Redistricting: The Art and Science of Gerrymandering

Click to Listen to the Show (24 MB MP3)

gerrymander 2

Illinois’ 4th Congressional District. [NationalAtlas.gov]

We’ve done three shows so far looking at various problems with America’s electoral process. (Here, here, and here.) The next question on the docket: how do you kill a gerrymander?

Redistricting is a necessary part of democratic upkeep: changing the boundaries of political districts to account for census-tracked population shifts every ten years. So when did redistricting go from being a necessary evil to just, well, evil?

As The Economist put it, “in a normal democracy, voters choose their representatives. In America, it is rapidly becoming the other way around.”

In most states, districts are drawn by incumbents — often the majority party — who move boundaries around populations to ensure predictable, non-competitive victories; make it almost impossible to unseat incumbents; and ensure that the legislative majority remains in the legislative majority. And now that the Supreme Court has ruled that state legislators can draw new maps whenever they want, we’re likely to see a lot more (partisan) redistricting a lot more often.

So does redistricting have to be a dirty word, or an intractable political problem? Does the political will exist to change the system? Who should draw the lines, and how should they be drawn?

Rob Richie

Executive Director, FairVote

Bruce Cain

Professor of Political Science and Director of the Institute for Government Studies, UC Berkeley

Redistricting consultant to Los Angeles County and the Justice Department

Nina Perales

Southwest Regional Counsel, Mexican American Legal Defense and Education Fund

Lead counsel for Latino plaintiffs in 2003 Texas redistricting case League of United Latin American Citizens v. Perry, Governor of Texas, decided by the Supreme Court in July

Extra Credit Reading

Mark Rose, Gerrymandering for me, but not for thee, Right Minded, July 10, 2006.

Karl, Fixing elections (the system, not the vote), Rite Wing Technopagan, November 16, 2004.

kos, Time for some Dem-friendly redistricting, Daily Kos, June 29, 2006.

How to Rig an Election, The Economist, April 25, 2002.

Juliet Eilperin, The Gerrymander That Ate America, Slate, April 17, 2006.

Lani Guinier, Tricks of virtual redistricting, The Boston Globe, March 13, 2006.

Martin Newhouse, Voting Rights and Voting Wrongs: An Interview with Lani Guinier, Mass Humanities, Spring 2006.

Faith, race, and Barack Obama, The Economist, July 6, 2006.

Fair redistricting is key to competitive races, Tennessean.com, July 10, 2006.


Comments

48 thoughts on “Redistricting: The Art and Science of Gerrymandering

  1. In Washington State we have a system (by state constitutional amendment) that a bi-partisan group redraw the boundaries every ten years. That is it, no legislative tinkering and both sides (no matter who controls the leg at any point) have to agree on the map.

    Whats even cooler is that the redistricting commision releases online the various option maps at every point in the process: http://www.redistricting.wa.gov/

    What bothers me about this option is that since the decisions are still being made by partisan politicians is that instead of one side dominating the other, we have both sides trying to get the best deal from the other. In that case, in places where we could have two competative districts, we instead get two safe seats. It seems like they’re making deals with eachother to prevent competition.

  2. A simple way to prevent gerrymandering is to pass a law requiring that every district’s borders be rectangular (which could extend meaninglessly outside of the state of course).

    This would severly limit the amount of playing around that could be done, and would also give the districts an understandable real-world feel for the constituents.

    Let’s put this competitive border-drawing to bed once and for all.

  3. avecfrites:

    Communities aren’t rectangular though, and legislative districts require there being somewhat equal numbers of people living in them.

    Drawing legislative boundaries is a very difficult job in terms of mapping and demography. If we really want to make competative districts, there will be as much research and deliberation as drawing gerrymandered ones.

  4. Wow, I’m really taking over this thread, somone please stop me. But, I think this guy would make an interesting guest: http://www.stealingdemocracy.com/facts.cfm

    From the book:

    PARTISAN REDISTRICTING

    -In 36 states, state legislators draw their own election districts and the districts for U.S. House members.

    -Following the 2000 Census, 49 competitive congressional districts were significantly redrawn. 92% of incumbents representing these areas obtained safer districts and only 8% received more competitive districts.

