« Ney & Abramoff | Main | Redistricting »

Democratic Peace

I agree with Justin Logan on this, especially the relevance of Wilhelmine Germany to casting doubt on what kind of claim democratic peace theory is supposed to be making. If you count WWI Germany as non-democratic and then apply the resulting very tight criteria for democratic-ness in a principled way, the theory winds up being pretty trivial because you don't get enough democracies. Ultimately, the problem here is that it's hard to think of examples of situations where a country's change in status from non-democracy to democracy exercised a dramatic impact on the likelihood that it would wind up in a war with a democratic country. The nature of US-Spanish relations changed in a bunch of ways as a result of the democratization there in the 1970s, but war was off the table as an option well before it happened for reasons that were clearly geostrategic/coldwar in nature. The same reasons that kept us out of war with Franco provide sufficient explanation for the lack of US-French or US-Dutch wars during the same period. You don't need to reach for democracy to explain the peace.

Probably the most important thing to put on the table to run the argument I'd like to run here is that the risk of US-Russian war plummetted dramatically several years before it transitioned away from being a CPSU dictatorship in the early 1990s. The key moment in risk-of-war terms was the shift Gorbachev undertook in Soviet foreign policy in the late 1980s and the Reagan administration's decision to embrace that shift and move toward a more cordial bilateral relationship. Relatedly, current trends away from democracy in Russia do not increase the chances of war in any appreciable way, nor does there seem to be any way they could. What would increase the chances of conflict would be either a Russian decision to try to once again hold sway over Eastern Europe (very unlikely to happen) or else a dramatic upswing in American efforts to curb Russian influence in Central Asia and the caucuses. The relevant issues, in other words, are what they always were -- questions about spheres of influence. Internal state structure has an influence on these sorts of considerations in that it would be untenable for a country to both be a democracy and simultaneously try to maintain perpetual control over Poland and Hungary against the will of those nations' populations, but the issues are still separate.

Ultimately, when puzzling over the fact that democracies seem historically to have only gone to war with one another very rarely, if ever, it's important to keep one thing in mind. It's only quite recently -- the last fifty, or maybe only thirty years, really -- that the world has seen a substantial number of democracies. During that period there have been very few international wars between countries of any sort. This makes the absence of conflict between democracies no less important, but significantly less surprising. One might think the causal arrows go the other way, and a general sense of international peace and security promote democracy.

May 10, 2005 | Permalink

TrackBack

TrackBack URL for this entry:
https://www.typepad.com/services/trackback/6a00d8345160fd69e200d83443a85d53ef

Listed below are links to weblogs that reference Democratic Peace:

Comments

Peace and security are not preconditions for the moment of transition between a totalitarian past and a democratic future. Surely Matt knows this.

A revolutionary opening, however, is created when a long feared security apparatus no longer appears to have the will to impose its order on society.

Causal arrows point more likely to things like geopolitical momentum and a sense that the lone superpower will, if anything, support the transition to democracy.

Posted by: JohnFH | May 10, 2005 2:48:55 PM

Also, it takes politics out of the equation. Are we to believe that if all the world's states were democracies there would be no war? Is every contentious issue divisible such that war can be avoided in all situations? I think it unlikely. And what counts as a democracy is no small decision (as Matt's notes of Wilheme Germany above). Say the United States, China, Russia, Britain, and most of Europe decides to recognize Palestine as a state. One, is it a democracy? And, if so, do you expect them to stop warring with Israel because of it? I doubt it very much.

Posted by: Zach | May 10, 2005 3:08:34 PM

Another point to consider when debating democratic peace theory is that while in the long run a democracy may become less likely to go to war (against other democracies anyway), during the the transition to democracy states often often become more aggressive rather than less.

Ohio State professor Edward Mansfield and Columbia professor Jack Snyder wrote a piece in Foreign Affairs in 1995 where they claimed that states in democratic transition were twice as likely to go to war as states with authoritarian regimes (including wars against other democracies). They later updated the article and gave examples such as Ethiopia vs. Eritrea in 1999, India and Pakistan again in 1999, and Indonesian agression in East Timor after the fall of Suharto. I would note that all of these conflicts had an ethnic component, but it's good food for thought. And it makes sense that nationalism would increase during democratic transition, with so many different factions competing to gain a constituency, a nationalistic appeal may be a shortcut to public support. Of course, the classic example of increased aggression during democratic transition is France.

