Can discrimination between nationals and non-nationals be justified?

Ways to think about multiculturalism when it goes off the rails

In brief, this piece is an engagement with the ideas of John Finnis, who argues (among many other things) that discrimination on the basis of both religion and nationality—but not race—is justified.

Finnis’ justification is not, however, one that has its origins in a form of political liberalism. Rather, Finnis is a special kind of political perfectionist: his philosophy is grounded in the ideas of Thomas Aquinas and, ultimately, Aristotle.

This strain of thinking holds that governments are instituted among men to make men good (and to pursue the good). They are not a ‘community’ as such, or a social contract intended to avert the depredations of man in the state of nature.

All that is by way of introduction. I’m at most a weak perfectionist—my only substantive non-neutral values are ‘framework’ ones, like free markets, personal autonomy, and various forms of negative liberty. If I’m to give credence to Finnis’ thought, I can’t do so only on his terms.

This is why I’ve drawn on arguments from one of Finnis’s intellectual rivals, another legal scholar by the name of Joseph Raz. I don’t agree with him either, but thanks to the pair of them (both were my tutors at Oxford many years ago), I’ve been forced to think my thoughts through to the end.

---

Liberalism is also a fighting creed 
— Charles Taylor

In a sense, this question is misconceived. Discrimination against non-nationals in most western states is a truism, occurring most commonly in the form of restrictions on the franchise and employment. New arrivals must wait for some time—or take out citizenship—to vote or stand for parliament. The US President must be a ‘natural born’ citizen, thus excluding from the nation’s highest office not only non-citizens, but also citizens with a particular characteristic. Arnold Schwarzenegger’s many fans probably stand to lose most from this arrangement.

I therefore take ‘discrimination’ and ‘non-nationals’ to be narrower terms, which Finnis uses in an immigration and refugee policy context as part of a series of important papers on the ability of western liberal states to accommodate large numbers of Muslim immigrants. Finnis challenges two common liberal pieties. First, he argues for a discriminatory immigration policy that halts certain would-be residents at the border and repatriates others. Second, he criticises various forms of the policy—common in many western liberal states—of multiculturalism.

Finnis correctly notes that public deliberation on these issues is increasingly difficult, so I wish to state at the outset that I take his arguments—particularly about illiberal aspects of the Islamic faith—seriously. However, I take them seriously as a classical liberal. That is, as a species of liberal far fiercer than the liberal who—in Robert Frost’s memorable phrase—couldn’t take his own side in an argument. Classical liberals sometimes observe that the First and Second Amendments are equally representative of ‘liberalism’. Even in a US context, people tend to forget the second, to forget that liberalism can also fight.

Justification and necessity

I think it’s possible for a liberal to justify discrimination in the sense Finnis describes, but it may not be necessary to do so.

First, I outline Finnis’ concerns with Islam and his arguments against multiculturalism, taking note of his debate with Joseph Raz on the latter. Next, I discuss how liberalism enjoins the defence of personal autonomy, including the use of coercion against religious (and other) traditions that undermine it. As part of this account, I discuss the distinction between weak and strong perfectionism.

I argue that what looks like anti-perfectionism—at least in the case of Hayekian classical liberalism—is actually a species of weak perfectionism, in that it presupposes certain non-neutral values such as individual liberty, the primacy of free markets, and personal autonomy. Finally, I suggest that Finnis’ specific policy proposals would at best form only part of a total package designed to assuage the problems he enumerates. I argue, for example, that it is possible to address some concerns by abolishing multiculturalism as a state-funded policy. This proposal is not discriminatory at all, although it may be controversial.

Islam, immigration and multiculturalism

Finnis founds his arguments for a discriminatory immigration policy in large part on his view that Islam—even in ‘moderate’ forms—is inimical to the common good. He makes this argument while pointing out that Western states themselves have moved away from endorsing many objectively valuable goods, even to the point of ‘embrac[ing] moral evils with at least as much complacency as Roman jurists preserved when contemplating slavery’.

Careful with those historical echoes, Ahmed.