    -An October 2004 Gallup/CNN/USA Today poll indicated that only 40% of Americans approved of the way Congress was handling its job. Nevertheless, in the 2004 elections, 98% of incumbents running for the U.S. House of Representatives retained their seats.

    -In 2002, 53% of Texas voters cast their ballots for Republican congressional candidates. Soon thereafter, Congressman Tom DeLay orchestrated an unprecedented mid-decade redrawing of districts in which Texas Republicans picked up 5 additional seats to control 66% of the Texas congressional seats.

    -Most nations of the world avoid extreme partisan gerrymandering by using proportional representation instead of single-member districts. In the January 2005 Iraq elections, for example, the Kurdish parties received about 26% of the nationwide vote, which entitled them to about the same percentage of seats in the national assembly.

    -Almost all democracies in the world that use single-member districts to elect legislators assign responsibility for drawing districts to independent commissions, including Australia, Canada, Germany, New Zealand, and the United Kingdom.

  5. emmett:

    To see what I mean, picture Massachusetts (or any other state) in your mind. Now just draw vertical lines to cut it into several slices, one for each district. You can vary the width of each slice to include a similar size population.

    This is easy to understand, and prevents politicians from bending district borders to include or exclude certain neighborhoods. The key is to limit their creative freedom, making gerrymandering ineffective.

    And it’s not like current district borders are drawn with the goal of preserving the character of neighborhoods; they are drawn to mathematically cleave people into predictable voting blocks and disenfranchise selected voters.

  6. avecfrites:

    Ok, now I see what you mean. In states like Washington though, that would work with congressional districts, but for state legislative districts that becomes a much harder job. You would need at least a few horizontal lines to prevent razor thin north to south LDs.

    Your in chips,

    Emmett

  7. The simple solution to gerrymandering is a constitutional change to an at-large ballot. You still get one vote for Congressman – if your state has 10 congressmen then the top 10 vote-getters state-wide for Congress get elected.

    Geographical districting, and its perversion, gerrymandering, are based on the antiquated idea that people’s identities and their self-interests, are based on where they live, physically. I’m a well-educated, socially-liberal, fiscally-conservative, relatively affluent, software engineer in Massachusetts who is also an environmentalist, painter, poet, figure-photographer, cook, web-designer, and gardener. My wife and I have no kids. In what way are my interests or voting patterns more aligned with my neighbors in my leafy-green suburb than they would be with voters in Cambridge, Boston, Amherst, or Cape Cod?

  8. “Avec” — the “grid” approach doesn’t fit the 48 states which are not rectangles… nor does it accomodate uneven population distributions.

    PL Nelson– Did you get a chance to look “The State I’m In by Robert David Sullivan in the Globe back in June? Sullivan created some curious districts across the state– and his “Left Fields” is a logical conglomeration of some of the places you mention.

    But consider that at-large would be unwieldly in large states. You’d get a lot of single-issue campaginers. And there’d be areas of a state which no one would be beholden to.

    No one on the thread above has yet stated the prime reason for compactness: it keeps costs down. With non-compact districts, campaigns take place over stretched areas; citizens have to travel further to see the candidate.

    Our current layout in Mass. is a reliably case in point: salamanders are still at home (while they’d have a really tough time squeezing into districts in Fla., Ill., N.Y., N.C., and others). A quarter of the state lives in Metro Boston, but half the of delegation hails from there– Boston, Newton, Quincy, Somerville, Malden. There’s no representative from Plymouth, Bristol, Barnstable or the islands. Five years ago this week, Speaker Tom Finneran tried to address this imbalance by creating moving the 5th there from its spot in Lowell (upon the assumption that Meehan was running for governor). He got a lot of flack for that, so he invited anybody to submit their own…

    Which, in August 2001, your eager-beaver citizen-cartographer supplied. I showed it to my friend Rick Klein of the Globe, then covering the State House; he looked at it and said, “Oh my God! You’ve killed Southie!” (by cutting it off from its Ponkapoag descendents and just slapping it with the rest of Boston) I also put Capuano into the 7th with Markey. Whoops, I tried. I had used Excel to align the numbers; I just wanted to keep municipalities together for . Most of the experts I spoke to as I put this together told me, of the pros who do this, “they’ve got bigger computers and custom software applications.” Seth Gittell ended up giving it a blurb in the Phoenix.

