Comments on National Self-Determination

In this article, the authors make a case for civic and cosmopolitan nationalism. They try to reconcile the perceived conflict between nationalism and cosmopolitanism and present a new understanding of nationalism. Do you believe the author's arguments are successful?

The Principle of Nationality

In Nations and Nationalism, Ernest Gellner says that “nationalism is a theory of political legitimacy which requires that ethnic boundaries should not cut across political ones.” Gellner’s characterization of nationalism should be understood as deriving from two premises: first, a definition of nation in terms of ethnicity; and second, the principle that national and political boundaries ought to coincide. Call the latter the principle of nationality. The two premises are of course independent.

Let's focus for now on the principle of nationality. In his account of nationality, David Miller states the principle as follows: “it is valuable for the boundaries of political units (paradigmatically, states) to coincide with national boundaries” (82). And he distinguishes two elements of the principle, and, correspondingly, two distinct strands of argument in its favor:

  1. It is “valuable” for nations to have states (political units)

  2. It is valuable for states to compromise a single nation (valuable to comprise one, and valuable to comprise no more than one).

Now Miller’s official statement of the principle of nationality (which I have quoted) is very weak, inasmuch as it says nothing about the importance of the value of having political and national units coincide. But elsewhere Miller suggests stronger formulations: for example that national communities living in a territory have a “good claim [emphasis added] to political self-determination; there ought to be put in place an institutional structure that enables them to decide collectively matters that concern primarily their own community” (11). So there is a right to national self-determination: not an absolute right, but a “good claim.”

Presumably, too, states have the authority to do what is required to build a nation. That is, presumably we should interpret the principle of nationality as claiming that the value of there being a single nation within a state is sufficiently great that states have the authority to foster a common national identity through law and public policy (in education, or regional development, or language policy, or policies on coinage). Of course some means – like expulsion or slaughter of national minorities to create homogeneity, or regulations on speech critical of the nation’s greatness – may be illegitimate, but the principle of nationality, as I am supposing Miller to interpret it, is more than an observation about what would be an attractive state of affairs. It is part of a normative account of the legitimate authority of the state.

Comments on the Principle

I want to explore the argument that states are helped by having nations later on. But before getting to it I want to make two background points.

First, the principle of nationality is a principle about political jurisdictions:

whether the political unit in question is a sovereign state or a subnational unit in a federal system, the point of the principle is to make the case for a correspondence between jurisdictional and national boundaries. It is thus a stronger thesis – in any case a different one – than we find in Yael Tamir’s defense of national self-determination in her account of liberal nationalism. Tamir associates the idea of national self-determination with a group’s managing its communal life in accordance with its own outlook, which may not mean group authority in a territory (since the group may not be territorially bounded). Though the principle of nationality under consideration here does not require an independent state, it is about rule in a territory.

Second it is important to distinguish Gellner’s ethnic account of nationality from the view endorsed I believe by Miller and common in the literature on nationalism. Thus it is common to distinguish at least two types of nationalism: civic and ethnic (see Brubaker, Greenfield, Snyder). In the case of ethnic nationalism, allegiance is to a group, which is assumed to have common characteristics – culture, language, religion, historical experience, or kinship. Possession of the traits determines membership in the nation. I the case of civic nationalism, loyalty is to a set of principles or institutions, and membership in the nation is a matter of, in the first instance, living in a territory governed by institutions that express the principles: thus Lincoln said about the United States that it is conceived in an idea and dedicated to a proposition. Here the nation is a demos not an ethnos: we may have awareness in a group of shared principles and institutions, but without shared language, culture, kinship, or religion.

Miller’s interpretation of the nationality principle, unlike Gellner’s, is intended to allow for both kinds of nationalism: to allow for multi-ethnic nations (see 21). A nation is a group of people who share a sense of common membership (each believes that he of she belongs, believes that others share that belief, believes that others are aware that the belief is shared, etc.); believe that they are entitled to a territory; are capable of acting as a group; have a common history; and a shared public culture. Now the requirement of shared public culture may seem to bring nations close to Gellner’s account. But a public culture need not be comprehensive: indeed, it may not extend much beyond political principles and social norms (26), and may thus comprise a variety of private cultures.

Three Observations

Three other brief observations about both kinds of nationalism are pertinent to what I will be saying later.

First, national identity of both the civic and ethnic kind may be more or less important. This observation raises especially important issues about the argument from nation to state, but I willing not be considering that argument here.

Second, both kinds of nationalism may be more or less determinate: thus the implications of the principles that define a civic nation may be uncertain (as Stanley Fish often points out), but the same is true of ethnically-defined national identities.

Finally, third, though national identity is distinct from institutional membership, identities of both kinds may be created by institutions rather than antecedent to them.