It is important to realise that he does not argue that states should refuse to tolerate the intolerant. Finnis has long questioned the ‘ship of state’ analogy, with its underlying insistence that there is a ‘determinable set of life plans’ towards which the state should direct its citizens. Rather, the state’s role is to set the background conditions that allow people to live valuable lives for themselves. Finnis’ concern with Islam springs from his view that many of its core tenets actively inhibit the possibility of living a good life, for both its own adherents and others.

He finds the routine intimidation of apostates particularly distressing, along with totalising aspects of the Islamic Ummah and the use of taqiyya to undermine proper public concern. The latter involves concealing one’s true objectives until achievement of a given goal. In Refah Partisi (No. 2) v Turkey (2003)—an ECHR case Finnis considers at length—this involved an Islamist political party that professed adherence to the rule of law and multiparty democracy until it became possible to overthrow both.

Those who give their allegiance to the Ummah, Finnis argues, sometimes see their nationality as a ‘form of alienage’ and seek to undermine Western states of which they may already be nationals. This last point is particularly important in light of Finnis’ discriminatory immigration policy proposals: these would not succeed against, say, the perpetrators of the 7/7 attacks, all of whom were British born or British nationals.

Finnis also finds Islamic attempts to propagate the faith by force troubling, citing Benedict XVI’s explicit repudiation of violence in Christianity’s cause. He also raises concerns with arranged marriages and denial of women’s educational opportunities. Public attacks on apostates like Ayaan Hirsi Ali and critics like Salman Rushdie are now a commonplace of global politics. These exempla are indicative, Finnis argues, of Islam’s general tendency to intimidate opponents. Of course, critics may argue that Finnis presents a monolithic Islam, and they may be right. That does not detract from the point that a state’s ability to distinguish between individual Muslims when it administers an immigration policy is limited.

Finnis on the Channel 4 discussion programme After Dark in 1987 © 1987 Open Media Ltd

Although both Raz and Finnis argue that the state has a duty to promote valid ideas of the good, they fall on either side of contemporary debates on multiculturalism. That said, Raz’s arguments are more nuanced than he appears to admit. His support for state funded multiculturalism has origins in his view that human well-being is linked to ‘social forms’, which he describes as ‘forms of behaviour which are in fact widely practiced in […] society’. These social forms ‘pervade important dimensions of one’s life’. He sounds almost communitarian when he defines them:

I have in mind the public perception of common forms of action, each of which has the internal richness and complexity which makes it into a possible comprehensive personal goal.

Raz is careful not only to note the extent to which people enjoy ‘the good life’ in company, but also to note that morality is not in conflict with well-being as often as commonly assumed. He points out, too, that certain lives are not possible without certain social forms developing—a university lecturer, to take an obvious example, needs the presence of ‘social forms’ that value not only education, but the institutional structures that support it. However—for Raz as for Finnis—all this is ‘provided those social forms are morally sound’. When it comes to multiculturalism, Raz argues that it is a ‘normative precept motivated by concern for the dignity and well-being of all human beings’ and calls for a radical reconceptualisation of the state as ‘constituted by a plurality of cultural groups.’

Raz argues for generous state funding of multicultural education for groups beyond a certain size threshold, a toleration regime for varied cultural practices, and a recognition and acceptance that public spaces will change in the face of large-scale migration. He does this based on dignity and self-respect, the lack of which undermines peoples’ well-being. Most crucially, he thinks that states that do not endorse multiculturalism in a form akin to his proposal ‘ha[ve] no respect for their culture, find it inferior and plot its elimination’. It is on this issue that Finnis takes issue with Raz’s support for multiculturalism.

First, Finnis notes that Raz is happy to have non-viable (read small) groups excluded from the ambit of state support. For some reason, he does not construe this as cultural elimination, but the effect is identical. Apply Raz’s reasoning to the tiny number of trans people (even allowing for recent exaggerations of group size) and it becomes clear that an entire, fraught debate—had Razian policies been in place—would simply not have happened. ‘There aren’t enough of you. We don’t care.’

Next, Finnis reiterates Raz’s point that ‘the willingness to share is not purchased easily,’ and argues that multiculturalism can be as or more divisive than many other redistributive policies. He also reminds us that regarding a given culture as inferior does not mean regarding the person who adheres to that culture as inferior.