    One of the maps I drew ended getting reproduced on the advocacy page (and accompanyingflyer– where I first noticed it) for redistricting by Common Cause. Some time ago I emailed them and thought I straightened it out (though credit is still missing from the web page). Actually, the image they copied was nothing quite remarkable– it was the existing districts, with legible numbers. So they probably couldn’t find any official map at that size.

    And on that point, if anyone can find the map of Massachusetts legislative districts, let me know. I’ve looked high and low. I last saw my rep district when I actually canvassed it for Tim Schofield’s race.

    Needless to say, check out the Common Cause page– if for not other reason than to see the original Gerrymander (whose artist, long dead, is also uncredited. But he was never credited in the original newspaper either.

    Lastly, I urge everyone to visit the common census website– truly one of the noteworthy experiments of autonomous identification. Site creator Michael Baldwin has asked Americans to identify with (on different maps) cities, regions, and sports teams, and he has mapped out where people think they belong. Red Sox Nation, we learn does not yield very much ground to Connecticut Yankees, and even reaches into the Northway stretch of NY State.

  9. I feel bad, hogging this thread, but one of the things we should think about when we talk about creating more competative districts in the interest of democracy is the intent of the Voting Rights Act. The VRA outlawed some kinds of at large districts and other gerrymandering that would dilute minority votes.

  10. “The VRA outlawed some kinds of at large districts and other gerrymandering that would dilute minority votes. ”

    Only if that was the intention. I’m proposing the complete elinination of districts altogether.

  11. pl, your idea is interesting, but it seems like it would moot the voices of minorities further. For example, I live in a liberal neighborhood in Utah (yes, they do exist) and short of putting every few houses in a different district (which they’ve come close to trying–but I digress) there is no way to change the fact that we are going to have a liberal rep/senator from this area. As much as I’d like it if the GOP didn’t try to gerrymander us out of the 15 seats we have every 10 years, I think it would be even worse if we didn’t have a voice at all because it was completely at large.

  12. I believe since we see the problem as a new unfairness characterized by districts that look like Chinese dragons, the solution actually is geometrical.

    You *absolutely* can have rectangular districts with equal populations. They need not be all the same size or all the same shape and they may encompass cities and towns or bisect them on certain streets but by limiting the corners to four and all of them exactly 90 degrees and the district lines to the four cardinal directions except at state borders, you would put an end to most of the monkey business without drastically changing the current system.

    Remember the current system enforces equal populations only within certain parameters, like 1000 or 10,000 voters as it is.

  13. This is utter frivolity now. The idea of rectangular districts was already raised and dismissed. All but 3 states in the union have irregular boundaries.

    Here’s an idea to you or Mark: pick a state and draw districts to your rules. Then present here. I only ask of this because there virtually no examples on record of any state carved up as rectangles as you suggest.

    Consider Iowa: its process is “praised by many critics of partisan gerrymandering and bipartisan imcumbent protection plans” according to the Almanac of American Politics. The five districts are compact and follow county boundaries. There are some quirks, as the Almanac points out, for Des Moines is separated from some of its suburbs, but they remain compact and simple.

  14. Iowa seems to have a good system. I would prefer a system by which districts are drawn by some sort of independent commission, possibly judges (but preferably not judges who are subject to partisan elections). There should be some basic rules. First and foremost, districts should be geographically compact and follow well-known boundaries, such as county lines, city limits, well-recognized neighborhoods, or zip codes. Perhaps computer software could be written to assist with the task. It should be strictly illegal for legislatures or other elected officials to draw districts. Since the courts don’t seem to understand the implications of the current problem, and the existing members of Congress prefer the current system, I doubt it will change unless the voters get really, really angry and seriously threaten the incumbents. Ideally, there should be a federal constitutional amendment to prevent legislatures from drawing districts – that way, all states would be forced to improve the system.

