You are on page 1of 24

Aliens and Citizens: The Case for Open Borders Author(s): Joseph H.

Carens Reviewed work(s): Source: The Review of Politics, Vol. 49, No. 2 (Spring, 1987), pp. 251-273 Published by: Cambridge University Press for the University of Notre Dame du lac on behalf of Review of Politics Stable URL: . Accessed: 23/10/2012 19:48
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact

Cambridge University Press and University of Notre Dame du lac on behalf of Review of Politics are collaborating with JSTOR to digitize, preserve and extend access to The Review of Politics.

Aliens and Citizens: The Case for Open Borders

Joseph H. Carens
Many poor and oppressed people wish to leave their countries of origin in the third world to come to affluentWestern societies. This essay argues that there is littlejustification for keeping them out. The essay draws on three contemporary approaches to political theory- the Rawlsian, the Nozickean, and the utilitarian-to construct arguments for open borders. The fact that all three theories converge upon the same resultson this issue, despite their significant disagreements on others, strengthensthe case for open borders and reveals its roots in our deep commitmentto respect all human beings as free and equal moral persons. The final part of the essay considers communitarian objections to this conclusion, especially those of Michael Walzer.

Borders have guards and the guards have guns. This is an obvious factof politicallifebut one that is easily hidden fromview- at Westleast fromthe view of those of us who are citizensof affluent ern democracies. To Haitians in small, leaky boats confronted by armed Coast Guard cutters,to Salvadorans dying fromheat and lack of air afterbeing smuggled into the Arizona desert,to Guatesewer pipes fromMexico to malans crawlingthroughrat-infested California- to these people the borders, guards and guns are all too apparent. What justifiesthe use of forceagainst such people? Perhaps borders and guards can be justifiedas a way of keeping out criminals,subversives,or armed invaders. But most of those to trying get in are not like that. They are ordinary, peaceful peoto ple, seeking only the opportunity build decent, secure lives for themselvesand their families. On what moral grounds can these sortsof people be kept out? What gives anyone the rightto point guns at them? To most people the answer to this question will seem obvious. The power to admit or exclude aliens is inherentin sovereignty and essentialforany politicalcommunity. Every statehas the legal and moral rightto exercise that power in pursuit of its own national interest, even if that means denying entry to peaceful, needy foreigners.States may choose to be generous in admitting but theyare under no obligation to do so.' immigrants, I want to challengethatview. In thisessay I will argue thatborders should generallybe open and that people should normallybe freeto leave theircountryof origin and settlein another,subject only to the sorts of constraintsthat bind currentcitizens in their new country.The argument is strongest, believe, when applied I 251



to the migrationof people fromthirdworld countriesto those of the first world. Citizenship in Westernliberal democracies is the modern equivalent of feudal privilege- an inherited status that privigreatlyenhances one's life chances. Like feudal birthright leges, restrictivecitizenship is hard to justify when one thinks about it closely. In developingthis argumentI will draw upon three contemporary approaches to political theory: firstthat of Robert Nozick; second that of John Rawls; third that of the utilitarians.Of the three, I find Rawls the most illuminating,and I will spend the most time on the argumentsthat flowfromhis theory.But I do not want to tie my case too closely to his particularformulations is (which I will modifyin any event). My strategy to take advanof three well-articulatedtheoretical approaches that many tage a people findpersuasiveto construct varietyof argumentsfor(relborders. I will argue that all three approaches lead atively)open for to the same basic conclusion: thereis little justification restrictEach of these theoriesbegins with some kind of ing immigration. assumption about the equal moral worth of individuals. In one way or another,each treatsthe individual as prior to the community.These foundationsprovidelittlebasis fordrawingfundamental distinctionsbetween citizens and aliens who seek to become citizens. The fact that all three theoriesconvergeupon the same basic result with regard to immigrationdespite their significant in differences other areas strengthens the case for open borders. In the finalpart of the essay I will consider communitarianobjections to my argument,especiallythose of Michael Walzer, the best defenderof the view I am challenging. contemporary

One popular position on immigrationgoes somethinglike this: "It's our country.We can let in or keep out whomeverwe want." as This could be interpreted a claim that the rightto exclude aliens is based on property rights, perhaps collective or national property rights.Would thissortof claim receive supportfromtheories in which propertyrightsplay a centralrole? I thinknot, because those theoriesemphasize individual propertyrightsand the concept of collectiveor national propertyrightswould undermine the individual rightsthat these theorieswish to protect. of Consider Robert Nozick as a contemporaryrepresentative



the propertyrightstradition. Following Locke, Nozick assumes that individuals in the state of nature have rights,including the All individuals have the same rightto acquire and use property. natural rights--thatis the assumption about moral equality that underliesthistradition althoughthe exerciseof those rights leads to material inequalities. The "inconveniences" the state of natof ure justifythe creation of a minimal state whose sole task is to protectpeople within a given territory against violations of their rights.2 Would this minimal state be justified in restricting immigration? Nozick never answers this question directly, but his argument at a number of points suggestsnot. Accordingto Nozick the state has no right to do anythingother than enforce the rights which individuals already enjoy in the state of nature. Citizenclaim. The state is obliged to proship gives rise to no distinctive tectthe rightsof citizensand noncitizensequally because it enjoys a defacto of monopoly over the enforcement rightswithinits terriIndividuals have the rightto enter into voluntaryexchanges tory. with other individuals. They possess this rightas individuals, not with such exchanges so as citizens. The state may not interfere as theydo not violate someone else's rights.3 long Note what thisimpliesforimmigration.Suppose a farmerfrom the United States wanted to hire workersfromMexico. The governmentwould have no rightto prohibithim fromdoing this. To prevent the Mexicans from coming would violate the rights of both the American farmerand the Mexican workersto engage in voluntary transactions. Of course, American workers might be disadvantaged by this competitionwith foreignworkers.But Nozick explicitly denies that anyone has a right to be protected against competitive disadvantage. (To count that sortof thingas a harm would undermine the foundations of individualproperty rights.) Even if the Mexicans did not have job offersfrom an American, a Nozickean governmentwould have no grounds for them fromenteringthe country.So long as theywere preventing or peaceful and did not steal, trespasson private property, otherwise violate the rightsof other individuals, their entryand their actions would be none of the state'sbusiness. Does this mean that Nozick's theoryprovides no basis for the exclusion of aliens? Not exactly.It means ratherthat it provides no basis forthe state exclude aliens and no basis forindividuals to to exclude aliens that could not be used to exclude citizensas well.



