The Institute for Objectivist Studies

Home » Commentary » TAS Graduate Seminar (third in a series): thought-experiments and open borders

TAS Graduate Seminar (third in a series): thought-experiments and open borders

They’re having a debate about open borders over at Bleeding Heart Libertarians. I don’t find the debate particularly illuminating.  It’s not clear whether the topic under discussion is supposed to be the nature of state sovereignty and borders, or the future of American immigration policy. In the first case, a reader would expect the articulation of general principles that had global scope in the present (i.e., well beyond American borders), handled historical cases, and was general enough to handle a range of future possibilities, including “fanciful” future possibilities that were at least nomologically possible.

Mostly, though, the discussion seems focused on American immigration policy. The examples are American, the “borders” referred to are American, the anxieties are American, and the “consequentialist” considerations adduced on either side seem parochially American. As far as American policy is concerned, I’m in favor of a basically libertarian position of the sort defended by Gary Jason in this piece in Liberty last December. But all the world is not 21st century America. Would open borders work for the borders of Gaza, whether on the Egyptian or the Israeli side? Have they worked for the Afghan-Pakistan border? Do they always work anywhere and everywhere in the same way? Perhaps some thought-experiments on open borders are in order–thought-experiments intended to take us beyond American shores, and beyond the early 21st century.

I gave a well-received talk on thought-experiments at the TAS Seminar a few weeks ago. I won’t rehearse my basic line on thought-experiments here, except to refer you to the first 21 pages of Kathleen Wilkes’s Real People: Personal Identity without Thought Experiments, with which I basically agree. Though Wilkes has been described as a thought-experiment “skeptic,” I think she’s more of a thought-experiment minimizer. She doesn’t want to get rid of thought-experiments (neither do I), but she wants to put them in their place, and offers some useful desiderata on good ones.

As I mentioned at the TAS seminar, Rand herself relied on thought-experiments, as have other Objectivists. In a sense, all of Rand’s fiction is thought-experimental because arguably fiction itself has a thought-experimental aspect to it; going further, Anthem and Atlas Shrugged are (or involve) specifically sci-fi thought-experiments. Besides that, Rand’s got her immortal robot (“The Objectivist Ethics”), and her eye lottery thought experiments (“Collectivized Ethics”) in The Virtue of Selfishness. Leonard Peikoff has his “atom-sized perceivers” in Objectivism: The Philosophy of Ayn Rand, and Allan Gotthelf has a similar thought-experiment in On Ayn Rand, also on perception.  I’ve always liked Eyal Mozes’s essay on Poul Anderson’s story “The Sharing of the Flesh,” intended to illustrate a point about Rand’s non-conflicts-of-interest principle. In a more general way, the very idea of capitalism as an “unknown ideal” requires thought-experimentation. Rand thought that genuine capitalism had never been instantiated anywhere in human history. To grasp the nature of Randian capitalism, then, we’re forced to employ historically-informed thought-experiments: the concept has no fully actualized historical or present-day referent.

I’d come to the TAS seminar with a (deliberately over-ambitious) list of a dozen or so thought-experiments to discuss. We had some great discussions within my session of Rand’s “immortal robot” thought-experiment (which I regard as two or three distinct thought-experiments), J. Charles King’s thought-experimental counter-example to it (in The Philosophic Thought of Ayn Rand), and Nozick’s experience machine (in Anarchy, State, and Utopia). After the “official” session ended, some of us stuck around to discuss Bernard Williams’s famous Jim/Pedro thought-experiment (in Utilitarianism: For and Against) and J. Charles King’s “life-long golfer” counter-example to Ayn Rand’s conception of productiveness (Philosophic Thought of Ayn Rand). I’ll try to blog some of this later.

Displaced schoolchildren, Schaunestein DP camp, 1946.

Displaced schoolchildren, Schaunestein DP camp, 1946.

Before the talk, I came up with a thought-experiment of my own on open borders. Apart from a brief conversation with one seminar participant, I didn’t have time to discuss it, so I thought I’d throw it out here. I call it Khawaja’s trade-off: 

Imagine a society that follows Objectivist principles. Call it O. Everyone is an ethical egoist in Rand’s sense, and the political system enshrines individual rights in Rand’s sense. O has an open border immigration policy.

The housing market in O is in near-perfect equilibrium. In fact, things are better than that description would imply. The poor have more than adequate housing; the middle class have excellent housing; the rich have fabulous housing. Many segments of the population have more than one domicile. The population of O is 24 million. Stipulate that the climate is such as to make outdoor living hazardous or impossible for any sustained period of time.

One day, a terrible political crisis in a far-off land creates 6.5 million refugees. No other country on Earth will take them in. Since O is known as an “open border” nation, O becomes the most obvious candidate for taking them in. Unfortunately, since their sudden appearance on O’s shores is an “exogenous shock” to O’s housing market, O can’t immediately build enough new housing to accommodate them. So it calls on the citizens of O voluntarily to take them in. Some do, but not nearly enough to absorb the remaining 6 million refugees. O can only voluntarily absorb 500,000 refugees.