  15. Why hasn’t the power of GIS and spatial data analysis been brought into the fray? It would be interesting to consider a model that identifies the minority interests and community identities that should be protected within cohesive districts, identify the census data parameters and values that best describe these groups, identify shape/size/density criteria, and then develop spatial analysis algorithms that select the polygons that individually optimize representation of a particular interest and collectively encompass the highest diversity of interests. Possible?

  16. Sopper14: I’m sure it is, and I’m sure it is what they’re doing now (at least in Washington State), but the intent isn’t to build fair and competative districts, but rather safe ones.

    The tech is there, just not the intent.

  17. Yes, agreed– the pros have all the tools. I remember when GoogleMaps came out, Chris Nolan had expounded on Personal Democracy Forum about the possibilities of mapping tools in the hands of average folks. But I responded that the pros have been using GIS mapping tools for a couple of decades now– and it’s still not at a price point for amateurs to play around with, to do things like re-draw districts.

    Heather– one thing to note about Iowa is that the only reason the system was every accepted is that the legislature and the governor still have the power to veto it.

    There’s another viewpoint I should bring to this discussion: the absurdity of the limit of 435 members. Michael Lind blew this open in the 1992 article A Radical Plan to Change American Politics. When he wrote the article after the 1990 reapportionment, House members on averaged represented 600,000 citizens; I think that it up another 10% now. More reps mean smaller districts, cheaper elections, closer constituent service, etc. Though I suppose we’d have even more goofballs elected.

    Lind noted also that “the U.S. House of Representatives is small by Western democratic standards. Germany’s newly revised Bundestag has 662 members, the British House of Commons 651 members, France’s National Assembly 577 members, and Japan’s lower house of the Diet 512 members.”

    Incidentally, Lind calls for proportional representation in the House (and, to protect against fringe parties, he suggests a 5% threshold, as Germany has).

    Lind later suggsted, in Mother Jones that big states divide themselves to increase their Senate representation.

  18. Increasing the number of House members would also effect real change in how the electoral collge works. It would make it more representative of the popular vote.

    This is interesting. If the number of representatives had been close to constitutionally mandated minimum of 1 per 30,000 people, Gore would have won.

  19. I’m unable, because I’m not an Atlantic subscriber, to access the article Jon G. links us to (at 12:00AM, July 24th). But even without reading it, I can surmise the advantages of proportional representation.

    Proportional representation would do much more than rendering unnecessary the skullduggery called gerrymandering. It would enable a much broader and healthily competitive national political conversation.

    I heard recently on NPR that the fastest growing party affiliation for registered voters is ‘Independent’. This implies that the country needs third (and fourth and fifth!) party options; yet these parties hardly ever form. Why? Because the constitutional mandate of representation by individuals instead of by parties necessitates a polarized political environment: Conservative vs. Progressive, with the stakes too high in that battle to leave any breathing-space for parties of true centrists. Hence the Constitution unintentionally creates a two party state – a duopoly.

    We take it as a given that monopolies are unhealthy for economies, and that one-party states are just as unhealthy for their peoples. Duopolies aren’t much better. Sure, the second entity enables competition, but the duopoly will also conspire to fix the rules, making the political collectivity in effect a two-headed monopoly.

    Is this what democracy should be? Is this the best we Americans, so proud of our innovative history, can do?

    Can anyone seriously argue that this strangulating duopoly isn’t the condition of the current US body politic? I realize I’m somewhat stretching the meaning of ‘body politic’, but it’s a metaphorically useful stretch: do human bodies prosper better on a diet limited to two starchy staples, or on a diverse diet?

    Proportional representation will not only rid us of gerrymandering, it will rid of us of PAC’s and of lobbyists – the corruptive bane of our republic-as-constituted. The countries with the most lobbyists per capita are the USA and Great Britain: the western industrial giants that elect their national legislatures by individuals instead of by party-proportional allotment.

    This is no coincidence. Why? Because each representative is a free agent: a policy-making entity of one. An entity, therefore, dependent on parochial financing and therefore prey to the parochial influences of the biggest contributors.