Poor aliens could not affordto live in affluent suburbs (except in the servants'quarters),but thatwould be true of poor citizenstoo. Individual propertyowners could refuse to hire aliens, to rent them houses, to sell them food, and so on, but in a Nozickean world they could do the same things to their fellowcitizens. In other words, individuals may do what they like with their own personal property.They may normally exclude whomever they want fromland theyown. But they have this rightto exclude as individuals, not as members of a collective.They cannot prevent other individuals from acting differently (hiring aliens, renting themhouses, etc.).4 Is there any room for collectiveaction to restrict entryin Nozick's theory?In the finalsection of his book, Nozick draws a discomtinctionbetween nations (or states) and small face-to-face munities. People may voluntarily constructsmall communitieson from the ones that govern the state so principlesquite different as individuals are freeto leave these communities.For examlong ple, people may choose to pool theirpropertyand to make collectivedecisions on the basis of majorityrule. Nozick argues thatthis sort of communityhas a right to restrictmembership to those whom it wishes to admit and to controlentryto its land. But such a communitymay also redistribute jointly held propertyas it its chooses. This is not an option that Nozick (or any otherproperty rightstheorist)intendsto grantto the state.5 This shows why the claim "It's our country.We can admit or exclude whomever we want" is ultimately incompatible with a propertyrightstheorylike Nozick's. Propertycannot serve as a is for the protection individualsagainst collectiveifproperty collectivelyowned. If the notion of collectiveownershipis used to justify keeping aliens out, it opens the possibilityof using the same notion to justifyredistributing income or whateverelse the majority decides. Nozick explicitly says that the land of a nation is not the collectivepropertyof its citizens. It followsthat the control that the state can legitimately exercise over that land is limited to the enforcement the rightsof individual owners. Prohibiting of people fromenteringa territory because theydid not happen to be born there or otherwisegain the credentialsof citizenshipis no part of any state'slegitimatemandate. The state has no rightto restrict immigration.




In contrastto Nozick, John Rawls provides a justificationfor an activist state with positive responsibilitiesfor social welfare. Even so, the approach to immigrationsuggested by A Theory of in justiceleaves little room for restrictions principle. I say "suggested"because Rawls himselfexplicitlyassumes a closed system in which questions about immigrationcould not arise. I will argue, however,that Rawls's approach is applicable to a broader contextthan the one he considers. In what followsI assume a general familiaritywith Rawls's theory,brieflyrecalling the main points and then focusingon those issues that are relevantto my inquiry. Rawls asks what principlespeople would choose to governsociknowety if theyhad to choose frombehind a "veil of ignorance," ing nothingabout their own personal situations(class, race, sex, natural talents,religiousbeliefs,individual goals and values, and so on). He argues that people in this original position would choose two principles. The firstprinciple would guarantee equal libertyto all. The second would permitsocial and economic inequalities so long as theywere to the advantage of the least well off (the difference principle) and attached to positionsopen to all under fairconditionsof equal opportunity. People in the originalpoto sitionwould give priority the first a principle,forbidding reduction of basic libertiesforthe sake of economic gains.6 Rawls also draws a distinction between ideal and nonideal theIn ideal theoryone assumes that, even afterthe "veil of ignoory. rance" is lifted,people will accept and generallyabide by the principles chosen in the original position and that there are no historicalobstacles to the realization of just institutions. nonIn ideal theory, one takes account of both historicalobstacles and the unjust actions of others. Nonideal theory is thus more immediately relevantto practical problems,but ideal theoryis more funthe and a badamental, establishing ultimategoal of social reform sis for judging the relativeimportanceof departuresfromthe ideal of (e.g., the priority liberty).7 Like a number of other commentators,I want to claim that many of the reasons that make the original position useful in thinking about questions of justice within a given society also make it useful forthinking about justice across different societies.8 Cases like migrationand trade, where people interactacross gov-



ernmental boundaries, raise questions about whether the background conditions of the interactionsare fair. Moreover, anyone who wants to be moral will feel obliged to justifythe use of force against other human beings, whether they are members of the same society or not. In thinkingabout these matters we don't want to be biased by self-interested partisan considerations, or and we don't want existinginjustices(if any) to warp our reflections. Moreover,we can take it as a basic presuppositionthat we should treatall human beings, not just members of our own society,as freeand equal moral persons.9 The original position offers strategy moral reasoning that a of helps to address these concerns. The purpose of the "veil of ignorance" is "to nullify of the effects specificcontingencieswhich put men at odds" because natural and social contingenciesare "arbiare traryfroma moral point of view" and therefore factorswhich not to influence the choice of principles of justice.'0 ought Whetherone is a citizen of a rich nation or a poor one, whether one is already a citizenof a particularstateor an alien who wishes to become a citizen-this is the sort of specificcontingencythat could set people at odds. A fairprocedure forchoosing principles of justice must thereforeexclude knowledge of these circumstances,just as it excludes knowledgeof one's race or sex or social class. We should therefore take a global, not a national, view of the originalposition. One objection to this global approach is that it ignores the extent to which Rawls's use of the original position and the "veil of of ignorance"depends upon a particularunderstanding moral perof sonalitythat is characteristic modern democratic societies but may not be shared by other societies." Let us grantthe objection and ask whetherit reallymatters. The understandingof moral personalityin question is essentially the view that all people are free and equal moral persons. Even if this view of moral personalityis not shared by people in other societies, it is not a view that applies only to people who share it. Many members of our own societydo not share it, as illustratedby the recentdemonstrations whiteracistsin Forsythe by County, Georgia. We criticize the racists and reject their views but do not deprive them of theirstatus as freeand equal citizens because of theirbeliefs. Nor is our belief in moral equality limited to members of our own society.Indeed our commitmentto civic equality is derived fromour convictionsabout moral equal-



