What town maintained records of everyone who lives in town? It turns out Massachusetts is more like Germany than I had realized.
This blog confirms that what R. believed is true ... in Massachusetts. It does not appear to be true anywhere else, which means it's probably true all over New England.
One of the message boards on snopes had a discussion about this a while back:
It seems to me that some conclusions can be drawn:
DMV + Voter registration is a common combination.
Voter registration alone survives as an older, probably discriminatory practice, largely in southern states.
Other records (like the town residence list, tax lists, unemployment rolls etc.) are also used in various places.
But before you get all over places like Mississippi for using voter registration alone as a basis for juror selection, it's worth noticing how the organization of local government and presumably limited IT support interact to make it difficult to select jurors in any other way. If you surf around Circuit Clerk descriptions on Mississippi county government web sites, you see over and over again that it's one man or one woman handling a lot of the county clerk duties, like the juror stuff AND voter registration -- and they don't necessarily handle data from other sources that would enable them to expand beyond that.
I want to know what a jury wheel is. Oh, look: someone has already explained all this over on wikipedia:
This provides a fantastic explanation for why some people get a "run" and are called every year for three years in a row:
"The Jury Act provides that each United States District Court shall create a list of names of prospective jurors, culled from voter registration lists or lists of actual voters, and supplemented through other sources of names if necessary to achieve a fair cross section of the community and prevent discrimination. The list must consist at least 1,000 names, and preferably at least one-half of 1 per centum of the total number of persons on the lists used as a source of names for the district or division. These names are loaded into a jury wheel, which must be emptied and refilled at least once every four years."
It seems to be the case that the shorter the probable service, the longer the reprieve you get after having done the service (that is, if checking in and being told No Trial Today Go Home constitutes service, you won't be call back for a few years; if servicing tends to involve a week's commitment, your guaranteed reprieve may be as short as a year). This makes sense, but you have to think a little first: the easier the service, the more likely everyone is to participate, so the longer it will be until you have to go back; the more onerous the burden, the more people exempt out, the more often the ones who can do it will be asked to do it. There are also some issues associated with prosecutorial culture (or whatever they like to call it) that can reduce the likelihood of a jury trial.
You can actually find much better sourcing than a snopes message board from several years back, but I just recently posted a ton of links to library donations guidelines page; I'll limit this to one example:
Anyway. Big takeaway here is that random doesn't necessarily guarantee the kind of distribution that people think it does, and people who claim that being early in the alphabet makes you more likely to be called are probably right in places that still use mechanicals like jury wheels, and who (stupidly) load from the beginning of an alpha list until they fill the wheel. Anyone who argues that they are a reasonable counter-example, because their last name starts with Z, and they got a run of three calls is NOT a counter-example, but I'll leave that as an exercise for the reader.
Sausage. Grind up all the nasty bits, stick it in a length of pig intestine with some herbs and spices, cook it and serve it on a toasty bun with condiments. Yum.