On Patriotism

Some commenters and others reacting to my post on nationalism raise the issue of its relationship to patriotism. Even if nationalism is an evil, perhaps patriotism can still be good. Patriotism is certainly distinguishable from nationalism, as I defined that term in my previous post: “loyalty to one’s own nation-state based on ties of language, culture, or ethnicity.” It is also differs from nationalism defined as a sense of moral obligation to members of one’s ethnic or racial group across national boundaries. In common usage, patriotism generally means loyalty to one’s government and/or its ideals regardless of ethnic or racial identity. For example, one can be a patriotic American even if you are a member of an ethnic minority, English is not your native language, you dislike mainstream American popular culture, and so on.

To the extent that patriotism simply means supporting your country when its government promotes good ideals and policies, I’m all in favor of it. Indeed, I place high value on the American political system because, despite serious flaws, it provides a great deal of freedom and happiness to large numbers of people. I also admire it because, unlike most other nations, it is not primarily based on ties of race, language, or ethnicity.

At the same time, I am opposed to patriotism in the sense of valuing a nation or government for its own sake. Unlike senior conspirator Eugene Volokh, I don’t believe that we should “love” our country in the same unconditional way that we love a spouse or family member. That kind of patriotism too readily leads people to support governments that are oppressive and unjust. More fundamentally, it loses sight of the principle that governments and nations are means, not ends in and of themselves. The Founding Fathers, I think, got it right when they wrote in the Preamble to the Constitution that they were creating a new government in order to “establish justice, insure domestic tranquility, provide for the common defense, promote the general welfare, and secure the blessings of liberty to ourselves and our posterity.” The Constitution and the United States generally are not ends in themselves, but means to the objectives laid out in the Preamble. The corollary is that the government deserves patriotic loyalty only in so far as it promotes those objectives better than the available alternatives. If I thought that freedom, happiness and other important values could be better achieved by replacing the United States with some other political entity or by breaking it up through secession, I would not support maintaining the status quo out of patriotism. To do so would be to exalt a mere means above the ends it is supposed to serve.

Some, like Glenn “Instapundit” Reynolds in his response to my earlier post, argue that we need “irrational affection” for government in order for it to work well. I am skeptical. A population that values its government for purely instrumental reasons can still give it the necessary support and defend it against external enemies. At the same time, it is less likely to tolerate abuses of government power on the grounds that we have a patriotic duty to support the state for its own sake. But even if some degree of “irrational affection” for government is necessary, it should still be regarded as a means to an end, not a value in itself.

Ultimately, I think the right attitude towards patriotism was best captured by Milton Friedman in his 1962 book Capitalism and Freedom:

In a much quoted passage in his inaugural address, President Kennedy said, “Ask not what your country can do for you—ask what you can do for your country….” Neither half of the statement expresses a relation between the citizen and his government that is worthy of the ideals of free men in a free society. The paternalistic “what your country can do for you” implies that the government is the patron, the citizen the ward, a view that is at odds with the free man’s belief in his own responsibility for his own destiny. The organismic, “what you can do for your country” implies that the government is the master or the deity, the citizen, the servant or the votary. To the free man, the country is the collection of individuals who compose it, not something over and above them… [H]e regards government as a means, an instrumentality, neither a grantor of favors and gifts, nor a master or god to be blindly worshipped and served.

The free man will ask neither what his country can do for him nor what he can do for his country. He will ask rather “What can I and my compatriots do through government” to . . . advance our several goals and purposes, and above all, to protect our freedom? And he will accompany this question with another: How can we keep the government we create from becoming a Frankenstein that will destroy the very freedom we establish it to protect?

UPDATE: Glenn Reynolds (AKA Instapundit) responds to this post here:

I BELIEVE THAT ILYA SOMIN MISTAKES MY POINT: “Some, like Glenn ‘Instapundit’ Reynolds in his response to my earlier post, argue that we need ‘irrational affection’ for government in order for it to work well. I am skeptical. A population that values its government for purely instrumental reasons can still give it the necessary support and defend it against external enemies.”

I don’t think this quite responds to my point. I was suggesting that, in an evolutionary sense, a state whose populace feels irrational loyalty is more likely to prevail against states whose populaces are purely rational. This doesn’t strike me as much of a leap. A parent who values a child for purely instrumental reasons can still give it the necessary support, but I suspect that evolution has favored those who feel irrational loyalty to their kinfolks, too.

Furthermore, a state whose populace feels irrational loyalty probably has greater threat-value when dealing with states whose populace is only rationally loyal. This is not a defense of nationalism on any sort of moral grounds, of course — merely a suggestion that efforts to get rid of it will be difficult. This is particularly true if, as seems likely to me, evolution has favored irrational group-loyalty (for basically the same reasons) over periods extending long before the development of the state, so that such traits are largely hardwired.

I thank Glenn for the clarification. There is, I think, less disagreement between us than I at first thought. I certainly agree that “irrational loyalty” can give a state an advantage in some conflicts, and that nationalism will be difficult to root out. At the same time, I’m not convinced that that advantage is necessarily decisive in a conflict. Indeed, it could be outweighed by the disadvantages created by that very same irrationality. For example, German and Japanese troops fought very hard in World War II, in part because of attachment to irrational nationalistic ideology. But that same ideology also led their leaders to grossly underestimate their enemies and ultimately caused their defeat. Also, relatively non-nationalistic states that limit the power of their governments as a result are likely to be more economically productive and therefore have more resources to commit to any conflict. This factor underlies a large part of America’s geopolitical success over the last century.

It’s certainly possible that evolution favored “irrational group loyalty.” But such loyalties need not be directed at a state or a nation. I think it is less dangerous if they are directed towards smaller groups, such as friends and family, or towards adherents of universal principles of freedom and justice. Granted, some universalistic ideologies, such as communism, are even worse than nationalism. But others are vastly better. In any event, there is nothing hardwired or inevitable about nationalistic commitments as such, as indicated by the fact that most people were not nationalistic for the vast majority of human history. I’m not sure how much of what I say in the update Glenn would disagree with. It’s possible that the difference between our views is actually very minor.

Powered by WordPress. Designed by Woo Themes