Ten Miles Square


February 06, 2012 9:09 AM Romney’s Next State Challenges: Why the Midwest is a myth

By Colin Woodard

If I were stranded on a desert island and allowed access to only a single source of U.S. political analysis, I’d request Nate Silver’s Five-Thirty-Eight blog at NYTimes.com. Timely and data-driven, it presents the stuff upon which opinions can be built, rather than simply opinion. True, Silver’s predictive engine completely fumbled the 2010 gubernatorial contest in my home state of Maine, but most of the time it’s remarkably accurate, and the commentary intriguing.

As the author of a book on American regionalism, yesterday’s post - “One test left for Romney - The Midwest”- was of particular interest. Looking at the results thus far - and polling data going forward - Silver argues that for Mitt Romney to lose the nomination, “it is likely to be because of the one region that has yet to give him a victory: the Midwest.”

A contiguous block of eight swing states containing 95 electoral votes — Indiana, Iowa, Michigan, Missouri, Minnesota, Ohio, Pennsylvania and Wisconsin — determine the winners and losers in most presidential elections…Only when [candidates] are about evenly divided, as in 2000 or 2004, do swing states in other parts of the country — like Nevada or New Hampshire or Florida — tend to make much difference.

Silver goes on to note that Romney has yet to win a “Midwestern” state, and that his challenger, Rick Santorum, beat him in Iowa and is polling well ahead of today’s caucuses in Minnesota, as well as Missouri and Ohio. “Mr. Santorum is, in many ways, a more dangerous opponent for Mr. Romney than Mr. Gingrich at this point,” he writes, with less baggage, an appeal to working class voters, and “a more compassionate side of conservatism when it comes to fiscal policy.” Unlike Newt Gingrich - strong only in the Deep South - Silver thinks Santorum has a possible - albeit remote - path to the nomination.

I’d offer one correction to this analysis: there’s no such thing as the Midwest.

As I argue in American Nations, the middle part of the country is divided between three regional cultures, each rooted in a distinct colonial culture founded on or near the eastern seaboard. Yankeedom, the Greater New England cultural space, encompasses the Western Reserve of Ohio, the northern fifth of Illinois, all of the Upper Great Lakes states, and portions of eastern Iowa and the Dakotas. The Quaker-founded, German-influenced Midlands occupy the next tier southward: a slice of Pennsylvania, central Ohio, Indiana, and Illinois, parts of northern Missouri, most of Iowa and the eastern halves of Kansas and Nebraska. Scots-Irish Greater Appalachia got its start in south-central Pennsylvania - very close to Mr. Santorum’s hometown - and dominates southern Ohio, Indiana, Illinois, and Missouri, half of Arkansas and Oklahoma, and northeastern and central Texas. These cultures -
fleshed out in more detail here- have never had much in common politically, socially, or ethnographically, a fact of considerable importance in interpreting “Midwestern” politics.

As I’ve previously noted in this space, Romney is a Yankee-born, Yankee-raised former Yankee governor who belongs to a faith with Yankee roots. He has the upper hand over his rivals across Yankeedom: he won New Hampshire and will almost certainly win Michigan, the other New England states, and if he loses in Minnesota tonight, it will be as much because of Ron Paul as Gingrich or Santorum. He should also be strong in Illinois (because Chicagoland and the north have such a large share of the electorate), but vulnerable in Ohio and Pennsylvania (where Yankees do not constitute a plurality) and, especially, Missouri (where they don’t exist at all.) He’d be in trouble in Indiana too, had Santorum not failed to get on the ballot there.

As Alec MacGillis noted recently at The New Republic, Santorum’s personal history and political powerbase have both straddled the great fracture line of Keystone State politics: that between Appalachia and the Midlands. His Catholic working class background plays well with conservative Midlanders, helped him achieve victory in Iowa, and accounts for his polling strength in (I suspect, Midlander and Appalachian) Ohio and Missouri (which is also split between these two cultures.) Gingrich, by contrast, plays poorly outside the Deep South, and even failed to win the counties of central Florida this year. Paul - a Midland-born, Appalachian-based iconoclast — does not have clear appeal in regional cultural terms, a fact that will always hamper his myriad candidacies.