Suppose that the 6 million remaining (potentially homeless) refugees are now about to enter O (they’re being held in a flotilla just offshore, perched half-way between entry and expulsion). If they are sent back, they die. If they are admitted, they will be homeless. Being homeless, and constituting a full fourth of O’s population, they will almost certainly create havoc designed to overturn the political system and demand a right to housing (and possibly other positive rights). They will demand that they be housed in the extra domiciles of the rich, and in the extra space in the domiciles of the rich, middle class, and poor. Considering these risks, O turns them away. They perish. O is forever execrated in human history as a heartless, hypocritical country. But its rights-respecting regime remains in place for a thousand years.

Upshot: The thought-experiment suggests that you can’t have an “open borders” policy that ignores the possibility of exogenous shocks to the economy of the country whose borders are supposed to be open. Either immigration policy has to be sufficiently “closed” to avoid the influx of refugees caused by an exogenous shock, or the inhabitants of O must be willing to court the subversion of their societies in order to maintain their commitment to an open border policy, even if the open border policy can be predicted to lead to a rights-violative regime.

The “upshot” is so weak that I worried at first that it might be thought trivial. But I don’t think it is, and I don’t think libertarians will take it that way.

My thought-experiment has a (remote) basis in three major historical events–immigration policy in Mandate Palestine and the United States during the Holocaust; housing policy in the Soviet Union just after the Bolshevik Revolution; and the partition of India and Pakistan.  (I don’t mean that the thought-experiment mirrors any of the three cases, but that reflection on those cases and others like them inspired the thought-experiment.)

The Holocaust claimed the lives of 6 million Jews, but it would have claimed the lives of more Jews if the Jews of Europe had not fled Europe en masse for other locations. Jewish immigration to the United States was closed, but immigration into Mandate Palestine was not. So European Jews fled to Mandate Palestine, which was under British control. The Arabs of Mandate Palestine protested the in-migration of Jews on the grounds that the Jews (the Zionists) intended to establish a Jewish state in a place where the majority population was non-Jewish. Some of this protest was mere anti-Semitism (or even Nazism), but not all of it was. Some of it was motivated by the justifiable fear of being dispossessed or turned into second-class citizens by a state governed by a minority ethnic population–a fear that in many cases was realized. (If you don’t like my putting things that way, just change the example and ask yourself whether you’d be willing to open Israel’s western border to Palestinian immigration and Palestinian enfranchisement. Philosophically speaking, nothing turns on whether we run the example the one way or the other.) Here, in my view, we have a legitimate basis for opposition to open borders. For an excellent discussion of part of this history, see Kenneth Stein, The Land Question in Palestine, 1917-1939.

It’s worth reflecting on the fact that while Mandate Palestine was expected to absorb millions of Jewish refugees, the United States was not. A comparison of the size of the two places, and the circumstances involved, makes one wonder what was going through the heads of the relevant decision-makers.

Anyone who’s read Ayn Rand’s We the Living will recall Rand’s depiction of the Domicile Norm in the early Soviet Union. For discussion, see Dina Schein Federman’s, “We the Living and the Rosenbaum Family Letters,” in Robert Mayhew’s Essays on Ayn Rand’s We the Living (pp. 67-70). I got part of my thought-experiment from that example.

Golden Temple, Amritsar, India

Golden Temple, Amritsar, India

Khawaja’s trade-off is also informed by my family’s experience during the partition of India and Pakistan in August 1947. During that event (in March 1947), my father’s family home and business in Amritsar, India were expropriated, and the family was forced to live for several months in refugee camps. In August 1947, they were forced to leave India essentially at gunpoint, and became refugees in the famous population transfer of that period, leaving Amritsar (India) for Lahore (Pakistan) and eventually settling in Jhang (Pakistan). (The family was Muslim, and the mobs forcing them out of India were Sikh.)  They found housing in Pakistan only because Hindus and Sikhs in Pakistan had themselves been expropriated or killed by Muslims, thereby leaving room for incoming Muslims who’d been forced out by Sikhs or Hindus in India. That was very fortunate for them, but you can’t always count on mass death or ethnic cleansing to free up space in cases like this. For a good discussion, see Yasmin Khan’s The Great Partition, or visit Reena Kapoor’s 1947 Partition Archive.

Questions about open borders, state sovereignty, migration, and immigration are complex. I don’t think they should be held hostage to the relative simplicity of the issues in the American debate.




  1. For versions of libertarianism that recognise only private property, your thought-experiment doesn’t seem to apply; there is nothing for immigrants to be granted permission to immigrate to, except individual plots of property, each owner of which can choose her own immigration policy.

    For those like me who recognise public (but non-state) property, matters are a bit more complicated, but the same basic principles apply.

    In other words, by libertarian standards “open borders” doesn’t mean “let everyone in” so much as “let people move onto any property whose owners are willing to accept them.”