Posted by: Ricky Barnhart | May 10, 2005 3:22:31 PM

...democracies seem historically to have only gone to war with one another very rarely...

Why do democracies attack, invade and occupy non-democracies? Like Iraq, Vietnam, Grenada, Egypt.

Ah, I know. Some dictators are much more difficult to bribe than most elected politicians.

Well, some elected politicians are also difficult to bribe, like Jean-Bertrand Aristide for example or Salvador Allende. Then they are to be smeared and declared non-democratic and removed, or just removed.

Posted by: abb1 | May 10, 2005 3:25:58 PM

to curb Russian influence in Central Asia and the caucuses

Did you mean the Caucasus? Or was Vladimir Putin responsible for Howard Dean's performance in Iowa last year?

Posted by: digamma | May 10, 2005 3:45:12 PM

"especially the relevance of Wilhelmine Germany to casting doubt on what kind of claim democratic peace theory is supposed to be making..."

Why is Wilhelmine Germany meant to be a killer counter-argument? The German government was appointed and dismissed by the Kaiser, the British government of 1914 (with which the comparison seems to be made) was appointed and dismissed by the House of Commons, even if the House of Commons does not exercise much day-to-day control after apppointment (as Bagehot observed/foresaw). These are radicially different systems and the British government was much more democratic than the German one.

Posted by: otto | May 10, 2005 3:46:06 PM

Why go to war with another democracy when you can just stage/support a coup instead?

Posted by: wml | May 10, 2005 3:46:31 PM

Just give it another 50 years and we'll plot the data points. I think democratic peace theory will hold up, not just because of institutional barriers to interstate violence, but also because of the promulgation of liberal norms. I do agree with Matt, though, that shifts in polarity and other geostrategic dynamics over the last half century definitely overshadow any clear causal links between democracies and peace so far.

Posted by: Jesse | May 10, 2005 3:49:10 PM

otto-

The Reichstag, elected by universal suffrage in free elections, voted war credits. That's why.

Posted by: Gareth | May 10, 2005 3:57:51 PM

What about the War of 1812? Almost 200 year old, but still two democracies (US and Britain) at war.

Posted by: WKH | May 10, 2005 4:32:11 PM

"it's hard to think of examples of situations where a country's change in status from non-democracy to democracy exercised a dramatic impact on the likelihood that it would wind up in a war with a democratic country."

Azerbaijan V. Armenia?

Posted by: praktike | May 10, 2005 4:36:43 PM

WKH,

I'm really not sure you could call the U.K. a democracy in 1812. Heck, I'm not sure that you could call the U.S. a democracy in 1812.

Posted by: Noel Maurer | May 10, 2005 5:05:42 PM

Matthew Yglesias:

the theory winds up being pretty trivial because you don't get enough democracies

No, if there are too few data points, the theory doesn't end up trivial (it is that for other reasons), it ends up undecidable.

The theory that democracies don't go to war with other democracies is trivial because it is dishonest and tendentious. It is formulated with the specific case of Iraq in mind. Not to guide our future behavior, but to justify the Iraq war.

This democratic non-aggression treaty sounds an awful lot like the 1980's conservative theory that totalitarian regimes wouldn't fall without violence; which of course they shortly did in spectacular fashion just as predicted by the lamentedly late George Kennan. A theory first publicly articulated by ex-UN ambassador Jeanne Kirkpatrick if I recall.

But wasn't Hitler (and I claim exception to Godwin's Law, this being on topic) elected democratically? And to claim that Germany stopped being democratic before starting WWII against demcoratic neighbors, begs the question. If the argument is we should make war to make other countries democracies because democracies don't make war, but we concede that democracies flip to dictatorships overnight, then make war, then democracies aren't necessarily more safe for us than dictatorships because they could just become dictatorships again, then make war on us. You must prove that once a democracy, always a democracy, to make the theory worth even considering. Fat chance.