ity,not vice versa. So, whateverwe thinkabout thejustice of borders and the limitationsof the claims of aliens, our views must be compatible with a respect for all other human beings as moral persons. A related objection emphasizes the "constructivist" nature of Rawls's theory, in The theory particularly its later formulations.'2 only makes sense, it is said, in a situationwhere people already share liberal-democratic values. But ifwe presuppose a contextof shared values, what need have we for a "veil of ignorance"?Why not move directly fromthe shared values to an agreementon prinThe "veil of ignociples ofjustice and correspondinginstitutions? a rance" offers way of thinking about principlesofjustice in a context where people have deep, unresolvable disagreementsabout matters of fundamental importance and yet still want to find a way to live togetherin peaceful cooperation on termsthat are fair to all. That seems to be just as appropriatea contextforconsidering the problem of worldwidejustice as it is consideringthe problem of domesticjustice. To read Rawls's theoryonly as a constructive of interpretation existingsocial values is to undermine its potential as a constructive critiqueof thosevalues. For example, racism has deep rootsin American public culture, and in the not-too-distant past people like those in Forsythe a County constituted majorityin the United States. If we thinkthe racistsare wrong and Rawls is rightabout our obligation to treat all members of our society as free and equal moral persons, it is surely not just because the public culture has changed and the racistsare now in the minority. gladly I concede that I am using the original position in a way that Rawls himselfdoes not intend, but I think that this extension is warrantedby the nature of the questions I am addressingand the virtues of Rawls's approach as a general method of moral reasoning. Let us therefore assume a global view of the original position. Those in the original position would be preventedby the "veil of ignorance"fromknowingtheirplace of birthor whethertheywere members of one particular society rather than another. They would presumably choose the same two principles of justice. (I will simplyassume that Rawls's argumentforthe two principlesis correct,though the point is disputed.) These principleswould apto ply globally,and the next task would be to design institutions of implementthe principles stillfromthe perspective the original include sovereignstates as they position. Would these institutions



exist? In ideal theory, where we can assume away historcurrently ical obstacles and the dangers of injustice,some of the reasons for of defendingthe integrity existingstatesdisappear. But ideal theory does not require the eliminationof all linguistic,cultural,and Let historicaldifferences. us assume that a general case fordecentralizationof power to respectthese sorts of factorswould justify the existenceof autonomous political communitiescomparable to That does not mean that all the existingfeatures modern states.1" of state sovereignty would be justified. State sovereignty would be (morally) constrainedby the principlesofjustice. For example, no statecould restrict religiousfreedomand inequalitiesamong states difference would be restricted an international principle. by What about freedomof movementamong states?Would it be regarded as a basic libertyin a global systemof equal liberties,or would states have the rightto limit entry and exit? Even in an ideal worldpeople mighthave powerful reasons to want to migrate fromone state to another. Economic opportunities particular for individuals might vary greatlyfromone state to another even if economic inequalities among states were reduced by an international difference principle. One might fall in love with a citizen fromanother land, one mightbelong to a religionwhich has few in followers one's nativeland and many in another,one mightseek cultural opportunitiesthat are only available in another society. More generally, one has only to ask whetherthe rightto migrate freelywithina given society is an importantliberty.The same sortsof considerationsmake migrationacross stateboundaries important.14

Behind the "veil of ignorance,"in consideringpossible restrictions on freedom, one adopts the perspective of the one who in would be most disadvantaged by the restrictions, this case the perspectiveof the alien who wants to immigrate.In the original position, then, one would insist that the rightto migrate be included in the systemof basic libertiesfor the same reasons that one would insistthat the rightto religiousfreedombe included: it mightprove essentialto one's plan of life. Once the "veil of ignorance" is lifted,of course, one might not make use of the right, but that is true of other rightsand libertiesas well. So, the basic agreementamong those in the original position would be to permit no restrictions migration(whetheremigrationor immigraon tion). There is one important qualification to this. According to



Rawls, libertymay be restrictedfor the sake of liberty even in ideal theoryand all libertiesdepend on the existenceof public order and security.'5 (Let us call this the public order restriction.) immigrationwould lead to chaos and Suppose that unrestricted the breakdown of order. Then all would be worse offin termsof theirbasic liberties.Even adopting the perspectiveof the worst-off those in the original posiand recognizingthe priorityof liberty, on tion would endorse restrictions immigrationin such circumstances. This would be a case of restricting libertyforthe sake of even libertyand everyindividual would agree to such restrictions one mightfindthat though,once the "veil of ignorance"was lifted, it was one's own freedomto immigratewhich had been curtailed. Rawls warns against any attemptto use thissortof public order argumentin an expansive fashionor as an excuse forrestrictions undertakenforotherreasons. The hypothetical on liberty possibilwould be of a threatto public order is not enough. Restrictions ity justifiedonly if there were a "reasonable expectation"that unlimwould damage the public order and thisexpectaited immigration tion would have to be based on "evidence and ways of reasoning would be justifiedonly acceptable to all."''6 Moreover, restrictions to the extentnecessaryto preservepublic order. A need forsome would not justifyany level of restrictions whatsoever. restrictions Finally,the threatto public order posed by unlimitedimmigration could not be the productof antagonisticreactions(e.g., riots)from currentcitizens. This discussion takes place in the contextof ideal theory and in this context it is assumed that people try to act justly. Rioting to prevent others fromexercisinglegitimatefreedoms would not be just. So, the threatto public order would have to be one that emerged as the unintendedcumulativeeffect inof dividuallyjust actions. In ideal theorywe face a world of just states with an international difference principle. Under such conditions,the likelihood of mass migrationsthreatening the public order of any particuto lar state seems small. So, there is littleroom for restrictions on immigration in ideal theory. But what about nonideal theory, where one takes into account both historicalcontingencies and the unjust actions of others? In the nonideal, real world there are vast economic inequalities among nations (presumably much larger than would exist under an internationaldifference principle). Moreover, people disagree about the nature of justice and often fail to live up to whatever