Although not sketched out in detail, Finnis also criticises Raz’s view that lack of a common language is merely a ‘nuisance’, making an argument for assimilation. He thinks that such a policy is not ‘motivated by judgments of disrespect for persons or their cultures, [nor does it] have a purpose of eliminating those cultures, even if the withering away of those cultures in this country is expected to be a side-effect of assimilation’. He argues that to hold otherwise involves ‘loosely extending the categories of “disrespect” and “offensiveness”.’

To be fair, Raz confines his support for multiculturalism to those cases where the cultures in question maintain a reasonable measure of liberality: they should abjure intolerances like homophobia and female genital mutilation and allow their members ‘rights of exit.’ They should also abandon historic racial prejudices. This is a weaker version of his earlier comments on illiberal cultures:

One particular troubling problem concerns the treatment of communities whose culture does not support autonomy. These may be immigrant communities, or indigenous peoples, or they may be religious sects. It is arguable that even the harm principle will not defend them from the ‘cultural imperialism’ of some liberal theories. Since they insist on bringing up their children in their own ways they are—in the eyes of liberals like myself—harming them. Therefore can coercion be used to break up their communities, which is the inevitable by-product of the destruction of their separate schools, etc?

If Raz’s arguments in Morality of Freedom are more representative of his ideas, then it may be that he and Finnis are closer than they appear. Even so, Raz’s mistake is the familiar one: in arguing for state funding of a particular conception of the good (multiculturalism), he forgets that it is almost impossible to ensure that the funding will lead to the creation of morally valuable opportunities, and the elimination of repugnant ones.

Liberalism and justification

Political anti-perfectionism has its origins in John Stuart Mill’s famous ‘harm principle’. That is, the only reason for governments—and individuals—to interfere with the liberty of any other person is self-protection:

The only purpose for which power can be rightly exercised over any member of a civilised community, against his will, is to prevent harm to others. His own good, either physical or moral, is not a sufficient warrant. He cannot rightfully be compelled to do or forbear because it will be better for him to do so, because it will make him happier, because, in the opinions of others, to do so would be wise, or even right.

Despite its familiarity, the principle is subtle. Mill recognised that tyranny ‘operating through the acts of public authorities’ is only one sort of oppression. He long thought that facilitating government power over the individual made it easier—legitimised, in a sense—other members of society in imposing ‘a tyranny of the prevailing opinion and feeling’. Finnis’ worries about Islamic intimidation—of apostates and those less serious about their faith—is an echo of this aspect of Mill’s harm principle.

While Raz does not go so far as to argue that ‘the legislator must labour to ensure that his citizens become good men’ (Aristotle), he thinks that the state has a duty to promote valid ideals of the good. By contrast, liberal thinkers traditionally focus on the importance of autonomy—personal, individual choice. That is, the ‘good society’ and individual happiness depends on people pursuing life goals independently selected from a range of alternatives. Inevitably, some of these alternatives may be valuable to the person making the choice but not valuable per se.

It’s highly likely that many such individual choices won’t be good in the moral sense Raz adumbrates. They will do no harm to others, however, which makes them none of the state’s business. Raz, by contrast, is explicit about defending autonomy, not merely making it available. He describes this process as follows:

Inasmuch as the liberal concern to limit coercion is a concern for the autonomy of persons, the liberal will also be anxious to secure natural and social conditions which enable individuals to develop an autonomous life. The liberal will seek […] to regulate the non-coercive effects that one person’s acts have over others in order to secure an environment suitable for autonomous life. In pursuing such goals the liberal may be willing to use coercion.

Interestingly, the above quotation amounts to a ‘weak’ or ‘thin’ perfectionism. It is quite possible to accept it without also accepting the ‘strong perfectionist’ aspect of Raz’s argument: that in addition to securing autonomy, the state should endorse, support, and provide morally valuable choices—and eliminate bad ones.

