1 International law distinguishes the existence of states from their recognition. As one international legal text puts it:

A State comes into existence when the Community involved acquires the basic characteristics associated with the concept of a State: a defined territory, an operating and effective government, and independence from outside control. … [However, a] State may exist for a long time, functioning as an operative political Community, without being a ‘state’ in the legal sense. Transformation into the legal personality known as a State comes about only through the admission of the entity in question as a recognized member of the Community of nations. Numerous communities possessing territory, Citizens, a government, and all other appurtenances commonly associated with the concept of a State were not ‘persons’ in international law until admitted to the Community.

Thus, in one sense, a State just is an entity that possesses certain political characteristics — like territory, a government, and independence — which make it eligible to receive recognition. But it does not become a State in the legal sense unless the international Community decides to recognize it as a fellow member, with certain rights and powers under international law. Following international law texts, I will call the former political entities (or simply, entities), to distinguish them from recognized states. (See Glahn, Gerhard Von Law Among Nations: An Introduction to Public International Law, 6th ed. (New York: Macmillan 1992): 87Google Scholar, 54.)

The role of a normative theory of recognition is to offer criteria for determining which political entities the international Community ought to recognize as states. For example, a theory might claim that all political entities, regardless of their moral quality, should be granted recognition automatically, or that they should be granted recognition if and only if they meet minimal requirements of justice, or that they should be granted recognition if and only if doing so would be the best means of achieving global peace and justice. Other positions are available as well, but my concern here is with the latter two.

2 Von Glahn, 125. Unfortunately, the term ‘recognition’ seems to suggest that to recognize a political entity as a State is merely to acknowledge that it possesses the characteristics of statehood. But in fact, to recognize an entity as a State involves much more than that: it confers legal personality upon that entity, with certain rights and powers under international law. It is ‘a political act with legal consequences’ — indeed, highly significant ones, like conf erring (or acquiring) the right to territorial integrity, the right to noninterference in their internal affairs, the power to make treaties, and the right (subject to certain restrictions) to enforce legal rules on those within their territory (Von Glahn, 85). A decision to grant or withhold recognition is therefore a highly important moral decision.

3 Technically, the granting of recognition is performed by individual states, not by the international Community as a whole (Von Glahn, 89). But here I will speak of the international Community as the agent that grants (or ought to grant) recognition, since a normative theory of recognition is a theory of the conditions under which any State (and hence all states) should recognize another entity as a fellow member of the international Community, with certain rights and powers under international law. Similarly, the granting of recognition by any State, or even by all states, technically does not commit any international body (like the United Nations) to admit the recognized State into its membership (ibid.). But here I will assume that, if the international Community should recognize another entity as a fellow member, with the rights and powers of statehood, then it should also be a member State in international organizations like the U.N. (Later, in section IV, I relax this assumption to see how it affects the argument.)

4 Besides distinguishing the existence of states from their recognition, international law also distinguishes the recognition of states from the recognition of governments:

Recognition of a government is a political act differing from the recognition of a State only in the nature of the entity being recognized. … What is involved is the authority of a new group or a new person to act as the governing agency of a State and to represent it (to act as its agent) in its international relations.

By recognizing a state's government, then, members of the international Community confer upon it the right to exercise the rights and powers of statehood for that State. They also indicate that they will enter into normal international relations with it, exchange diplomats, and hold it accountable for that state's international obligations (Von Glahn, 94, 99).

5 The distinction I have in mind in the examples below is roughly analogous to the one Aristotle draws at 1097b25-1098a20 of the Nicomachean Ethics, where he distinguishes the good or virtue of someone who performs a particular function (like that of a carpenter, leatherworker, or harpist) from the good or virtue of a human being. The main difference is that I am concerned with one's minimal goodness (or virtue) at a particular function, or as a human being.

6 Notice that even if we hold different ideals and decide to set the minimum where they overlap, our individual judgments will still be reached by comparison with some ideal or other —just different ones. Each one of us will have decided to draw the line (so to speak) where the overlapping consensus exists. Since the minimum is still determined by its distance from an ideal — which is the basic point — it will simplify the exposition if we continue our discussion under the assumption that we share the same ideal, but keep this qualification in mind.

7 They would be disagreeing, that is, over whether possession of certain qualities (in certain degrees) is morally sufficient for being minimally decent — which is a Substantive moral disagreement.