    Is it any wonder then that the Republicans consistently troll up the biggest campaign treasuries while the Democrats are stuck with the less affluent leftovers? And is it any wonder then that the Republicans, whose true constituents are but a very small percentage of the electorate, nevertheless out-propagandize their way to one victory after another?

    Parties in proportional representation architectures create detailed platforms of policy and then compete by advertising their detailed policies to the voters. The parties craft the policies, not lobbyists.

    The competition of ideas in multi-party states innovates many more policy options than any monopoly or duopoly can. Such innovation is the ‘healthy diet’ for the body politic.

    Proportional representation is the best imaginable solution – even though it will splinter the Democratic Party. Because the GOP will splinter too – some suggest it’s on the verge of such a split even now. I doubt this, but only because the Elephants must continue to huddle together so long as the Donkeys continue to hang together in their barely articulate loose herd.

    We’re prisoners, in other words, of an 18th century artifact: a constitutional structure that belongs in a museum of political innovations once laudable but now obsolescent.

    Since the 18th century, Democracy as a form of governance has evolved everywhere else in the world but here.

    Let’s join the 21st century.

    Here’s how it works: The most widely appealing parties gain pluralities of votes, and then form governing coalitions truly representative of the People.

    Isn’t that what democracy should be? Collective instead of by the most moneyed of the special interests?

  20. Dear Jon Garfunkel, Yes, legislating rectangular districts was was brought up and dismissed…by you.

    Now stop being a bully and think like a mathematician. You dismissed the suggestion without even reading it because twice you say that state borders are not straight, but both posters who have suggested rectangular districts have both specifically talked about borders and that would be the exception.. obviously.

    Think and read closer, please.

    If the question is how to get closer to what Iowa looks like and less like Massachusetts (or apparently Texas) then it is easy to do with simple rules and the same damn computers they use now to create their gerrymandered masterpieces.

    I know this might be a nice jumping off point for lots of discussion about all the changes that would make our system better, but I would rather just solve one problem first and then address all that other stuff later.

    Right now we have politicians with computers picking the voters instead of voters picking the politicians and rectangular districts would put a nice big ole wrench in their works.

    It’s fair and non-partisan; the kind of thing the Dems would do if they get a chance.

    I’ll see if I can find more nerdy detail on how it’s done…

  21. Pfui. If you were thinking like a mathemetician and wanted to create geometric districts you would not be “limiting the corners to four and all of them exactly 90 degrees and the district lines to the four cardinal directions.” You would first suggest the form bees use, which is a a honeycomb made up of hexagons.

    But then you would realize, with many irregular natural boundaries and municipal boundaries, and then you find that there would be no need at all to put a straight line or a 90 degree angle (or even a 120 degree one) where none existed.

    Still, as I said before, you are free to produce for me a map of your design using your rules.

  22. Jon:

    Let’s not get lost in the details. The basic principle is to use simple rules to limit the degrees of freedom that politicians have in creating district boundaries. It doesn’t matter much if we use hexagons, squares, whatever. Any reasonable geometric rules would do. But of course the simpler the better, so voters can sit at their kitchen tables and say “damn right!” to the idea.

    The geometric rules do a lot to limit flexibility, in a very understandable way. And any natural feature such as a river that a boundary crosses is a minor deviation from perfection compared to the horrors of the current system.

    What is the big objection to this simple idea? Let’s not let the Perfect be the enemy of the Good Enough.

  23. We tie ourselves in knots to preserve the idea of having a “local” representative. Great notion, but kind of archaic: I look at my Representative’s district boundaries, and they’re just incoherent, I happen to live in a zip code that happens to lie in his district, that’s all. And because the district boundaries are so screwy, you end up with a skewed pool of candidates to select from: there may be several strong candidates the next district over & nobody much in my district, but I don’t get the choose one of those strong candidates just because they live half a mile away in the wrong direction.

    Wouldn’t it be simpler, and more democratic, if we went with proportional representation & did away with district boundaries altogether? Each state gets its number of House seats, each party puts together a slate with up to that number of candidates, and you can vote for the whole slate or for individual candidates, regardless of what part of the state you live in. If the state has 25 House seats, the top 25 vote-getters win – that’s it. I’m sure there are scads of problems with this approach, too, but without getting into all of them, just think about how many problems this solves – all just by letting go of the outmoded idea of the “local” district?