In short, Romney has two tests left in the primary: the Midlands and Greater Appalachia. And he’ll need to appeal to voters in both if he’s to win the general election against President Obama.

Back to Home page

Colin Woodard is State and National Affairs Writer at the Portland Press Herald / Maine Sunday Telegram and author of American Nations: A History of the Eleven Rival Regional Cultures of North America.


  • Equal Opportunity Cynic on February 07, 2012 8:33 AM:

    Woodard is such a one-trick pony, of course, but it's an interesting theory. I'd love to see cluster analysis of voting patterns (or other data, but here we're discussing politics) to strengthen or weaken his case.

  • John on February 07, 2012 8:53 AM:

    God, what a tired shtick.

  • scott_m on February 07, 2012 10:04 AM:

    Give Woodard a break--he's on the hedgehog side of fox-hedgehog continuum, but that's cool.

    If the One Big Thing he knows has predictive value, that's very cool indeed.

  • Russell Sadler on February 07, 2012 10:56 AM:

    This is a refreshing shtick, John. Regional historians have been saying this stuff for years and being outshouted by the Left, Center and Right crowd. There are NOT just two poles to American politics. The political fight in our country is between the Slave-owning South and the Yankees of New England over the control of the federal government in Washington. The winner is the side that makes the best alliances with the remaining American cultures. The mass media era tended to bury these regional differences. The recent fragmentation of the media into demographically selected slices allows these regional differences to re-emerge and the fight for their loyalty is what you are seeing in these regional primaries. The South created super-Tuesday in an effort to lock up the Republican nomination for a southern-sympathizing presidential candidate before the convention. It's not working well this season because the southern-sympathizing candidates have little appeal beyond that region -- so Romney is still alive politically despite being the ultimate Yankee.

    Pretty asute analysis.

  • Colin Woodard on February 07, 2012 1:32 PM:

    Here at Ten Miles Square I do, indeed, tend to post about the implications of the American Nations paradigm, it being a presidential election year and all.

    If you don't fancy this fare, I do have other tricks and interests -- from the golden age pirates to marine policy to ethnic conflict in the Balkans -- which your friends at Google can help point you to.

  • Sean Scallon on February 07, 2012 1:59 PM:

    Sounds like someone who reads a lot of Kevin Phillips.

    It's too bad Paul doesn't have that distinct regional identity, largely because his libertarian ideology doesn't allow for it. Certainly he could poll well among the same Midwest groups Taft always polled well with: anti-war Germans.

  • critic on February 07, 2012 8:02 PM:

    Yea, it might be a good idea for Woodard to consider writing about one of those other topics here.

  • Morton Nadler on February 08, 2012 2:46 PM:

    Oh you tiresome nit-pickers. You don't need to tell how tired you are about Colin's case studies in support of his thesis. Just unsubscribe.

  • Rory Larson on February 08, 2012 9:16 PM:

    Colin, I've recently read your 'American Nations' book, and I was quite blown away by your thesis. I'm sure I will never look at American history the same way again, and my perspective will be the richer for it.

    The one piece I choke on is your claim that the Upper Great Lakes area is a "Western Reserve" of Yankeedom. I was born in Utah, grew up in Upper Michigan, and have spent most of my adult life in eastern Nebraska with visits to Michigan, Minnesota and Wisconsin. This area all seems to be basically Midlands to me. Scandinavians may be less Catholic than Germans, and perhaps they are more enthusiastic about government, but they are not at all the Yankees you describe for New England. In any case, Germans are all over the Upper Great Lakes as well, and Scandinavians are among the main settlement groups in eastern Nebraska.