principlestheyprofess.Most statesconsiderit necessaryto protect of themselvesagainst the possibility armed invasion or covertsubversion.And many statesdeprivetheirown citizensof basic rights and liberties. How does all this affectwhat justice requires with regard to migration? First, the conditions of the real world greatly strengthenthe case forstate sovereignty, especially in those statesthat have relaNational security a crucial form is tively just domesticinstitutions. of public order. So, statesare clearlyentitledto preventthe entry of people (whetherarmed invaders or subversives)whose goal is of the overthrow just institutions.On the other hand, the strictures against an expansive use of the public order argumentalso apply to claims about national security. A related concern is the claim that immigrantsfrom societies where liberal democraticvalues are weak or absent would pose a threatto the maintenanceof a just public order.Again the distinction between reasonable expectations and hypotheticalspeculations is crucial. These sorts of argumentswere used during the nineteenthcenturyagainst Catholics and Jews fromEurope and against all Asians and Africans. If we judge those arguments to have been provenwrong (not to say ignorantand bigoted) by hiswe them in another guise. tory, should be wary of resurrecting A more realisticconcern is the sheer size of the potential demand. If a rich country like the United States were simply to open its doors, the number of people frompoor countriesseeking to immigratemighttrulybe overwhelming, even iftheirgoals and beliefsposed no threatto national securityor liberal democratic values."7 Under these conditions,it seems likelythat some restrictions on immigrationwould be justified under the public order principle. But it is importantto recall all the qualificationsthat would apply to this. In particular,the need for some restriction not justify any level of restriction whatsoever or restrictions for other reasons, but only that level of restriction essential to maintain public order. This would surelyimply a much less restrictive in policy than the one currently forcewhich is shaped by so many otherconsiderationsbesides the need to maintain public order. Rawls asserts that the priorityaccorded to liberty normally holds under nonideal conditions as well. This suggests that, if thereare restrictions immigration public order reasons, prion for orityshould be given to those seeking to immigratebecause they have been denied basic libertiesover those seeking to immigrate



simply for economic opportunities.There is a further complicaof tion, however.The priority libertyholds absolutelyonly in the long run. Under nonideal conditions it can sometimesbe justifiable to restrict libertyforthe sake of economic gains, if that will and speed the creation of improve the position of the worst-off conditionsin which all will enjoy equal and fullliberties.Would it be justifiableto restrict for immigration the sake of the worst-off? We have to be wary of hypocriticaluses of this sort of argument. If rich statesare reallyconcernedwiththe worst-off poor in resources states, they can presumably help more by transferring and reforming international economic institutions than by restricting immigration.Indeed, there is reason to suppose more open would help some of the worst-off, hurt them. At not immigration the least, those who immigratepresumably gain themselvesand oftensend money back home as well. however.It is Perhaps the ones who come are not the worst-off, don't have the resourcesto plausible to suppose that the worst-off leave. That is still no reason to keep others fromcoming unless their departure hurtsthose leftbehind. But let's suppose it does, as the brain-drainhypothesis suggests.If we assume some restrictions on immigrationwould be justifiedforpublic order reasons, this would suggestthat we should give priority the least skilled to because their departure would preamong potential immigrants on sumably have littleor no harmfuleffect those leftbehind. It also suggestthat compensation was due to poor countries might when skilled people emigrate. But to say that we should actually tryto keep people fromemigrating(by denyingthem a place to go) because theyrepresenta valuable resource to theircountryof originwould be a dramatic departurefromthe liberal traditionin general and fromthe specificprioritythat Rawls attaches to libertyeven under nonideal conditions.'8 Consider the implicationsof this analysis for some of the conventional arguments for restrictions immigration. First, one on could not justifyrestrictions the grounds that those born in a on or given territory born of parentswho were citizenswere more entitledto the benefitsof citizenshipthan those born elsewhere or born of alien parents. Birthplace and parentage are natural confroma moral point of view."One of tingenciesthat are "arbitrary the primarygoals of the originalpositionis to minimize the effects of such contingenciesupon the distribution social benefits.To of assign citizenship on the basis of birth might be an acceptable



procedure,but only ifit did not preclude individualsfrommaking different choices later when theyreached maturity. on Second, one could not justifyrestrictions the grounds that would reduce the economic well-beingof currentcitiimmigration zens. That line of argumentis drastically limitedby two considerations: the perspectiveof the worst-off and the priority liberty. of In order to establish the currentcitizens'perspectiveas the relevant worst-off position, it would be necessaryto show that immiwould reduce the economic well-beingof currentcitizens gration below the level the potentialimmigrants would enjoy if theywere not permitted immigrate.But even if this could be established, to it would notjustifyrestrictions immigration on because of the priof liberty. the economic concerns of currentcitizens are So, ority essentiallyrenderedirrelevant. of Third, the effect immigrationon the particularculture and of history the societywould not be a relevantmoral consideration, so long as there was no threatto basic liberal democraticvalues. This conclusion is less apparent fromwhat I have said so far,but it followsfrom what Rawls says in his discussion of perfectionwould require social instituism.19The principleof perfectionism tions to be arranged so as to maximize the achievementof human excellencein art, science, or cultureregardlessof the effect such of on equality and freedom.(For example, slaveryin arrangements ancient Athens has sometimesbeen defendedon the grounds that it was essentialto Athenian culturalachievements.)One variantof this position mightbe the claim that restrictions immigration on would be necessaryto preservethe unity and coherence of a culture (assuming that the culturewas worthpreserving).Rawls argues that in the original position no one would accept any perfectionist standard because no one would be willing to risk the possibilityof being required to forego some importantrightor freedom for the sake of an ideal that might prove irrelevantto one's own concerns. So, restrictions immigration the sake of on for culturewould be ruled out. preservinga distinctive In sum, nonideal theoryprovides more grounds for restricting than ideal theory, but these groundsare severely limimmigration ited. And ideal theoryholds up the principleof freemigrationas an essential part of the just social order toward which we should strive.