    Are there any researchers or advocates out there promoting this type of approach?

  24. TW:

    If we do away with local districts and make all elections state-wide, it will make it more expensive to run, especially in large states. This will reduce the chance of a non-wealthy person getting elected. It will also make all of the campaigns resemble eachother, with a handful of hot-button issues instead of a variety of issues of local interest. I think we need to preserve a way for non-wealthy people to get elected, and localized districts are a good way of helping with that.

  25. TW– plnelson had brought this up earlier, and I noted that Michael Lind had published this idea in outlets like the Atlantic Monthly and Mother Jones. And, yes, avec is right, one of the pitfalls is potentially larger campaigns.

    And, “avec”– I understand your point. I had mostly objected to your and chuck’s certitude in the idea of rectalinear districts. To give TW some credit, he or she post his or her suggestion with a good sense of uncertainty. I speak with a fair degree of confidence here as I’ve done a little bit of work in this field.

    To bring up a fresh angle, I suppose a good question to ask a researcher is actually whether the research bears out that a compact district is cheaper to campaign in, and/or cheaper to serve (e.g., # of district offices).

    And furthermore, one could research how much a representative pays attention to his or her districts by logging visits and communications, etc.

  26. A point I haven’t seen addressed in this thread has to do with the near disenfranchisement of any third-party candidate. Recall Private Willis’s “every boy and every girl that’s born into this world alive is either a little liberal, or else a little conservative!” We seem only interested in how to split access to the spoils between the two. As a libertarian (small ‘l’), I resent the exclusivity, though I see little prospect for improvement. Nor do I have a suggestion. Does anyone else care?

  27. weldcjr: This post deals at length with the ‘third party freeze-out’.

    But forget about third parties: Tavis Smiley did a segment this weekend with the authors of the new book One Party Country: The Republican Plan for Dominance in the 21st Century – http://www.powells.com/biblio/61-0471776726-0 – which implies that we’re headed for gerrymandered and financially insured single party dominance.

    Hear that sucking sound? That’s the last vestiges of our 18th century constitution’s attempt to offer democracy spiraling down the toilet.

    Here’s the relevant link to the Smiley show: http://www.tavistalks.com/TTcom/TSradio/H2_Wallsten_n_Hamburger072806.html

  28. There is a mathematical, quantitative method to measure just how severely gerrymandered a given district is.

    This was proposed by Isaac Dolom, host of the “Science Hour” segment of the Dailysonic.com podcast, a while back:

    http://www.dailysonic.com/segment1039

    Check out his “Gerry Index” page here for the mathematical details and a simple calculator that lets you punch in any existing state and district number for a quick look:

    http://www.dailysonic.com/gerrymander/

    IMO this would be a reasonable way to regulate districts. Require all districts in a given state to have “Gerry” scores under a certain hard limit, as well as within a given range of one another.

  29. A simple way to prevent gerrymandering is to pass a law requiring that every district’s borders be rectangular (which could extend meaninglessly outside of the state of course).

    — avecfrites

    Avec, this is a fantastic idea! Basing the boundaries in a mechanical/algorithmic/purely geometric method is clearly the only way out of the madness.

    I am guessing that when you were in school you solved geometric problems and chose not to throw your notebook and pencil on the ground declaring them “utter frivolity”.

    I am still awaiting patiently for a genuine critisism of your suggestion.

  30. ####So does redistricting have to be a dirty word, or an intractable political problem? Does the political will exist to change the system? Who should draw the lines, and how should they be drawn?####

    This conversation assumes that citizens have an influence over redistricting — obviously a fallacious argument. It also assumes that gerrymandering is a ‘problem’; obviously those who draw the lines don’t look at their actions as problems, but rather as solutions….. The issue for those drawing the lines is NOT fairness, democracy, reasonability, representativeness, or all the other issues raised by posters. The issue is simply one of seizing power without resorting to violence. Those who redistrict have the power to control the joystick, and are not going to give it up to anyone without a no-holds-barred, dirty, underhanded fight. No amount of ‘citizen indignation’ will make these folks give up their ace up the sleeve. I am simply being realistic. Welcome to 21st century America.

