an indication of the social control issues that were in play at the time.
Here's my current theory. This is a time frame in which the materials (paper, ink) and skills (literacy) needed to keep rudimentary records become available in a very wide context. And they weren't available in a wide context a hundred years earlier (feel free to argue with me on this one. I'm interested. But I want more than a bald assertion. I've already produced one of those). There are some other characteristics of this time period:
(1) No one (effectively) is tied to the land any more. If someone doesn't like what's happening to them, they can walk somewhere else and try their luck there. And a decent chunk of people are doing exactly that.
(2) There has been a substantial breakdown in the legitimacy of a lot of social control mechanisms; this is an effort to re-institute control and surveillance in a manner more in keeping with Modern Times (circa 1550).
(3) Birth, marriage, and death are _still_ considered the province of the church. Keeping records of members activities is a way of getting a sense of how big a group one is operating and may have been a factor in lobbying for recognition by secular authority. May. I'm kinda hazy on this one. I think this might have taken another century to become an important issue.
Basically, I'm blaming the labor shortages following plagues, the printing press and the Reformation, in approximately that order.
So why that subject line?
I don't know about you, but I think this is actually kind of important. I'm not just saying that because I want the records as an amateur genealogist. I'm saying that as a person who has been trying to make sense of her world by understanding how people have lived in it -- as an amateur historian. Recordkeeping defines a lot of what we can know and therefore what kinds of theories can be tested. What we can understand about the past is all there really is for predicting the future and making better decisions in the present. Not even _knowing_ when recordkeeping underwent various transformations, much less _why_, seems like a gaping hole in my understanding of the world.