States and Nations: Instrumental Argument

Miller, I said, distinguishes two arguments in support of the nationality principle. The first argument is that nations are entitled to states because states enable members of nations to fulfill their obligations, help nations to preserve their cultures, and enable nations to achieve collective self-determination. I will not be focusing here on this first line of argument.

The second argument is that (what I will refer to for convenience as) the “successful conduct of politics” depends on a common national identity. More precisely, the provision of public goods, the achievement of distributive justice, and the operation of deliberative democracy all are promoted by national identity: that is, when the state coincides with a single nation, and is not multinational (though it may be multiethnic or multicultural). Call this the instrumental argument for the coincidence of state and nation: Beitz suggests it, when he says that the best case for national self-determination is that a correspondence of national and political boundaries may be an aid in the cause of justice. [1]

And in his response to Nussbaum, Charles Taylor urges that achieving the justice embraced by cosmopolitans appears to require national attachments, because justice requires democracy and therefore some limits of equality. And both democracy and equality require “a high degree of mutual commitment[s]” to compatriots (Taylor in Nussbaum, 120).

Cosmopolitan Nationalism

As the reference to Beitz indicates, and as Taylor suggests in his response to Nussbaum, this instrumental argument may be found attractive even by moral cosmopolitans, at least if we think of moral cosmopolitans as essentially being moral universalists, whose first principle is that the well-being of each person in the world matters equally. (Nussbaum’s cosmopolitanism seems ambiguous as between an endorsement of moral universalism and an endorsement of a personal ideal of thinking of one’s first allegiance as to all persons everywhere, and taking that thought as a guide to conduct). Starting from that premise, the cosmopolitan may think that it is good for the achievement of justice that there be separate states that aim to protect the basic rights and interests of members (or at least that that is an acceptable status quo). But then the moral cosmopolitan may also think that if the best way to achieve justice is through a world of separate states in which there are strong national allegiances.

I describe this case for national identity as instrumental, because of the nature of the justification: national identity is here seen as a means to the achievement of independently cognizable goods (common good, justice, and democracy). But while the theorist may see the justification this way, and nationalizing political elites who are aiming to foster a shared national identity, may treat it this way, it seems plausible that they must present it some other way in order to encourage it: to promote national identity successfully, they must encourage individuals to see their shared allegiance to a group or to a set of principles as something other than a means for achieving independent goals.

Nationalism and Political Success

In any case, Miller claims that the linchpin in the instrumental argument – binding shared national identity to political success – is trust. The idea is that success on the three dimensions mentioned earlier (public goods, justice, democracy) is fostered by high levels of trust – roughly, confidence that others will do their part in the political society, and will not free ride. Moreover, trust in turn is fostered by the sense of loyalty (whether to principle or to group) associated with national identity: “I take it as virtually self-evident that ties of community are an important source of such trust between individuals who are not personally known to one another and who are in no position directly to monitor one another’s behavior. A shared identity carries with it a shared loyalty, and this increases confidence that others will reciprocate one’s won cooperative behavior” (92). In the case of public goods, for example, people are more likely to do their share in supporting the provision of such goods, and in complying with rules designed to provide them (paying taxes, keeping the parks clean, getting rid of the SUV), if they have trust that others will do so as well. And they are more likely to go in for sincere and reasoned argument that deliberative democracy requires if they can trust others to do the same.

I have four points to make about this argument.

First, in explaining the link between national loyalty and political success Miller vacillates between emphasizing trust – confident expectations about the compliance of others – and emphasizing group solidarity – commitment to ensuring the success of others out of a sense of common group membership. The difference between trust and solidarity seems clear, as a matter of common sense: I trust Norwegians, but have no particular sense of solidarity with them: trust is epistemic, and concerns my beliefs about what others can be expected to do; solidarity is about motivations. The story about public goods is essentially about trust, whereas the story about distributive justice concerns solidarity (Miller shifts from “trust” to “solidarity” in a single paragraph [see 93] but seems not to notice the shift). The claim with respect to distributive justice is that the willingness to forgo advantages that one could win in less regulated markets in order to ensure a decent life for others who do not succeed in the market depends on having a sense of being in community with those others: a sense of obligation to them as members of one’s own nation.

The idea that distributive justice depends on a sense of solidarity and loyalty is an argument advanced by social democrats who are opposed to the European Union, precisely because they see it as undermining the national solidarities required for distributive justice. It is also the crux of Michael Sandel’s critique of Rawls: the idea that there is a tension between Rawls’s liberalism and his egalitarianism. Thus Rawlsian egalitarianism is said to require that we reject the idea of the priority of “plurality over unity,” and think of community allegiance as fundamental and constitutive, though – Sandel argues – such rejection stands in tension with the idea of antecedent individuation which is the basis of the liberal idea of the priority of individual rights.