  31. oolitic– yes, it is preferable to properly call the approach that avecfrites champions as not “rectangular” but “geometric.” But, as I explained, populations are not equally distributed by area. So it’s not quite a solution. People have had eight weeks to demonstrate here how this would work. I remain the only poster so far who has taken the time to draw a redestricting map at one point to counteract gerrymandering.

    There is a mathematical measure called compactness, and Johnathan brought up a website which calculates it for you (the Massachusetts districts, btw, seemed a bit off). And yes, in various reform efforts, district reformers have called for compactness as an explicit goal. Similarly, as I pointed out, choosing a straight line comes at the expense of recognized municipal boundaries, which, here in the East, tend to be irregular.

    tbrucia: your point is nonsense. Any decision made by elected officials in a democracy must be responsible to the people who elected them. By your argument, there is no reason to discuss any policy decision of the government, like war, because well, we have no influence over it. Furthermore you argue that the minority can’t even label something a problem, because somebody else calls it a solution. “No amount of citizen indignation…” whatever. It may likely be highly difficult in Massachusetts. But a study of American history shows that reforms have taken root in the West: Wyoming (women’s suffrage, 1890), Iowa (nonpartisan redistricting, 1980), Oregon (vote by mail, 2000).

    Ok, you are “simply being realistic.” You could have argued that there isn’t much actual harm done as a result of noncompact or gerrymandered districts. But you didn’t.

  32. Here is a non-rectangular state that is entitled to nine representatives in Congress.

    http://www.math.uiowa.edu/~treadway/rectdist.png

    First, the two horizontal lines are adjusted north and south to divide the state’s population into exact thirds.

    Then, the vertical lines (six of them) are adjusted west and east to divide each third into thirds.

    In each stage, there is no wiggle room if you start with a map showing the residence of

    each resident.

    You can adjust in straightforward ways for the curvature of the earth, the way the Census reports on tracts rather than individuals, and the like; but this map shows

    the principle of the rectangular-district rule in a non-rectangular state. (There is

    probably a similar way to adjust hexagons, but it is not as easy to illustrate.)

  33. bft: ok, now try states where one factor is more than three times the next: 7,10, 11, 13, 14, 17, 19, 21, 23, 29. Your districts aren’t compact. Why do people on this thread keep trying to prop up an idea which has had virtually no support in the annals of redistricting?

  34. Here’s a question: What of Michael Lind’s proposal for “no stop loss” — no reduction in seats, which would lead to a growing House size? never in a million years would a sitting Congress propose such an amendment, but would we have our first ever state legislature-initiated constitutional convention?

  35. Why no mention of approval voting, advocated by the libertarians.

    You vote for every candidate that you approve of. You don’t vote for any candidate you don’t approve of. The one with the largest majority vote wins.

    That addresses the either-or dilemna of say Kerry-Nader, Perot-Clinton and Perot-Bush. If you want change, and you are comfortable with a president without any constituency in Congress, you vote for Nader and Perot, as well as the safe Republican or Democrat that you hate to see win, but at least you don’t vote for the evil Democrat or Republican.

  36. Since I don’t live in the US, I was a bit surprised by the Illinois’ 4th Congressional District map above. I didn’t realize it was really that bent out of shape. There must be examples of fairer redistricting process from other countries that could be well applied in the US.

    Here is how it is done in Canada.

    http://www.mapleleafweb.com/features/electoral/boundaries/boundaries-canada.html

    http://www.mapleleafweb.com/features/electoral/boundaries/readjustment-canada.html

    And here is a map of ridings for one province. It is not perfectly geometrical, but at least there are less of the imprint of long, crooked fingers.

    http://upload.wikimedia.org/wikipedia/en/6/6a/Bc2001.PNG

  37. sidewalker, thanks.

    BTW, here’s the answer toChicago:

    Cook County covers mostly 7 congressional districts; each is neatly drawn so that its representatives, well, represent the makeup of their districts: 3 black, 2 Jewish, 1 Polish-American, 1 Hispanic… in the 4th. Geographically “bent” but ethnically quite orderly.

