This has been flying around a bit, but I've not seen it pop up here, yet. Forgive me if it is old news. (http://sensoryoverload.typepad.com/sensory_overload/2004/11/free_states_vs_.html)

I'd hoped to write a long and insightful commentary on it, but the real bottom line is this:

You're all going to have your own insights from the map. Is it really about slavery? Does it more accurately reflect rural vs. urban states? Is it more about religious states vs. more secular states? Does it have to do with Puritainical ideas of society vs. Cavalier ones (as noted in the book Seeds of Albion).

Maybe something else.

I really think it is more of a population density matter. I remember someone (Rover I think) posted a map comparing district by district which went to Kerry vs which went to Bush side by side with a map of the US taken at night from space. The areas there was light - Kerry. The areas there wasn't - Bush.

Areas with large population densities have a greater variety of people and a greater exchange rate of ideas. Colleges have long been the corner stone of liberal movements because they are places where higher learning supposedly starts, generally away from your home town. Homosexual marriage is a no-brainer when you rub elbows with a few hundred homosexual people every day on your way to and from work. Religion isn't as much of a driving factor when you have so many more methods of input. Social justice for all is easier to digest when you truly have a vision of the "all" every day instead of just on documentaries on TV. So when the notion of slavery being wrong began to permeate our membranes it started in the higher population centers and drifted to the countryside. Same to be said for the suffrage movement, civil rights, and hopefully soon homosexual acceptance.

On that note, it was the Republicans that ushered in the Anti-Slavery movement and the Republicans were the first party to back the Suffrage movement. Its just been lately that they began to shift from "Individuals, not Government" to "GOD SAYS DON'T THINK FOR YOURSELF AND HATE THOSE DIFFERENT FROM YOU, YAY!"

At least in the case of Atlanta, I think that correlation of "areas with light = Democrat" grossly misconstrues the data. If you look at the 2004 results (http://sos.georgia.gov/elections/election_results/2004_1102/001.htm), the counties that contain the low rent areas voted Kerry, while the rest of the counties in town were strongly Bush. It's really that simple here.

Counties like Cobb, Gwinnett, Forsyth, Coweta, and Fayette are where you'll find the sorts of people that Democrats often try to identify themselves with (successful, intelligent/educated, cultured, not-rednecks-in-small-town-american). However, those counties voted decisively Republican in Atlanta, not Democrat.

The only tough one is Fulton. It's a tall, oddly shaped county that contains downtown Atlanta in the South, but also some of the most affluent areas in the North. So, it's difficult to get accurate statistics for it. It's basically a given that North Fulton is strongly Republican and South Fulton is strongly Democrat though. If the county were split, it would follow the same trend that the rest of the area does.

Point being, I would be careful about what conclusions you're actually drawing with those sort of assertions.

Atlanta, Columbus, Macon, and Augusta are 4 of the 5 biggest cities in Georgia. All of them trended Democrat. I don't care and never commented on places "that Democrats often try to identify themselves with", just that the places with the highest population densities trend liberally ... and thats exactly what is the case in Atlanta as well as most of the country.

Again this is just what I think, but I hardly think its a coincidence that higher population density trends liberally 9 times out of 10 (and maybe more than that).

Atlanta, Columbus, Macon, and Augusta are 4 of the 5 biggest cities in Georgia. All of them trended Democrat. I don't care and never commented on places "that Democrats often try to identify themselves with", just that the places with the highest population densities trend liberally ... and thats exactly what is the case in Atlanta as well as most of the country.
I narrowed the 2004 Presidential vote results down to the Atlanta MSA, as defined by the 2000 Census. The vote was 55% Bush, representing just over half of Bush's 1,914,256 votes in all of Georgia. That's only 3 points more Democrat than the state-wide Georgia vote, including all of the rural areas.

I don't buy it.

To make Atlanta Democrat, you would have to use a very biased criteria to include only the low rent areas of town. I hate to say it that way, but it's simply how the city is currently divided.

Just ran the Augusta MSA. 56.29% Bush (with only 65,314 votes) to 43.71% Kerry.

Where are you getting your information? It is incorrect, or heavily biased to appear to support a particular assumption.

edit:

Columbus MSA: 52.47% Bush (with 42,303 votes).

Macon MSA: 53.66% Bush (with 46,510 votes).

It's worth noting that all three of these smaller MSAs combined aren't as many votes as a few single counties in the Atlanta MSA. Using them to make your point about population doesn't seem to make a lot of sense.

You do know Georgia's electoral votes went to Bush in 2004, right? Compare those cities to the rest of your state and you'll find they all were much closer races than the countryside.

You also should have learned by now that counties generally are geographically much bigger than cities. However, the population in cities is denser ala "population density"

The closest to home example I can think of would be the Allen / Webb race in Virginia. Allen crushed in 90% of the state, but the population difference of the state compared to the population just outside DC was pretty staggering and the seat went to Webb. Virginia went to Bush in 2004, its safely no longer a garunteed victory in 2008 for the Republicans.

You do know Georgia's electoral votes went to Bush in 2004, right? Compare those cities to the rest of your state and you'll find they all were much closer races than the countryside.
Try reading my post. That's exactly what I did.

55% Bush in Atlanta vs 58% Bush for the state is not "much closer".

You also should have learned by now that counties generally are geographically much bigger than cities. However, the population in cities is denser ala "population density"
"Generally", you're right. However, you would be incorrect yet again, when speaking about Atlanta.

This is why I'm only talking about an area where I've lived for years and am intimately familiar with. In Atlanta, we have many counties contained by the city area (which is larger than the land area of Massachusetts). Georgia is weird as has more, smaller counties than any other state in the Eastern half of the Country.

Even the downtown, city limits of Atlanta spans multiple counties.