Why not just argue that part of the definition of a true democracy is that it is peaceful unless provoked, and that cases to the contrary in our history were examples of undemocratic action, then the argument is the full-blown tautology it is striving to be anyway.

More tendentiously, the US was arguably, like Germany in 1938, NOT a democracy. Bush lost the popular vote, his partisans threatened the vote-counters with a mob, and he took over by transgressing his own avowed principle of strict constitutional construction.

Posted by: epistemology | May 10, 2005 5:15:03 PM

"No, if there are too few data points, the theory doesn't end up trivial (it is that for other reasons), it ends up undecidable.

The theory that democracies don't go to war with other democracies is trivial because it is dishonest and tendentious. It is formulated with the specific case of Iraq in mind. Not to guide our future behavior, but to justify the Iraq war."

Epistemology,

Right on the first point, wrong on the second. Modern scholarship on the "democratic peace" dates back to the 70s, got a major boost with the work of Michael Doyle in the mid-1980s, and was a major topic among IR scholars from the late 1980s through the 1990s.

It was Clinton, not Bush, who first drew on this scholarship to justify a policy of promoting democratization.

Posted by: KS | May 10, 2005 5:30:59 PM

The Democratic Peace argument was not formulated with the Iraq war in mind. It was, in fact, an issue debated in political science before a second war in Iraq was even an issue. it's really a theory that attempts to explain a phenomenon, not some sort of ideological justification.

a big point that's being missed here is that while democracies do not go to war with one another they still do have conflict (i.e. UK and US in the Suez Canal). it's not simply closely allied interests that keep them from fighting. There are a number of reasons why democracies resist war even if there are conflicts of interest with other democracies.

the significance of a democratic leaders re-election is not that people in democratic countries don't like war, therefore the leader should not go to war. democracies go to war as often as autocracies, just not with other democracies. the significance is that when a democratic leader has a conflict with another country, the difference (in terms of his re-election chances) between settlement and defeat is huge, while the difference between victory and settlement is not, so if he's going to go to war he better be damn sure he's going to win. so, when a democratic leader goes to war he tends to put a good deal more effort into making sure he wins that war. so if you have two democratic leaders, each planning on putting full effort into the war, one of them will likely realize that he is unlikely to win. so, because winning is so much more important for a democratic leader's chances of remaining in power, it is unlikely that two democracies will both have the belief that they CAN win, and thus, they are unlikely to end up fighting a war.


Posted by: b. schac | May 10, 2005 5:36:48 PM

If we explictly say that unlimited democracies are the only proper objects of study for 'democratic peace theory' then Matthew is correct. Democracies have only existed for 30-40 years. Under the shadow of thermonuclear weapons. There's no way to separate out the variables. If we extend the period backwards to include republic (implicitly defining them as limited democracies, which I would contend would apply to both the US and UK in the early 19th century) then democracies have gone to war with each other, although apparently not quite as often. However, when they do fight, they tend to fight TOTAL wars (Civil War, WWI, Napoleonic wars). (I wouldn't class Tsarist Russia of the Crimean War period as democracy - no regular elections.)

In either instance, democratic peace theory sounds like a combination of noting that civilian populations that live under democratic rule tend not to want to go wage serious war, and wishful thinking/just so stories ala Norman Angell.

What would increase the chances of conflict would be either a Russian decision to try to once again hold sway over Eastern Europe (very unlikely to happen) or else a dramatic upswing in American efforts to curb Russian influence in Central Asia and the Caucasus.

Not to mention the western part of the former Soviet Union. Which I am afraid is exactly what the WH is leaning towards trying to do.

ash
['Well, this should be an interesting war.']

Posted by: ash | May 10, 2005 6:22:10 PM

It seems to me authoritarian leaders have just as great an incentive to avoid losing wars as democratic ones do. Losing wars is a good way to get overthrown, as General Galtieri could tell you.

Posted by: Gareth | May 10, 2005 6:23:09 PM

Dude, you buried your lead! The last sentence was the essence of your argument, and a damn fine essence it was, too.

Posted by: Michael Gee | May 10, 2005 6:26:36 PM

Gareth, there is a statistical and theoretical basis for saying Democratic leaders have more to lose from losing a war than do Autocrats.