  38. I like district candidates I guess because we have a super rep in the House: Jim Mc Govern willing to stick his neck out and go against the pack. But I don’t see why we cannot retain the districts, allow people to vote for the reps that are running in their district OR cast their vote for one in another district if it is felt that that candidate better represents their views. Because I am black or Jewish or Latino and live in a certain district does not means that I will want the rep that everyone in my district wants. I have to listen again but was Chris trying to make this last point?

  39. I should have noted that there is a certain amount of pride involved in having your rep, perhaps your neighbor (who you can meet at an uncrowded pancake breakfast) in Congress.

  40. JonGarfunkel: Try those other numbers yourself. The same method will work for a set of rows for 8 = 3 + 3 + 2, for example. The horizontal lines are adjusted to give 3/8, 3/8, and 1/4, and then the vertical lines are adjusted to give thirds, thirds, and halves of those strips respectively. Draw the map. I drew the one you asked me to draw. I am not trying to “prop up” the idea, only to show that the idea you dismissed as not feasible for non-rectangular states is in fact feasible. The method is not worthy of dismissal on that basis. It is true that an effort to give maximum compactness will not give rectangles, or partial rectangles, and there is no reason to exclude the more compact possibilities just because rectangles are easier to explain the method for. On the other hand, it is not correct to say the rectangles and partial rectangles are “not compact” to any degree. The formal measure of compactness will show a better value for these partial rectangles than for many districts that are in effect now.

  41. Jon: But, as I explained, populations are not equally distributed by area.

    Jon, this is a laughable. It is a completely irrelevant point.

    Did you take avec’s suggestion to _also_ mean one of the following:

    a) Each rectangle is the same shape

    b) No rectangle can overlap

    c) math is hard

    In retrospect, I think the real problem of a geometric/mathematical constructions is that the innumerate masses are too confused by “complicated math”.

    If your’re local, and care to meet, I can explain how this simple, very workable, plan is quite achievable.

  42. One significant issues that has not been addressed is the relative undercounting of poor and minority households during each decennial census. Statistically speaking, certain ethnic and racial minorities tend to be more transient and more incented to stay under the government radar. The Census Bureau attempted to rectify this shortfall in 2000 by applying a sampling adjustment to the enumeration, but the plan was shot down by the Supreme Court. Until these populations are truly represented by the count, they will never be fully represented by the congress.

  43. oolitic: Yes, I concede the point. I spent too long contesting the rectalinear proposal using weak arguments; for what it’s worth, I withdraw my silly point about uneven factor. Also, of course, population distribution is not a hurdle for the rectalinear districting.

    But history, sociology, geography, and practicality are.

    Let’s start with practicality.

    As I had posted, I had re-mapped around with the 8 congressional districts of Eastern Massachusetts back in 2002. No software in existence could have provided me with population per square grid. I did have the population for each municipality, which were readily available from online sources. As I explained then, I wanted to aim for compactness and the respect of municipal boundaries.

    And this is what Open Source is all about. If I didn’t have the data, there was no way I could have produced a candidate map (and without the Internet, I could have, at best, mailed it to Seth Gitell at the Phoenix). If there were a law which required rectalinear districts, there would be very few software tools to allow people to draw competitive maps, and with fewer tools, they’d be more expensive, and be in the hands of fewer people.

  44. This discussion is (mostly) based on the idea that elected officials are responsible to those who elected them, and that decisions of government are greatly influenced by the public. But mobilizations of disciplined ‘base’ are closer to the way the system really works… An energized base and control over how the rules are written trump the democratic myth (myth = legendary narrative that expresses the ideology of a culture). One can call gerrymandering a solution (if one holds power) or a problem (if one is trying to seize power) — but after everything is said and done, most frequently the Golden Rule applies: “He who has the gold, rules.” Fortunately there are exceptions to this, but those who seek to depose those who hold the power to write rules benefitting themselves should not kid themselves that those holding power will relinquish it without a dirty, tough, underhanded struggle… No ‘mathematical formulae’ will do the trick!

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>