F. A. Hayek—while enjoining great caution—makes a related weak perfectionist point:

There is a need for certain common standards of values, and, although too great emphasis on this need may lead to very illiberal consequences, peaceful co-existence would be clearly impossible without any such standards. If in long-settled communities with a predominantly indigenous population, this is not likely to be a serious problem, there are instances, such as in the United States during the period of large immigration, where it may well be one.

He goes on to note that the US could never have become such an effective ‘melting pot’ without a degree of ‘Americanization’ through its education system. He also argues persuasively that much of the bitterness attending race (Hayek was writing shortly after Brown v Board of Education was handed down) arose thanks to racially separate schooling.

For a weak perfectionist (such as a classical liberal), many Islamic values are strongly antithetical to personal autonomy. This is particularly notable with respect to homosexuality, for example, and at least in some versions, to women. Some values abrogate not only Raz’s ‘rights of exit’, but also endorse a degree of political dishonesty that goes far beyond mere spin.

Of all systems of governance, liberalism (like the free market) depends on the provision of accurate information. Taqiyya virtually ensures information asymmetry. Thus, to the extent that large numbers of Muslim immigrants undermine essential elements of liberal governance, then discrimination may be justified. Of course, the discrimination need not be in the form Finnis proposes, but it could be.

Necessity, rent-seeking and empirical policy

Are Finnis’ policy proposals likely to be effective? His interpretation of UK immigration law in favour of deportation or detention of non-nationals would only solve part of the problem. The Council of Ex-Muslims of Britain has repeatedly made the point that state-funded multiculturalism offers those who want a platform unparalleled access to one.

When the state funds community bodies, it exposes itself to the problem of ‘rent seeking,’ whereby the community leaders in question respond to non-market incentives in order to collect ‘rents’. In the case of ‘Islamic community leaders’, obtaining status as an ‘official spokesman’ is like obtaining a taxi licence in pre-Uber days. It walls out competition and allows for wealth transfers from both the group the lobbying outfit purports to represent and the taxpayer.

This is, of course, a wider problem, exemplified by Michael Gove’s recent, ham-fisted attempt to define extremism in such a way that Islamic loonies can’t get their hands on government largesse. It has roots in the state handing out money to the charitable sector, forgetting that the point of charity is that it isn’t the government. No charity, of any sort, should get funding from HMGov.

It is entirely consistent with liberalism to end multiculturalism’s access to taxpayer funds, and yet it’s not discriminatory in the sense Finnis adumbrates. This would simultaneously undermine the capacity of groups (comprising both nationals and non-nationals) to advocate illiberal and coercive religious values and encourage assimilation to the host culture.

Other policy suggestions may well be discriminatory, albeit in a less radical sense. Adoption of immigration quotas—as in Australia—only dilutes the existing population with immigrants who assimilate more readily. Discouraging or preventing the immigration of persons with few or no skills takes the sting out of resentment common in areas where poor locals and low-skill immigrants compete for jobs at the market clearing rate. Constricting access to welfare entitlements discourages the immigrant equivalent of ‘forum shopping.’ While discrimination may be justified—and may even be necessary—there are many ways to alleviate the problems Finnis outlines.

Concluding comments

As a concluding empirical point, I wish to observe that strong perfectionism—of the type endorsed by both Raz and Finnis—is vulnerable to government failure. State funded multiculturalism is a signal example. Multiculturalism’s supporters had a clear conception of the good, and successfully recruited the state in its support. Raz has long argued that a general risk of policy failure cannot lead to anti-perfectionism: instead, it should lead only to general caution. His endorsement of state-sponsored multiculturalism, then, must count as incautious.

This is not to argue that it is impossible to conceive of an objective account of the good, or to suggest that it is appropriate for states to abjure all promotion of any goods. Rather, it is to note that even if it’s possible to give substantive, objective content to the common good, this is not always a good argument for the state to support it, especially in the form of taxpayer subsidy.

States fail far too often, in ways so spectacular as to be almost inconceivable. Liberalism leaves decisions up to individuals. Paternalism—even very mild paternalism—requires the state to be systematically more competent than it is likely to be, especially over time.

I do not want to be the government official who paraphrases Star Trek when she tells Finnis ‘it’s the good, Professor, but not as you know it’.