A utilitarianapproach to the problem of immigrationcan take into account some of the concerns that the original position excludes but even utilitarianism does not provide much support for the sorts of restrictions immigrationthat are common today. on The fundamentalprinciple of utilitarianismis "maximize utility" and the utilitariancommitmentto moral equality is reflectedin the assumption that everyoneis to count for one and no one for more than one when utilityis calculated. Of course, these broad formulationscover over deep disagreementsamong utilitarians. For example, how is "utility" be defined?Is it subjectiveor obto jective? Is it a question of happiness or welfareas in classical utilitarianismor preferences interests in some more recentveror as sions?20 However these questions are answered, any utilitarian apimproach would give more weightto some reasons forrestricting than Rawls's approach would. For example, ifmore immigration migration would hurt some citizens economically, that would count against a more open immigrationpolicy in any utilitarian theoryI am familiarwith. But that would not settlethe question of whetherrestrictions werejustified,forothercitizensmightgain economicallyfrommore immigrationand thatwould count in favor of a more open policy.More importantly, economic effects the of more immigration noncitizenswould also have to be considon ered. If we focus only on economic consequences, the best immigrationpolicy froma utilitarianperspectivewould be the one that maximized overalleconomic gains. In thiscalculation, currentcitizens would enjoy no privilegedposition. The gains and losses of aliens would countjust as much. Now the dominant view among both classical and neoclassical economistsis that the freemobility of capital and labor is essentialto the maximizationof overalleconomic gains. But the freemobilityof labor requiresopen borders. So, despite the factthat the economic costs to currentcitizens are morally relevantin the utilitarianframework, theywould probato bly not be sufficient justifyrestrictions. Economic consequences are not the only ones that utilitarians consider. For example, if immigrationwould affectthe existing culture or way of life in a society in ways that current citizens found undesirable, that would count against open immigrationin many versions of utilitarianism.But not in all. Utilitarians dis-



agree about whetherall pleasures (or desires or interests)are to count or only some. For example, should a sadist's pleasure be given moral weight and balanced against his victim's pain or should that sort of pleasure be disregarded? What about racial prejudice? That is clearlyrelevantto the question of immigration. Should a white racist'sunhappiness at the prospectof associating withpeople of color be counted in the calculus of utilityas an argument in favorof racial exclusion as reflected, say, in the White Australia policy? What about the desire to preservea distinctive local culture as a reason for restrictingimmigration? That is sometimeslinked to racial prejudice but by no means always. will answer these sortsof questions in difutilitarians Different ferentways. Some argue that only long-term,rational, or otherwise refinedpleasures (or desires or interests)should count. Others insistthat we should not look behind the raw data in making should count, not merely our calculations. Everyone'spreferences someone else findsacceptable. I favorthe former the preferences or but I approach to utility, approach, a reconstructive filtering won't tryto defend that here. Even if one takes the raw data approach, which seems to leave more room for reasons to restrict immigration,the final outcome is stilllikelyto favormuch more than is common today.Whateverthe methodof open immigration the concerns of aliens must be counted too. Under calculation, currentconditions,when so many millionsof poor and oppressed people feel theyhave so much to gain frommigrationto the advanced industrialstates, it seems hard to believe that a utilitarian calculus which took the interests aliens seriouslywould justify of limitson immigrationthan the ones entailed significantly greater by the public order restriction implied by the Rawlsian approach.

The threetheoriesI have discussed conflict withone anotheron issues but not (deeply) on the question of immimany important gration. Each leads on its own termsto a position far more favorable to open immigrationthan the conventionalmoral view. It is true that, in terms of numbers, even a public order restriction might exclude millions of potential immigrantsgiven the size of the potentialdemand. Nevertheless, the argumentsI have develif oped here were accepted, theywould require a radical transforma-



tion both of current immigration policies and of conventional moral thinkingabout the question of immigration. Some may feel that I have wrenched these theoriesout of context. Each is rooted in the liberal tradition.Liberalism, it might be said, emerged with the modern state and presupposes it. Liberal theorieswere not designed to deal with questions about aliens. They assumed the contextof the sovereignstate. As a historical observation this has some truth, but it is not clear why it should have normativeforce. The same wrenchingout of context complaintcould as reasonablyhave been leveled at those who first liberal argumentsfor the extensionof full citizenship constructed to women and membersof the workingclass. Liberal theoriesalso assumed the rightto exclude them. Liberal theoriesfocus attention on the need to justifythe use of forceby the state. Questions about the exclusion of aliens arise naturally from that context. Liberal principles(like most principles)have implicationsthat the original advocates of the principlesdid not entirelyforesee. That is part of what makes social criticism possible. the Others may thinkthat my analysis merelyillustrates inadeof liberal theory,especially its inability to give sufficient quacy weight to the value of community.21 That indictmentof liberal theorymay or may not be correct,but my findingsabout immigration rest primarilyon assumptions that I thinkno defensible and public polmoral theorycan reject: thatour social institutions as moral persons and thatthis icies must respectall human beings respect entails recognition, in some form, of the freedom and equality of everyhuman being. Perhaps some otherapproach can accept these assumptions while still making room for greaterreI strictions immigration.To test that possibility, will consider on the views of the theoristwho has done the most to translatethe vision: Michael communitariancritique into a positivealternative Walzer. Unlike Rawls and the others, Walzer treats the question of membershipas central to his theoryof justice, and he comes to the opposite conclusion about immigrationfrom the one that I have defended: thatare made, states are Across considerable a rangeofthedecisions in free simply to takestrangers (or not).22 I Walzer differs fromthe other theorists have considered not only