If you'd like the stats, I can send you an article by Bruce Bueno de Mesquita, James Morrow, Alastair Smith and some other guy that really well outlines the strategic theory of the Democratic Peace. If you don't mind I can sum it up for you here.

Democratic leaders stay in power by provision of public goods (security, economic growth etc.) because they need a large winning coalition to stay in power. A winning coalition is the set of people keeping a leader in power. So, in a democracy it is generally 51% or more of the selectorate (defined as those people who could be in the winning coalition). For an autocrat, the winning coalition is understandably smaller. The autocrat doesn't stay in power thanks to the voters, but rather by some smaller set of powerful people who support the autocrat based on his provision of private goods. Basically, public goods are things that EVERYBODY benefits from, while private goods can be given to specific people. So, clean air is a public good while a bunch of oil fields is a private good. Losing a war directly hurts the democratic leader's chances of staying office because he loses public goods from it. An autocrat doesn't necessarily lose private goods from losing a war, in fact, the settlement can end up giving him more private goods, even if the people may suffer. For example, after the first Iraq war, Saddam remained in power even though he lost, and part of the settlement (Oil for Food) actually ended up tightening his grip on power. I probably did a crap job of explaining this, but, in short, losing a war is always a negative for a democrat but not necessarily for an autocrat because of the different sizes and natures of their winning coalition. (also, sorry for the stupid polisci speak, but i learned all this junk in my polisci class so it's the only way i know how to explain it).

Posted by: b. schac | May 10, 2005 6:37:28 PM

The Democratic Peace Theory dates (in its theoretical underpinnings, though there wasn't much data at the time) to Perpetual Peace, which Immanuel Kant wrote in - I think - 1795. In Section II of that work, he writes that the first definitive article for perpetual peace is that "The Civil Constitution of Every State Should Be Republican," that is, we can have peace if - along with a norm of universal hospitality and an international federation of free states - every country has elected representative government domestically.

If he was writing with the specific intention of justifying the 2003 conflict between the United States and Iraq, we should stop mentioning him with Mill and start mentioning him with Nostradamus.

(I express here no opinion as to the theoretical or empirical validity of Democratic Peace Theory; I'm not a quantitative social scientist. I'm also not expressing any opinion on the legal or moral validity of the Iraq war. I'm not expressing any opinion on whether the war in Iraq fit in with the rest of Kantian political philosophy. I'm not even expressing an opinion on whether anyone involved with the decision to go to war there had ever read or heard of Kant. All I'm saying is, the theory is about 170 years older than even KS suggests, and more than 200 years older than epistemology wants to make it out to be. And I'll just add to b.schac's generally strong explication of the factors that make Democratic Peace logical the fact that in democracies, the people who choose the government - that is, everyday working folks - also bear the most costs of war, so you're going to see a lot more reluctance to take on casualties than in a polity whose leader isn't democratically accountable. Oh, and epistemology, the Hitler argument is totally batshit insane. The only way the reduxio you draw works is if we don't have factors other than aggression to draw on to determine whether Germany had stopped being democratic before invading Poland. Here are a few: The roundup of members of the Communist party, February 1933. The arrest of anyone criticizing the Nazi party, March 1933. The law of March 23, 1933 legally making Hitler dictator. The abolition of the presidency, August 1934. The Nuremberg Laws of 1935, restricting citizenship based on ethnicity. Shortly after, elections were abolished. By 1939, of course, there were cascades of other violations, including Kristallnacht, but restricting citizenship to people of a certain ethnicity, abolishing elections, and naming a dictator really ought to be sufficient to prove a country is undemocratic).

Posted by: You Kant Re-Write History | May 10, 2005 6:43:15 PM

Good points on the latter post too, b.schac - sorry, was typing while you were.

Posted by: You Kant Re-Write History | May 10, 2005 6:44:21 PM

Why is Wilhelmine Germany meant to be a killer counter-argument? The German government was appointed and dismissed by the Kaiser, the British government of 1914 (with which the comparison seems to be made) was appointed and dismissed by the House of Commons, even if the House of Commons does not exercise much day-to-day control after apppointment (as Bagehot observed/foresaw).

The British government of 1914 was appointed and dismissed by the King, based on whether the government could command a majority in the house of commons. Parliament had absolutely no control over who got what job in the government (and the majority of the Liberals almost certainly would not have chosen Sir Edward Grey as foreign secretary), and no real control over foreign policy. Royal prerogative was, and is, a real thing.

It should be added that a German Chancellor who could not control the Reichstag would be out on his ass pretty quickly.

I'd like to add a number of other examples -

Italy in 1915, which was more democratic, at least, than Wilhelmine Germany, engaged in a completely unprovoked war against Austria-Hungary (which was not, of course, a democracy, but see below).

Austria-Hungary, in 1914, again, was not a full democracy, but it had two representative chambers. The Austrian chamber was prorogued at the time of the war starting, so it doesn't count, but the Hungarian government, despite very limited suffrage, was a genuine constitutional enterprise. Tisza, the Hungarian prime minister, agreed to war somewhat more reluctantly than others in the government, but nevertheless agreed. Serbia, I would add, was a constitutional monarchy of sorts, and an election campaign had been in progress as of June/July 1914.

Bulgaria, in 1915, declared war on Serbia and other Allied powers. Bulgaria, too was a constitutional monarchy. Bulgaria, Greece, and Serbia, all constitutional monarchies, had declared war on the Ottoman Empire, which was in the process of becoming a constitutional monarchy, in 1912. Then in 1913, Romania, Greece, and Serbia, still constitutional monarchies, made war on Bulgaria. Constitutional Italy had, in 1911, also declared war on the nascent Turkish constitutional experiment.

Spain in 1898 had a reasonably democratic government. I'm not sure about Mexico in 1846, but 1846 was during the time between Santa Anna dictatorships.

The War of 1812 was between two countries with representative governments.

Finland was at war with all of the allies, including Britain, the US, and so forth, for most of World War II.

Now, one of these countries precisely fits as a democracy, but I think the utter opportunism of Italy in 1911 and 1915, and that of the United States in 1898 and 1846, goes a long towards relieving one of any idea that democracies behave better than anybody else. Whether or not the governments of Turkey, Austria-Hungary, Spain, and Mexico were democratic at that time, it was aggressive and essentially unprincipled behavior on the part of the more democratic states which brought about war.

Posted by: John | May 10, 2005 6:55:14 PM

Regarding democracy and peace, I think there is a good case that the causality mainly goes the other direction. That is, peace makes democracy much more likely. I don't think it is an accident that the countries that first developed democracy in the modern era, namely the US, the UK, and Switzerland, all were, thanks to geography, much less vulnerable to invasion than the other countries of the day.

A. B. Schmookler explains in his book The Parable of the Tribes: the Problem of Power in Social Evolution, that anthropologists and historians have found that war leads to authoritarian government. That's because if you are going to succeed at war, you need a strong leader who can decide to go to war, even if much of the population opposes it. The result is that, ever since the development of agriculture lead to chronic warfare, there has been a selection process that eliminated democratic regimes in favor of autocratic ones.

In the modern era there have been some forces in the opposite direction, but what really made the difference is when nuclear arms eliminated war between great powers, and then the Cold War ended.

Posted by: Les Brunswick | May 10, 2005 7:38:13 PM

Can there be democracy without capitalism?

Bush, right now, blathers on about bringing democracy to the former Soviet republics. Is it democracy he is really intent on bringing, or capitalism? Both, of course. They are the same.

And when economies are tied through capitalist relations, fighting each other is not in the interest of production or consumption. It would be better called the "capitalist peace", one that leads, is leading, to the ecological holocaust.

Posted by: RIPope | May 10, 2005 7:44:03 PM

Of course capitalism and democracy tend to correlate. At least as much as, historically, a planned economy and totalitarianism tend to correlate.

As Chernobyl taught us, the historical alternative to capitalism if anything takes us even more speedily down the road to ecocatastrophe.

So the Pope's comments don't lead anywhere, so far as I can see.

Posted by: JohnFH | May 10, 2005 7:56:07 PM

The comments to this entry are closed.