8 See note 6.

9 Ronald Dworkin makes a similar point in a different context. In a recent article, Dworkin argues that if security concerns require us to treat legal immigrants unfairly, then we should at least be candid about that and admit that our treatment of them is unfair (yet required for security reasons) instead of insisting that such treatment meets lower Standards of fairness. See Dworkin, Ronald ‘The Threat to Patriotism,’ New York Review of Books 28 (February 2002), 44–9.Google Scholar In the next section, I will have more to say about the content of the pragmatic aims mentioned.

10 Buchanan, Allen ‘Recognitional Legitimacy and the State System,’ Philosophy & Public Affairs 28, 1 (1999), 51.CrossRefGoogle Scholar In his latest work on this topic, Justice, Legitimacy, and Self-Determination: Moral Foundations for International Law (New York: Oxford University Press 2004) — released just as this article was going to press — Buchanan develops more fully the argument of his earlier article. I discuss this work in notes 16 and 40.

11 For two philosophers who accept the view that accounts of recognition belong to nonideal theory but who otherwise disagree over which one is correct, see my ‘Recognition and Legitimacy: A Reply to Buchanan,’ Philosophy & Public Affairs 28, 3 (1999), 257; and Buchanan, Allen ‘Rule-governed Institutions versus Act-Consequentialism: A Rejoinder to Naticchia,’ Philosophy & Public Affairs 28, 3 (1999), 259.CrossRefGoogle Scholar

12 As we shall see, there are two different versions of the justice-based account that accept these claims. There are strong versions, according to which failing to meet minimal requirements of justice automatically defeats a claim to state-recognition. (See, for example, Buchanan, Allen ‘Recognitional Legitimacy and the State System’ Philosophy & Public Affairs 28, 1 (1999) 46–78.CrossRefGoogle Scholar) There are also weaker versions, according to which failing to meet minimal requirements of justice merely creates a presumption against state-recognition that can be defeated. My focus here is on both kinds of justice-based theory, and in arguing that we should reject them in favor of the pragmatic account, I shall be embracing these same four claims.

13 This substantially develops and extends the reasoning first sketched in my ‘Recognition and Legitimacy: A Reply to Buchanan,’ Philosophy & Public Affairs 28, 3 (1999) 242-57.

14 Notice that this argument is perfectly general. It does not rest on the assumption (used for purposes of Illustration) that the best theory of justice is a noncomprehensive liberalism. It can also be easily extended to cover Rawlsian ‘overlapping consensus’-type defenses of minimal justice. If an overlapping consensus among reasonable conceptions of justice establishes x as minimal justice, but Standard y (lower than x) is needed to bring enough entities into the processes for formulating, adjudicating, and implementing international law, we reach the same result.

15 Another way of putting this is that its notion of minimal justice carries no independent weight. It can be lowered to whatever level is necessary to bring enough entities into the processes for formulating, adjudicating, and implementing international law. Its level, in other words, is merely a consequence of a decision we make on pragmatic grounds. Surely, though, it would be more straightf orward simply to admit this fact, as the pragmatic account does.

16 Depending on how broadly we construe it, then, minimal democracy could include, say, an Afghan-style loyajirga. The strong Standard for state-recognition is defended by Allen Buchanan in ‘Recognitional Legitimacy and the State System,’ 53; and ‘Justice, Legitimacy, and Human Rights,’ in The Idea of a Political Liberalism: Essays on Rawls, Davion, Victoria and Wolf, Clark eds. (Totowa, NJ: Rowman & Littlefield 2000) 73–89.Google Scholar As I explain in note 40, however, Buchanan seems to lower his Standard for state-recognition in Justice, Legitimacy, and Self-Determination.