    I'd also wonder about the geographically substantial North Woods part of the Upper Great Lakes. You define the Far West primarily in economic rather than in ethno-cultural terms, as an area that isn't easily settled by Euro-American farmers, and which therefore tends to be heavily dependent on extractive corporations and support from the federal government. It seems to me that this would fit the North Woods area where I grew up as well, and I've always felt that there is a cultural line between it and the densely farmed country to the south.

    Could we consider the North Woods to be an eastern finger of the Far West, and the agricultural land south of it to be Midlands (with a Scandinavian flavor)? How would that affect the analysis?

    At any rate, it was certainly one of the most fascinating books I've ever read. Thanks so much for writing it!

    Best, Rory

  • Josh on February 11, 2012 4:46 PM:

    Yes, this is a schtick, well put. It's not a totally empty schtick, but it's not rigorous either.

    The map he relies on for his analysis should put you on your guard. Click on the link, in the article, to his book "American Nations" and you'll see the cover, which superimposes his theorized cultural zones onto a map of the continental United States. Look carefully and you'll observe that the map is a rather poorly chosen historical map with many basic errors, such as... the Great Lakes and the midwestern states are sort of in the wrong places.

    I don't know if he picked this map out or if his publisher did, but this same map has been featured in all the online articles that have put Woodard into the public discussion during he past year. For him to be writing about the midwest right now is therefore especially amusing, since he didn't seem to have applied any rigor to how he defined the boundaries of these zones in that very region--if he used that old map, anyway.

    It just makes you wonder how serious the guy actually is.

  • Rory Larson on February 13, 2012 11:08 AM:

    Thanks, Josh. I have the book. The map on the front is obviously an old map that would date to some time before the Civil War. The land isn't totally accurately surveyed, and a lot of the interior states haven't even been created yet. That map is an eye-teaser. Inside though, just before the text begins, he has an accurate modern map of the United States down to the county level, with each county shade-coated to indicate what regional culture it belongs to. I'm assuming this comes from analysis of electoral returns, but I'd like more of a discussion on this and on what the margins of error look like, especially between his "Midlands" and his Western Reserve of "Yankeedom".

    The book is fascinating as a history, and he has a powerful idea that is a huge relief from the old Left-Center-Right conception of politics. But I agree with you that it needs to be backed up better.


  • John on February 14, 2012 9:39 AM:

    These kind of regional divisions often (but not always) make sense within states, but once you start applying them on a macro level, they actually don't make any sense.

    Just look at the map (you can find the detailed one on Amazon Look Inside). Supposedly, Philadelphia is supposedly more like Des Moines or even Amarillo, Texas, than it is like New York City or even Trenton. Cleveland is more like Boston than it is like Toledo or Youngstown. Pittsburgh is more like Little Rock or Dallas than it is like the western Pennsylvania counties immediately to its north. New Haven is more like Minneapolis than it is like Bridgeport.

    This stuff is idiotic. It's obvious nonsense, and trying to make any kind of predictions from it is a fool's errand.

    Of course some elements of American politics can be understood from looking at the original settlers of the region, and some of Woodard's "nations" make intuitive sense (the Deep South, for instance, or the Left Coast - although in the former case, I'm not really sure Tampa and Orlando aren't more like nationless Miami than they are like Mississippi). But, especially in the northern part of the country, looking only at first settlers makes very little sense if you want to understand contemporary politics. The cities of the Great Lakes region are of a piece with one another, but they're not really of a piece with Boston. There were later waves of immigrants, and later changes to the economic basis of these regions, that makes them very culturally different from New England.

  • Missourian on February 16, 2012 12:00 AM:

    St. Louis, the most populous and economically important area in Missouri, is certainly a better fit for your "Yankeedom" paradigm than either of the others. Its very Catholic and traditionally a manufacturing city and now has an important financial industry. Its not an "Appalachia" by any stretch of the imagination and has never been defined by a "German" population or culture like the "Midlands."