I will return to the issue of solidarity later. Suffice to say now that with this clarification, we have two basic ideas in play: first, that political success depends on trust and solidarity, and second that trust and solidarity depend on shared national identity.

Before exploring those ideas, I make the second observation, which is that national solidarity, especially of the ethnic-nationalist kind, can be in deep tension with justice, democracy, and the broad provision of public goods, because ethnic nationalism is essentially exclusive: loyalty is to the group, and if the group is defined as essentially unjoinable (as in the traditional conception of the German nation and German citizenship, as defined by blood ties), then the loyalties are limited in scope and may not extend to everyone in the territory and complying with the institutions. So political success – defined as Miller defines it – may not be helped by some forms of national allegiance. Whether it is or not will depend on both the content of the allegiance, how determinate it is, and its strength. A strong form of ethnic nationalism, that does not leave much room for interpretation on its claims about membership in the nation, will raise troubles for political success.

It seems plausible that political success depends on trust – mutual confidence that others will do their part. But the force and significance of the idea that political success depends on trust is uncertain, because it is not clear whether trust needs to be prior and historically generated – with deep roots lost in the mists of time, as the national identity theory suggests, and as Robert Putnam has argued with his account of the emergence of trust in Northern Italy from 600 years of associational life – or is relatively malleable and capable of construction. The idea that trust is deeply dependent on and fostered by prior shared loyalties – the idea that if you want it, you should “get a history” – faces three difficulties: first, there is a vast literature on how trust emerges from efforts to cooperate on common projects among people without prior histories of shared loyalty, whether the loyalties are defined in terms of an ethnos or in terms of principles. Though the precise conditions for the emergence of trust are uncertain, there is some evidence for the construction of trust from joint projects. So, if for example, we agree that political success depends on trust, and are interested in political success for the European Union, we should not aim to promote a European political identity, but to encourage common projects that promise practical success.

Second, while loyalties may help trust, they can make it hard to sustain trust because violations may be experienced as profound betrayals – betrayals of the group – and thus be very hard to repair: this could be a particularly severe problem for a highly determinate collective, national identity. Weaker ties may, on balance, make for more stable trust.

Third, people can have reputations for being trustworthy even with those with whom they have no history of cooperation. Thus, I trust Norwegians and Swedes, for example. Even if the reputation for being trustworthy is founded in the first instance on an aspect of national identity, the reputation itself may extend to settings in which cooperation grows outside the circle of compatriots.

Finally, it not clear why distributive justice depends on national solidarities: why a principled commitment to fairness will not do. Recall again the distinction between civic and ethnic nationalism. In the case of civic nationalism, the allegiance that binds the political society is to a set of principles. Suppose then that we can have a political society with a civic form of nationalism in which the content of the civic nationalism is given by principles of justice that require some form of egalitarian distribution. If that is possible, then suppose that we have a multinational state, and that each of the nations shares an allegiance to principles that require fair distribution of resources: the principles are different, and by all measures the groups do form different nations. Nevertheless, the content of the loyalties that define them require them to ensure a relatively egalitarian distribution in the multinational society.

So if an egalitarian civic nation is possible, it is not clear why an egalitarian multinational state is not possible. But, then, you might reject the antecedent. And the fact that Miller shifts, as I noted earlier, from trust to solidarity in the course of his discussion of distributive justice suggests that Miller might have trouble with the antecedent. That is, he might say (in effect with Sandel) that a political society will not embrace an egalitarian view of distributive justice unless it is founded on a richer, more substantial group bond than is characteristic of civic nationalism – not simply a unity founded on shared commitment to principles, but a unity that sustains the thought “one of us should not be left to live that way.” You need group loyalty, not simply shared loyalty to principles. So thinner, civic nationalisms will all be more liberal and less egalitarian in content than ethno-cultural nationalisms. Even if, as Rawlsians suppose, a compelling argument is available within the framework of liberal principles for such an egalitarian view of justice, a more form civic form of nationalism will lack the motivational resources – the “high degree of mutual commitment” – required to sustain limits on inequality.

Maybe this is right: because principles are motivationally weak, civic nationalism must be thin in content, and therefore multinational states are limited in the justice they can achieve. But two points about this: first, it would be worth considering whether there really is a correlation between forms of nationalism and content of views of justice on the dimension under consideration here (views about equality). And second, the conclusion should not be that states need nations, but that multinational states are in no more trouble with respect to justice than civically national states.


[1] For criticism, see Onora O’Neill, Bounds of Justice, pp. 179-181.

Source: Joshua Cohen, Thomas Scanlon, and Amartya Sen,
Creative Commons License This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 License.

Last modified: Friday, September 10, 2021, 12:03 PM