17 Here I rely primarily on the U.S. State Department (DOS) Country Reports on Human Rights Practices, which, while having certain limitations, are not as limiting as United Nations pronouncements or Nongovernmental Organization (NGO) reports, the main alternatives. U.N. pronouncements are limiting because the U.N. only reviews the human rights practices of a few countries each year, whereas the DOS reports cover all countries annually. In addition, the U.N. is notoriously reticent to condemn countries for human rights violations and its language is painstakingly meek. When it does condemn them in Security Council (S/RES), General Assembly (A/RES), or Commission on Human Rights (UNCHR) resolutions, we may rely on these official pronouncements, but we should understand that there will probably be other countries that violate the strong (or weak) Standard as well. NGO reports are a bit better, but because the DOS reports rely on multiple sources (such as domestic and international reports as well as reports from individuals, academics, and embassy employees), they are more likely to be complete than a report drafted by an independent organization using its own sources and contacts. Further, the State Department has access to intelligence information, which gives it an independent way of verifying accusations that a country violates human rights (accusations often raised by NGOs themselves). The main drawback to relying on the DOS reports, of course, is that they do not address whether human rights are adequately honored in the United States. Consider this Statement from the U.N. Human Rights Committee, Subcommittee on the International Covenant on Civil and Political Rights (ICCPR): ‘The Committee… regrets the easy availability of firearms to the public and the f act that federal and State legislation is not stringent enough in that connection to secure the protection and enjoyment of the right to life and security of the individual guaranteed under the Covenant’ (Concluding Observations of the Human Rights Committee: United States of America. 03/10/95. CCPR/C/79/Add. 50; A/50/40, para. 282). (Although this Statement was issued in 1995, I doubt that circumstances have changed enough since then to render it inapplicable.) Using this Statement as authority, we would have to conclude that the U.S. violates the weak Standard — something the DOS reports could hardly acknowledge. Still, for the purpose of assessing the justice-based account, I shall stipulate that the United States would now receive recognition under the strong Standard, since that is the Interpretation most favorable to establishing the feasibility of the strong Standard elsewhere and worldwide (and hence the Interpretation most favorable to the justice-based account).

18 See Table 1.

19 See Table 2.

20 In response to this argument, someone perhaps might want to press the distinction between recognizing states and recognizing governments. That is to say, one might reply that my objections concern a too stringent Standard for recognizing governments, not states. I will examine this reply in detail in section IV.

21 The basic idea, according to the justice-based account, is that if we recognize internally unjust entities, we confer upon them the right to non interference in their internal affairs and thus support their ability to continue wielding unjust coercive power within their territories — and this, claims the justice-based account, makes us an accomplice in injustice. I will examine this claim in detail in section V. For now, we need only note that, for the purposes of this section, we are attempting to characterize a ‘best version’ of the justice-based account that respects this commitment. So, for present purposes, references to conduct that makes us an accomplice in injustice should be understood as referring to what the justice-based account claims or implies.

22 S/RES/661 (1990).

23 In addition, allied coalition members Egypt, Saudi Arabia, and Syria, which fail the strong Standard, would have been ineligible to implement S/RES/661.

24 ‘Voting was restricted to male Citizens over 21 — about 3.5 percent of the population’ prior to the war, so Kuwait failed the strong Standard back then too. See Middle East Report (Nov./Dec. 1990), 14.

25 Middle East Report (Nov./Dec. 1990), 13

26 ‘Invading Iraqis Seize Kuwait and Its Oil; U.S. Condemns Attack, Urges United Action,’ The New York Times (August 3, 1990), AI (citing government analyst).

27 Anthony Cordesman, ‘Symposium on Dual Containment: U.S. Policy Toward Iran and Iraq,’ Middle East Policy (Fall 1994) 11-13

28 Hassan Hamdan al-Alkim, ‘The Gulf Subregion in the Twenty-First Century: U.S. Involvement and Sources of Instability,’ American Studies International (February 2000) 72-95

29 Middle East Report (Nov./Dec. 1990), 13. Notice that one may consistently accept the points made in this paragraph whatever one's positions are regarding (i) whether sanctions should have been given more time to work before the war began; (ii) whether the postwar sanctions were justified; (iii) whether the particular way the war was conducted (like the ‘turkey shoot’ on the ‘highway of death’ at war's end) was justified; (iv) whether members of the allied coalition overconsume oil.

30 Shlomo Gazit, ‘Israel and the Palestinians: 50 Years of War and Turning Points’ Annais of the American Academy of Political & Social Science (January 1998) 82-97

31 Egypt failed the strong Standard in 1978-79 due to a State of emergency that had been in force since 1967, which curtailed rights to liberty. Israel also failed it in 1978-79 due to discriminatory laws against Arabs. See report by U.N. Human Rights Committee, Subcommittee on the International Covenant on Civil and Political Rights, Concluding Observations of the Human Rights Committee: Israel, 18/08/98. CCPR/C/79/Add. 93 (covering the period 1978-1998).

32 Stephen Zunes, ‘The Strategie Functions of U.S. Aid to Israel,’ Middle East Policy (October 1996) 90-1

33 Stephen Zunes, ‘The Israeli-Jordanian Agreement: Peace or Fax Americana?’ Middle East Policy (April 1995) 57

34 Shibley Telhami, ‘From Camp David to Wye: Changing Assumptions in Arab-Israeli Negotiations,’ Middle East Journal (Summer 1999) 380-2

35 See ibid.; also see A. Safty, ‘Sadat's Negotiations with the United States and Israel: Camp David and Blair House,’ American Journal of Economics & Sociology (October 1991) 473-85.

36 See Table 3.

37 See Table 4.

38 The U.N. Peacekeeping Operations Reports by the Secretary General, May 2004, State that there were 55, 470 total U.N. peacekeepers, of whom 14, 020 were African. Of the contributing African countries, only Djibouti, Namibia, Niger, and South Africa pass the weak Standard. Together, they supply 2, 574 peacekeepers, roughly 18.4% of African peacekeepers and 4.6% of the total.

39 Only one (Namibia) passed the weak Standard. See reprint of report by S. Mullen and J. Woods, in ‘Special Report: Peacekeeping in Africa,’ United States Institute of Peace (February 13, 2001), 5

40 It might be objected that the theory is only meant to apply to future cases of recognition, not to past cases. But it is important to see that this position is incompatible with anything that deserves to be considered justice-based, since it allows scores of countries which have rarely, perhaps never, met its minimum Standards — many of them massively unjust — to enjoy the rights and powers of statehood, and an entitlement to participate in the processes for formulating, adjudicating, and implementing international law, while denying those same rights and entitlements to entities that may actually be morally better.

Similarly, it is important to see how the argument of this section actually predicts (see note 10 above) the way in which Buchanan's argument evolves in his latest work on this topic, Justice, Legitimacy, and Self-Determination. In four critical respects, he lowers his Standard for state-recognition from the strong Standard, which he embraced in his earlier work (see note 16 above). First, he Claims repeatedly that, in Order to merit recognition, states need only make credible efforts to satisfy minimal Standards of justice (see 70, 187, 234, 247, 257, 260, 331, 336, 375, 434). But making credible efforts to satisfy minimal Standards of justice falls short of actually satisfying those Standards. Second, he permits states to be recognized even if they fail to be minimally democratic — which he regards as a minimum Standard of justice — and even if minimally democratic institutions are not feasible in the short run, provided that such institutions could be developed in the long run with the benefits of recognition (235, 254, 257, 259). Third, he permits states to be recognized even if their agent, the government, violates minimal Standards of justice, provided that their constitution guarantees these Standards (pp. 354-5). But despite the fact that we often speak of constitutional ‘guarantees,’ the constitution itself does not guarantee that these Standards are satisfied. The constitution is just a piece of paper. It would be more accurate to say that the constitution promises to satisfy these Standards. But promising to satisfy minimal Standards of justice falls short of actually satisfying them. Fourth, and most dramatically, he insists that his theory is only meant to apply to future cases of recognition, not to past cases (273-8), since it would result in withdrawing recognition from too many existing states. But this is problematic for reasons explained in the paragraph above. All of these moves appear to be motivated by the need to bring enough entities into the processes for formulating, adjudicating, and implementing international law. The result is not a plausible Standard of minimal justice but rather an acknowledgment that, if we are to make moral progress, we must (unfortunately) recognize many entities that fail to satisfy minimal Standards of justice (277) — exactly as the pragmatic account admits.

41 See the first paragraph of this article for this characterization of global justice.

42 By ‘too unjust,’ I mean a Standard that falls farther below the Standard for minimal justice.

43 Notice that this argument applies no matter how we identify what counts as being too unjust. Whether we identify it with a failure to protect certain rights and liberties, or with a failure to garner a certain level of popular support (however we might determine that in the absence of democracy), if too few entities qualify under that Standard, we must lower it further.

44 United Nations Development Program, Human Development Report (2002) 10

45 United Nations Development Program, Human Development Report (2002) 151-2

46 United Nations Development Program, Human Development Report (2002) 7, 8, 113

47 Presumably, the justice-based account is committed to something like the following. A political entity that meets minimum requirements for a sufficient period of time qualifies for state-recognition. But then, brief lapses in human rights protection do not automatically trigger derecognition of the State. Instead, they first trigger derecognition of the government as a step in that direction. Persistent failure to protect basic human rights, however, requires derecognition of the State. I maintain that this last condition is satisfied.

48 This is because, by hypothesis, the relations that best promote peace and justice are the ones that they would have if they were recognized, as the pragmatic view (by hypothesis) would recommend in these cases.

49 Where to draw the line is itself a Strategie decision resting on a balance of two considerations: (1) the extent to which conferring any benefits of recognition at all on unjust entities weakens incentives for moral progress (since an entity may be satisfied with those benefits and interested in no more); and (2) the extent to which conferring further benefits of recognition strengthens those incentives. I assume that the more unjust an entity is, the less interested it is in acquiring further benefits, and so at some point granting recognition to entities that are too unjust does not advance but hinders the pursuit of global peace and justice. For further discussion of how to balance these incentives, see my ‘Recognition and Legitimacy: A Reply to Buchanan’ 252-3.

50 There are other kinds of deontological view with which the pragmatic account is consistent. For instance, it is consistent with deontological views that require us to protect basic interests without requiring us to maximize value (a policy of recognizing some unjust entities may best protect basic interests even if it doesn't maximize value). Depending on how we are to frame and universalize maxims in (extremely) nonideal circumstances, it may even be consistent with deontological views that reach moral judgments by testing maxims. The only type of theory it is inconsistent with is the (in my view, oversimplified) deontological type which Claims that some acts are wrong no matter what the consequences — and then only on the questionbegging assumption that recognizing an unjust entity is one of those acts. Of course, which of these views (if any) the pragmatic account should accept depends on which of them is independently best justified. Nothing said here implies that the pragmatic account must accept the strongest (oversimplified) deontological view if indeed it is implausible.

51 See, for example, Allen Buchanan, ‘Rule-governed Institutions versus Act-Consequentialism: A Rejoinder to Naticchia,’ 264-7.

52 I am assuming that the pragmatic account need not use a case-by-case approach to granting recognition, and that it can and should use rules for that purpose if there are good pragmatic grounds for doing so (such as that decisions granting recognition would likely be more predictable, impartial, and easily reached).

53 Again, see the first paragraph of this article for this characterization of global justice.

54 It would be mistaken to infer from this that the pragmatic account admits no moral constraints at all on the pursuit of peace and justice. Despite the fact that it regards peace and justice as goals to be pursued, it does not follow that any means whatsoever may be used to pursue them — that ‘the ends justify the means.’ Suppose, for instance, that the best theory of justice includes nonconsequentialist restrictions (or even rule-consequentialist ones) on the pursuit of the ends it takes to be valuable. In that case, there are such moral constraints — though they would apply to acts other than acts of granting recognition.

55 See ‘Blood Spills to Keep Oil Wealth Flowing,’ Los Angeles Times (Sept. 15, 2002), AI; and ‘Cutoff of Aid a Sign to Bogota,’ Los Angeles Times (Jan. 14, 2003), A1.

56 This is clearest in the case of consequentialist theories. But it is also consistent with many types of deontological view, for reasons explained earlier (see note 50).

57 Rawls, John Political Liberalism (New York: Columbia University Press 1993), 9Google Scholar

58 O'Neill, Onora ‘Political Liberalism and Public Reason: A Critical Notice of John Rawls, Political Liberalism,’ The Philosophical Review 106, 3 (1997) 411–28,Google Scholar at 416

59 To be sure, there is a slight difference in the two cases. In the domestic case, it is the politically unrealistic attempt to achieve the Standard of domestic justice itself that is claimed to ground the alteration of the Standard. In the international case, by contrast, it is the politically unrealistic attempt to achieve, not the Standard of minimal justice itself, but other Standards — global peace and justice — that is claimed to ground the alteration of the Standard of minimal justice. Nonetheless, they share the central defect of compromising our Standards of justice or minimal justice so as not to make some goal politically unrealistic. Even if we consider the amended justice-based account, where recognition does not depend on minimal justice but where f acts about injustice still matter, the way in which such f acts matter still rests solely on what is politically practical (as we saw in section IV). Such facts matter only insofar as they affect political strategy and do not carry the kind of independent weight that we would expect facts about injustice to possess.

60 See, for example, Buchanan, ‘Recognitional Legitimacy and the State System,’ 55-6, where he suggests that a normative theory of recognition should not be ‘too 'utopian,’ lest it not be taken seriously or provide a useful guide for practice. He repeats this in justice, Legitimacy, and Self-Determination, 268.

61 Alternatively, with the amended justice-based account, the bad news is that facts about injustice matter only insofar as they affect political strategy. They don't carry any deeper normative significance.

62 Many people were involved in helping me prepare this article, and here I would like to acknowledge my deep debts of gratitude to them. For their comments on a prior version of this paper, I am much indebted to Tony Roy, Tom Moody, members of the Law and Philosophy Discussion Group of Southern California, and especially — for his generous and valuable guidance — David Miller. For research assistance, I am indebted to Donovan Rinker-Morris and Michael Stieg. I am also grateful for my many discussions with Rinker-Morris, whose ideas, knowledge, and suggestions made the writing of section III — a critical part of the paper — possible. Of course, the views expressed in this paper, and any errors made in defending them, remain mine alone.