in his conclusions but also in his basic approach. He eschews the search foruniversal principlesand is concerned instead with "the He particularismof history,culture, and membership."'23 thinks that questions of distributive justice should be addressed not from behind a "veil of ignorance"but fromthe perspectiveof membership in a political communityin which people share a common cultureand a common understandingabout justice. I cannot do fulljustice here to Walzer's rich and subtle discussion of the problem of membership,but I can draw attentionto the main points of his argumentand to some of the areas of our disagreement.Walzer's central claim is that exclusion is justified The by the rightof communitiesto self-determination. rightto exclude is constrainedin three importantways, however.First, we have an obligation to provide aid to otherswho are in dire need, even ifwe have no establishedbonds with them, provided thatwe can do so without excessive cost to ourselves. So, we may be or obliged to admit some needy strangers at least to provide them with some of our resources and perhaps even territory. Second, once people are admitted as residents and participants in the economy, they must be entitled to acquire citizenship, if they wish. Here the constraint flowsfromprinciplesofjustice not mutual aid. The notion of permanent"guest workers"conflicts with the underlyingrationale of communal self-determination which justifiedthe rightto exclude in the first place. Third, new statesor governments may not expel existinginhabitantseven if they are regarded as alien by most of the restof the population.24 In developinghis argument,Walzer compares the idea of open stateswithour experienceof neighborhoodsas a formof open asBut in thinking about what open stateswould be like, sociation.25 we have a bettercomparison at hand. We can draw upon our experience of cities, provinces, or states in the American sense. These are familiarpolitical communitieswhose borders are open. Unlike neighborhoodsand like countries,theyare formally organized communitieswith boundaries, distinctions between citizens and noncitizens,and elected officials who are expected to pursue policies that benefitthe members of the communitythat elected them. They oftenhave distinctive culturesand ways of life. Think of the differences between New York City and Waycross,Georgia, or between California and Kansas. These sortsof differences are oftenmuch greaterthan the differences across nation-states. Seattle has more in common with Vancouver than it does with many



American communities. But cities and provinces and American states cannot restrict immigration(from other parts of the country). So, these cases call into question Walzer's claim that distinctivenessdepends on the possibilityof formal closure. What makes for distinctiveness and what erodes it is much more complex than political controlof admissions. This does not mean that control over admissions is unimportant. Often local communitieswould like to restrict immigration. The people of California wanted to keep out poor Oklahomans during the Depression. Now the people of Oregon would like to keep out the Californians. Internal migrationscan be substantial. the They can transform characterof communities.(Think of the migrationsfromthe rural South to the urban North.) They can to place strainson the local economy and make it difficult maintain locally funded social programs. Despite all this, we do not thinkthese political communitiesshould be able to control their borders. The rightto freemigrationtakes priority. Why should this be so? Is it just a choice that we make as a the self-deterlarger community(i.e., the nation state) to restrict minationof local communitiesin thisway? Could we legitimately interpermitthem to exclude? Not easily. No liberal state restricts nal mobility.Those states that do restrictinternal mobility are criticizedfordenyingbasic human freedoms.If freedomof movementwithinthe stateis so important thatit overridesthe claims of local political communities,on what grounds can we restrict freedom of movementacross states?This requires a stronger case for the moral of distinctiveness the nation-stateas a formof communitythan Walzer's discussion of neighborhoodsprovides. Walzer also draws an analogy betweenstatesand clubs.26Clubs may generallyadmit or exclude whomever they want, although any particulardecision may be criticizedthroughan appeal to the characterof the club and the shared understandingsof its members. So, too, with states. This analogy ignores the familiardistinction between public and private, a distinction that Walzer makes use of elsewhere."27 There is a deep tension between the rightof freedomof association and the rightto equal treatment. One way to address thistensionis to say that in the privatesphere freedom of association prevails and in the public sphere equal treatment does. You can pick your friends the basis of whatever on criteriayou wish, but in selectingpeople foroffices you must treat all candidates fairly. Drawing a line between public and privateis



oftenproblematic,but it is clear that clubs are normally at one end of the scale and states at the other. So, the fact that private clubs may admit or exclude whomever they choose says nothing about the appropriate admission standards for states. When the state acts it must treatindividualsequally. Against this, one may object that the requirement of equal of treatmentapplies fullyonly to those who are already members the community.That is accurate as a descriptionof practice but the question is why it should be so. At one time, the requirement of equal treatmentdid not extend fullyto various groups (workers, blacks, women). On the whole, the historyof liberalism reof a flects tendencyto expand both the definition the public sphere and the requirementsof equal treatment.In the United States today, forexample, in contrastto earlier times,both public agencies and private corporationsmay not legally exclude women simply because they are women (although private clubs still may). A white shopkeepermay no longer exclude blacks fromhis store(althoughhe may exclude them fromhis home). I thinkthese recent developments,like the earlier extension of the franchise,reflect The somethingfundamentalabout the inner logic of liberalism.28 the same logic: equal extensionof the rightto immigratereflects of treatment individuals in the public sphere. As I noted at the beginningof this section, Walzer assertsthat the political community is constrained by principles of justice fromadmittingpermanentguest workerswithoutgivingthem the opportunityto become citizens. There is some ambiguity about whetherthisclaim is intendedto apply to all politicalcommunities or only to ones like ours. If stateshave a rightto self-determination, broadly conceived, theymust have a rightto choose political formsand politicalpracticesdifferent fromthose of liberal democracies. That presumablyincludes the rightto establishcategories of second-classcitizens(or, at least, temporaryguestworkers)and also the rightto determineother aspects of admissions policy in accordance with theirown principles."29 if the question is what But oursociety(or one with the same basic values) ought to do, then the matteris different both forguest workersand forotheraliens. It is rightto assertthat oursocietyought to admit guestworkers to full citizenship. Anything else is incompatible with our liberal democratic principles. But so is a restrictive policy on immigration. Any approach like Walzer's that seeks its ground in the tradi-



tion and culture of our communitymust confront,as a methodological paradox, the fact that liberalism is a central part of our culture. The enormous intellectualpopularityof Rawls and Nozick and the enduring influence of utilitarianismattest to their abilityto communicate contemporary understandingsand shared meanings in a language thathas legitimacyand power in our culture. These theories would not make such sense to a Buddhist monk in medieval Japan. But their individualistic assumptions and theirlanguage of universal,ahistoricalreason makes sense to For people us because of ourtradition,ourculture,ourcommunity. moral tradition, one that assumed fundamental in a different between those inside the societyand those outmoral differences restrictions immigrationmightbe easy to justify.Those on side, who are other simply might not count, or at least not count as much. But we cannot dismiss the aliens on the ground that they are other,because we are the productsof a liberal culture. The point goes stilldeeper. To take ourcommunityas a starting point is to take a community that expresses its moral views in this. termsof universalprinciples.Walzer's own argumentsreflect When he asserts that states may not expel existing inhabitants whom the majorityor the new government regards as alien, he is making a claim about what is rightand wrong for any state not just our own or one that shares our basic values. He develops the argument by drawing on Hobbes. That is an argument froma one thatmay not be shared by new statesthat particulartradition, want to expel some of their inhabitants. Nonetheless, Walzer makes a universal claim (and one I consider correct). He makes the same sort of argumentwhen he insiststhat statesmay not leThis applies to all political comrestrict gitimately emigration.3? munitiesnot just those that share our understandingof the relation of individual and collective. of Recognition of the particularity our own culture should not preventus frommaking these sorts of claims. We should not try to forceothersto accept our views, and we should be ready to listen to othersand learn fromthem. But respectforthe diversity of communities does not require us to abandon all claims about what other states ought to do. If my argumentsare correct,the general case foropen borders is deeply rooted in the fundamental values of our tradition.No moral argumentwill seem acceptable to us, if it directlychallenges the assumption of the equal moral worthof all individuals. If restrictions immigrationare to be on



justified, they have to be based on arguments that respect that principle. Walzer's theoryhas many virtues that I have not explored here, but it does not supply an adequate argumentforthe state'srightto exclude.

Free migrationmay not be immediatelyachievable, but it is a goal towardwhich we should strive.And we have an obligationto open our borders much more fullythan we do now. The current on restrictions immigrationin Westerndemocracies- even in the most open ones like Canada and the United States- are not justifiable. Like feudal barriersto mobility, theyprotectunjust privilege. between aliDoes it followthatthereis no room fordistinctions ens and citizens, no theoryof citizenship,no boundaries for the community?Not at all. To say thatmembershipis open to all who betweenmemwish to join is not to say thatthereis no distinction in bers and nonmembers.Those who choose to cooperate together the statehave special rightsand obligationsnot shared by noncitithat inzens. Respectingthe particularchoices and commitments dividuals make flowsnaturallyfroma commitment the idea of to for equal moral worth.(Indeed, consent as a justification political is least problematicin the case of immigrants.)What is obligation notreadily compatible with the idea of equal moral worthis the exclusion of those who want to join. If people want to sign the social contract,theyshould be permitted do so. to borderswould threatenthe distinctive characterof differOpen ent political communitiesonly because we assume that so many people would move if they could. If the migrants were few, it would not matter.A few immigrantscould always be absorbed withoutchangingthe characterof the community. And, as Walzer observes, most human beings do not love to move.3' They normally feel attached to theirnative land and to the particularlanguage, culture, and community in which they grew up and in which theyfeelat home. They seek to move only when lifeis very difficult where theyare. Their concernsare rarelyfrivolous.So, it is rightto weigh the claims of those who want to move against the claims of those who want to preservethe communityas it is. And ifwe don't unfairly the scales, the case forexclusion will rarely tip triumph.



People live in communitieswith bonds and bounds, but these kinds. In a liberal society, the bonds and may be of different bounds should be compatible withliberal principles.Open immibut it would grationwould change the characterof the community not leave the communitywithoutany character.It might destroy old ways of life,highlyvalued by some, but it would make possible new ways of life, highlyvalued by others. The whites in Forto sytheCounty who want to keep out blacks are trying preservea way of lifethatis valuable to them. To deny such communitiesthe rightto exclude does limit theirabilityto shape theirfuturecharbut it does not utterly acter and destiny, destroytheircapacity for self-determination. Many aspects of communal life remain potentially subject to collective control. Moreover, constraining the kinds of choices that people and communitiesmay make is what principlesofjustice are for.They set limitson what people seeking to abide by these principlesmay do. To commitourselvesto open borderswould not be to abandon the idea of communal character of but to reaffirm It would be an affirmation the liberal characit. ter of the communityand of its commitment principlesof justo tice.
NOTES This paper was firstwrittenforan APSA seminar on citizenship directed by Nan Keohane. Subsequent versions were presented to seminars at the University of Chicago, the Institute for Advanced Study, and Columbia University.I would like to thank the members of these groups for their comments. In addition I would like to thank the following individuals for helpful comments on one of the many drafts: Sot Barber, Charles Beitz, Michael Doyle, Amy Gutmann, Christine Korsgaard, Charles Miller, Donald Moon, JenniferNedelsky, Thomas Pogge, Peter Schuck, Rogers Smith, Dennis Thompson, and Michael Walzer. 'The conventional assumption is captured by the Select Commission on Immigration and Refugee Policy: "Our policy-while providing opportunityto a portion of the world's population - must be guided by the basic national interests of the people of the United States." From U S. Immigration Policyand theNaon The and tionalInterest: Final Report RecommendationstheSelectCommission Immiof and States(1 March and of Policyto theCongress thePresident theUnited gration Refugee 1981). The best theoretical defense of the conventional assumption (with some modifications) is Michael Walzer, SpheresofJustice(New York: Basic Books, 1983), pp. 31-63. A few theoristshave challenged the conventional assumption. in See Bruce Ackerman, SocialJustice theLiberalState(New Haven: Yale University Press, 1980), pp. 89-95; Judith Lichtenberg, "National Boundaries and NationalAutonomy and Moral Boundaries: A Cosmopolitan View" in Boundaries.: Its Limits,ed. Peter G. Brown and Henry Shue (Totowa, NJ: Rowman and Littlefield, 1981), pp. 79-100, and Roger Nett, "The Civil Right We Are Not Ready For: The Right of Free Movement of People on the Face of the Earth," Ethics81:212-27. Frederick Whelan has also explored these issues in two inter-



esting unpublished papers. 2Robert Nozick, Anarchy, State, and Utopia (New York: Basic Books, 1974), pp. 10-25, 88-119. 3Ibid., pp. 108-113. Citizens, in Nozick's view, are simply consumers purchasing impartial, efficient protectionof preexistingnatural rights. Nozick uses the terms "citizen;' "client" and "customer"interchangeably. 4Nozick interpretsthe Lockean proviso as implying that property rights in land may not so restrictan individual's freedom of movement as to deny him effectiveliberty.This furtherlimits the possibility of excluding aliens. See p. 55. 56Ibid., pp. 320-23. 6John Rawls, A TheoryofJustice(Cambridge, MA.: Harvard University Press, 1971), pp. 60-65, 136-42, 243-48. 7Ibid., pp. 8-9, 244-48. 8The argument fora global view of the original position has been developed most fullyin Charles Beitz, PoliticalTheory International Relations and (Princeton, NJ: Princeton University Press, 1979), pp. 125-76, especially 129-36 and 14353. For earlier criticismsof Rawls along the same lines, see Brian Barry, The Liberal TheoryofJustice (Oxford: Clarendon Press, 1973), pp. 128-33 and Thomas M. Scanlon, "Rawls's Theory of Justice," University Pennsylvania Law of Review121, no. 5 (May 1973): 1066-67. For more recent discussions, see David A. J. Richards, "International DistributiveJustice,"in Ethics,Economics, and the Law, eds. J. Roland Pennock and John Chapman (New York: New York UniversityPress, 1982), pp. 275-99 and Charles Beitz, "Cosmopolitan Ideals and National Sentiments,"TheJournalof Philosophy no. 10 (October 1983): 59180, 600. None of these discussions fullyexplores the implications of a global view of the original position for the issue of immigration, although the recent essay by Beitz touches on the topic. 9Respecting others as free and equal moral persons does not imply that one cannot distinguish friends from strangers or citizens from aliens. See the conclusion for an elaboration. '0Rawls, Justice, pp. 136, 72. "John Rawls, "Kantian Constructivism in Moral Theory,"TheJournalofPhi77, losophy no. 9 (September 1980): 515-72. as '2Ibid. See also John Rawls, '"Justice Fairness: Political not Metaphysical," and 14 Philosophy PublicAffairs (Summer 1985): 223-51. 13Compare Beitz, PoliticalTheory, 183. p. 14For more on the comparison of mobility within a country and mobility across countries, see Joseph H. Carens, "Migration and the Welfare State" in and State,ed. Amy Gutmann (Princeton: Princeton UniverDemocracy theWelfare sity Press, 1987). pp. 212-13. 15Rawls,Justice, 16Ibid.,p. 213. 17For statisticson current and projected levels of immigration to the U.S., see Michael S. Teitelbaum, "Right Versus Right: Immigration and Refugee 18For the deep roots of the right to emigrate in the liberal tradition, see Frederick Whelan, "Citizenship and the Right to Leave," American PoliticalScienceReview75, no. 3 (September 1981): 636-53. 'gRawls,Justice, pp. 325-32. 20Forrecent discussions of utilitarianism,see Richard Brandt, A Theory the of Goodand theRight(Oxford: Oxford UniversityPress, 1979); Peter Singer, Practical Ethics(Cambridge: Cambridge University Press, 1979); R. M. Hare, Moral Thinking(Oxford: Oxford University Press, 1981); and Amartya Sen and Bernard Williams, eds., Utilitarianism Beyond and (Cambridge: Cambridge Uni-

59 Policyin theUnitedStates," Foreign Affairs (1980): 21-59.



versityPress, 1982). 21Forrecent communitarian critiques of liberalism, see Alasdair MacIntyre, AfterVirtue (Notre Dame: Notre Dame University Press, 1981) and Michael Sandel, Liberalismand the Limits ofJustice(New York: Cambridge University Press, 1982). For a critique of the critics, see Amy Gutmann, "Communitarian Critics of Liberalism" Philosophy PublicAffairs (Summer 1985): 308-322. and 14 22Walzer,Spheres, 61. p. 23Ibid.,p. 5. 24Ibid.,pp. 33, 45-48, 55-61, 42-44. 25Ibid.,pp. 36-39. 26Ibid.,pp. 39-41. 27Ibid.,pp. 129-64. 28I am not arguing that the changes in treatment of women, blacks, and workers were brought about by the inner logic of liberalism. These changes resulted fromchanges in social conditions and frompolitical struggles,including ideological strugglesin which arguments about the implications of liberal principles played some role, though not necessarily a decisive one. But froma philosophical perspective, it is importantto understand where principles lead, even if one does not assume that people's actions in the world will always be governed by the principles they espouse. 29Compare Walzer's claim that the caste systemwould be just if accepted by the villages affected(ibid., pp. 313-15). 30lbid.,pp. 39-40. 31Ibid.,p. 38.