  2. I think Eyal’s piece on the Anderson story is an (unintended) reductio ad absurdum of the instrumentalist/survivalist version of Randian ethics.

  3. Incidentally, the Anderson story is back in print, in Flandry’s Legacy, the last volume of the Technic Civilization series from Baen Books.

  4. And I don’t know what he thought of Rand, but he was certainly a libertarian of some sort.

  5. irfankhawaja says:

    No, I think the thought-experiment applies–possibly with minor adjustments–on both of your scenarios, and virtually any libertarian scenario.

    1. Anarchist private property only interpretation. Strictly speaking, this interpretation violates the “political system that enshrines individual rights in Rand’s sense” proviso (because Rand’s conception of rights wasn’t anarchist). So delete “in Rand’s sense” and replace it with “in ___’s sense” where the blank stands for an anarchist; you can still run the thought-experiment. Ex hypothesi, 6.5 million refugees are headed for O, where O is an anarchy but no political entity has sovereignty over O, so that “O” is just a conventional name for a certain geographical area. Ex hyp, half a million refugees are granted permission to immigrate to individual plots of land within O, but no more than that. The other 6 million die. Presumably the denizens of O have some mechanism for excluding any of the 6 million who try to enter. The first sentence of the Upshot, and the first disjunct of the second sentence of the Upshot, hold.

    2. Anarchist private/public property mix. Again, replace “Rand” with someone else. Here, the difference with (1) is trivial. The point is, whatever the private/public mix, 500,000 are let in, so 6 million perish. Same Upshots as (1).

    3. Limited government private property only interpretation. This is the Randian interpretation. Rand defines capitalism as a system in which all property, presumably including government property, is privately owned. So on this interpretation, you have a government of limited powers which has sovereignty over a geographic area. You have consenting citizens, and you have non-consenting denizens living in a state of nature. The consenting citizens own the government and its property. The denizens own their own property but not the government. But the government controls the borders. The “limited powers” of government then have to be defined, codified, and enforced in such a way as to be sufficiently closed to avoid exogenous shocks of the kind described in the thought-experiment. On this view, the government is doing the work “from above” that is done “from below” on an anarchist view. In my thought experiment, the government excludes the 6 million that are excluded by private owners in the anarchist iterations of the thought-experiment. But ex hypothesi, the same work somehow has to be done by someone, unless you bite the bullet of the second disjunct, and regard it as a moral imperative to let the 6 million in (which you could do whether you’re an anarchist or limited state person). At this point, some will say that given the egoist provision at the outset, there is no reason for O to care if the 6 million die. Some will regard that a reductio of egoism. And others will think egoism requires incorporating the 6 million. But then we’re onto a different topic, egoism.

    Related point: the Randian view requires that its defenders distinguish sovereignty and ownership as distinct relations between agents and things in the world, and then relate both things back to immigration policy in a single coherent way. For instance, if I own JFK airport, and the government has sovereignty over the borders (and JFK is a point of entry and therefore a border), the Randian view implies that my ownership rights have to be conceived so as to exclude the government’s legitimate authority to police the borders. That is, I may own the International Terminal at JFK, but I cannot, even as its owner, dictate policy to the Immigration and Customs Enforcement officials who sit in the terminal. That’s a matter of sovereignty. And yet, assuming that I’m a consenting citizen, since all ownership is private, the property of ICE is private as well–and indirectly mine as well. Both things–the airport and the institutional property of ICE–are mine under different legal descriptions. So my ownership rights have to be conceived–and restricted–accordingly. I think this is a coherent view, but don’t think Objectivists have really grasped the complexity involved in spelling it out.

    4. BHL libertarians agnostic as between limited government and anarchy and agnostic as between private-only and public/private mix. I leave this as an exercise (for them), but to start, they’d have to acknowledge that the thought-experiment was a legitimate one to think about.

    On my view, thought-experiments don’t do a lot of theoretical work. They clarify or precisify our thoughts, and they give us a map of the theoretical alternatives and associated “price structure,” but they don’t push us to any given conclusion. Hence the disjunctive Upshot and my postscript about how some might regard the thought-experiment as trivial. I don’t, of course.


  6. […] by Vipul Naik, May 27, 2014, linking to TAS Graduate Seminar (third in a series): thought-experiments and open borders by Irfan Khawaja, Institute for Objectivist Studies, August 12, 2013. 5 likes, 1 […]

  7. […] the borders of your country and be integrated with your border/immigration policy. If you confront dogmatists who insist on ‘open borders’ ask them whether open borders as they conceive of them require a […]

  8. […] the borders of your country and be integrated with your border/immigration policy. If you confront dogmatists who insist on ‘open borders,’ ask them whether open borders as they conceive of them require a […]

  9. […] post. For my skepticism on the anarchist-literalist version of “Open Borders,” read this post and the comments on this […]

  10. […] the borders of your country and be integrated with your border/immigration policy. If you confront dogmatists who insist on ‘open borders,’ ask them whether open borders as they conceive of them require a […]

Comments closed

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: