Skip to content

Social Pluralism and the Principle of Subsidiarity

Subsidiarity is one of the four basic principles of Catholic social teaching (along with dignity of the individual, common good, and solidarity). Principles however do not create their own facts, and subsidiarity presupposes a certain set of social facts: first, that there exists a comprehensive community called polity (put to one side the issue of international polity) that has a twofold common good: common end or ends, and a common form (a regimen) for shared action in achieving the end. The common form is called “the bond.” If that obtains, we have what the tradition called a “true society” that transcends aggregate intersubjectivities and is thus entitled to hold itself out as “one.” It is a social agent, a locus of rights and responsibilities.

But the principle of subsidiarity immediately depends upon yet another set of social facts. Namely, that within the bosom of the comprehensive community there exist still other communities, each of which possess its own unique common good and a social form that constitutes “the bond.” These bonds, including the political, differ qualitatively and not merely quantitatively. The political bond, for example, is not just a large family.

This formulation goes back to the medieval appropriation of the dispute between Aristotle and Plato on the unity of the city. Plato argued for a unitary bond—a single common good and social form. Wives and children must be regarded as bona communia, a common stock along the lines of a municipal water system. The bond is entirely political. Aristotle contended that political forms come into existence through the other social bonds: marriage, family, extended families, villages, etc. To relinquish these other social bonds is unnatural and defeats the purpose of polity.  For his part, Thomas Aquinas argued that a polity consisting of other social forms is “better and more divine” than a unitary polity.1 It is more perfect precisely because it includes a diversity of social bonds (not reducible to one another) that are key causes of human eudaimonia.

Without a set of social facts along these lines the principle of subsidiarity has no useful work to do. But on the supposition that we have a community of communities, differing qualitatively, the principle is fairly easy to understand. What do these communities owe to one another? The answer is subsidium (aid, assistance), here by analogy to the body. What do the organs owe to the whole and to one another? Subsidium according to the social competencies of each. Charles Taylor has accurately called this social scheme “hierarchical complementarity.”2 The social parts need one another and recognize that one cannot substitute for the other.

The principle of justice governing such a system is straightforward: given the obligation of providing subsidium, it must be done in such a way that does not cancel, absorb, or confuse the social function of the parts. It is a limit on distributive justice, for the sake of the parts and the whole. Although subsidiarity is often invoked as an obligation and a limit on the higher community, the principle is best understood as circular.3 This cybernetic (or “circular”) model is especially apt for describing how domestic, political, and ecclesial societies are subsidiary to one another without existing in an assembly-line type of hierarchy. In fact, a multilateral and communicative hierarchy would seem to follow from the very idea that domestic, political, and ecclesial communities are “necessary” for human eudaimonia.

Modern Times

The principle was asserted with urgency in modern times. The treaties of Westphalia (1648) understood sovereignty as absolute, indivisible, and perpetual power over a certain geographical territory. This made it difficult to conceive of a comprehensive society that includes qualitatively different social forms. For one thing, the indivisibility of sovereign power tended toward a Platonic monistic model. For another, some social forms are extraterritorial. This was, and still is, true of some religious bodies, especially the Catholic church. Baptism and other sacraments do not constitute members on a strictly territorial basis. Its social functions and forms, from the episcopacy, to religious orders, to charitable organizations cannot be described in such a way.  The problem persisted into the revolutionary era post-1789. On the French, and later the German models, the state claims a monopoly on fraternity, at least for public purposes.

Modern state-making regimes understood at least part of the problem. Indivisible sovereignty over a territorial social body can prove very inefficient. Thomas Hobbes (1588-1679) provided an early, and very clear, model for understanding the relationship between groups and the sovereign state. It has been called the “concession” theory.  The term concession is traced to the Edict of Gaius in the Digest 3.4.1, where collegia or other social bodies are conceded “on the pattern of the state.”4 The sovereign does not have to do all the work. A sovereign can outsource some of the functions of sovereignty, just as Queen Elizabeth empowered a privateer, Francis Drake, to lead the royal navy against Spain.

While the state does not deny its own plenitude of power or sovereignty, it does recognize the contingent fact that it cannot efficiently reach all of the objects within its formal power. Accordingly, it will outsource power, by way of a quasi-delegation, to other groups for the purpose of efficiently creating and distributing certain goods and services:  education, charitable relief of the poor, and the orderly transfer of property or investment, to mention only a few. The complexity and scale of modern states practically guarantee that the sovereign must make concessions. It must, in this sense, learn to devolve.

While despotic regimes will tend to be stingy, liberal regimes will tend to be generous in giving concessions. England was the model for a liberal regime, jealous of its sovereignty, but ever ready to outsource certain functions to corporations and to unincorporated groups and to trusts, even to pirates acting as auxiliaries of the royal navy. Contemporary China perhaps stands at the opposite pole of this spectrum; in 2007, for instance, China’s State Administration of Religious Affairs announced Order No. 5, a law covering the “Management Measures for the Reincarnation of Living Buddhas in Tibetan Buddhism.” The state prohibits Buddhist monks from returning from the dead without government permission. No one outside China can influence the reincarnation process, and only monasteries in China can apply for the concession. But in both cases, we are presented with a theory of intermediate societies as state concessions: the Halliburton principle of subsidium, so to speak.

Given these developments, the principle of subsidiarity was made to do a quite different work. It became a principle of efficiency, often interpreted as mandating action at the lowest possible level. It is on this basis that neoliberals can find the principle somewhat agreeable. In many cases, the lowest and most efficient (in dollar terms) level at which goods can be produced and allocated is a free market, which is no society of any sort. This is why Charles Taylor contrasts hierarchical complementarity with the modern “social imaginary” of the market. It is not as though we moderns cannot understand the original principle, but it is difficult to understand it in relation to our social facts. On the traditional view, a social bond other than the state does not mean other than the whole.

The Power-Checking Model

We may call the modern liberal version of subsidiarity the power-checking model, because it estimates the value of groups other than the state chiefly in terms of a check upon untrammeled power. Montesquieu wrote: “Political liberty is found only in moderate governments. . . . It is present only when power is not abused, but it eternally has been observed that any man who has power is led to abuse it; he continues until he finds limits. . . . So that one cannot abuse power, power must check power by the arrangement of things.”5 Hence, the famous idea of civil society as “intermediate powers.” What interested Montesquieu was not the specifically social landscape, or the milieu intérieur, of corporate persons (an ontology that perhaps he took for granted) so much as the general distribution of “powers.”  Civil society is useful as an arrangement that checks abuse of power and thereby inclines a political society to moderation. As Justice Brennan once noted, one of the reasons motivating American protection of associational liberty is to “foster diversity and act as critical buffers between the individual and the state.”6 He calls this reason “instrumental” because the association protects other values.7

The logic of devolution and privatization of so-called subsidiary groups orients much contemporary debate about how distinct societies share a common social order. Take, for example, the debates in Europe and the United States over the issue of marriage. What kind of juridical person is a marriage? In 1992, the Hawaiian Supreme Court defined marriage as “a partnership to which both parties bring their financial resources as well as their individual energies and efforts.”8

This point was reiterated in the controversial 2003 decision of the Massachusetts Supreme Court, which prohibited the legislature from giving legal title of marriage only to one man and one woman. Let’s put to one side the moral issue of whether marriage ought to be exclusively heterosexual. This puts the cart before horse. First, we want to know whether there are group-persons distinct from partnerships, and second what reason the state has to recognize them.

We begin by considering the nature of civil marriage itself. Simply put, the government creates civil marriage. . . . Civil marriage is created and regulated through exercise of the police power. . . . Civil marriage anchors an ordered society by encouraging stable relationships over transient ones. It is central to the way the Commonwealth identifies individuals, provides for the orderly distribution of property, ensures that children and adults are cared for and supported whenever possible from private rather than public funds, and tracks important epidemiological and demographic data.9

We read that the state does not merely regulate, but creates marriage through the exercise of the police power. What aspects of good order move the state to allow married people to be a right-and-duty social unity? The Court mentions economic reasons (property), sociological reasons (stable relationships), health reasons (care for the old or indigent), and scientific reasons (collection of epidemiological data). Stability of a relationship echoes something of the older notion of a society, but the gist of the court’s position is the utility of marriage. The state breathes into the dust of sexual relationships and private aspirations to intimacy, and creates a “person” at law.

This person, then, becomes the site or occasion for bringing about more efficiently certain results which are in the interest of the state.10 All that remained for the Massachusetts Court to do was to judge that one sex or two sexes are immaterial to the state’s interest in having other agents procure the publicly desirable results, and therefore not to favor one arrangement over the other. The Massachusetts decision is a pure example of concession theory. The Court leaves untouched the question whether this juridic person is a society or a transient partnership. The public efficiencies falling within the purview of the positive law could be attached to either a partnership or a society.

Either way, the state enjoys some monopoly over public things. But, where the concession model seeks to protect and maintain state sovereignty by outsourcing its power to other groups, the power-checking model endeavors to shrink its scope—at least materially and politically. Let the state be sovereign in the “modern” sense of the term; but let this sovereignty be materially diminished by intermediate groups. These groups are not estates in the old sense of the term, for they have no representative power or authority; they do not constitute political bodies. Rather, the intermediate powers constitute a vast sphere of private judgments, choices, and actions by individuals and associations. Understanding that the state was no longer limited from above, it followed that its power is limited either from within (e.g. the division of powers), or from below.

The power-checking model treats devolution as privatization. It, too, wants the state to devolve for reasons of efficiency, but with a value-added purpose. For example, private schools are useful not only because they efficiently allocate educational resources, but also because they check the untrammeled state power over education. This double efficiency has always proved essential to the liberal social theory. Like the two faces of Janus, it looks in one direction toward private competition, organization, and efficient distribution of resources, and it looks in the other direction toward the negative liberty accruing from private initiative. The state is put in the position of having to justify, on cost-benefit grounds, why the private sphere should not prevail whenever there is a concurrent jurisdiction or interest in a common thing: fisheries, education, capital investment, etc.

In both the United States and Europe, this model is often associated (more or less explicitly) with “subsidiarity.”11 In the next section, I will explain why we should resist the equation. But for now I call attention to the fact that the intermediate powers analysis is not principally interested in the sociality of diverse group-persons. It focuses rather upon efficiency, which is the common coin of public policy and discourse. It matters little whether the efficiencies have this or that social form. The key insight is that the state be limited by the private sphere. If we ask why the state should be so limited, the answer will be that it increases liberty and that such an arrangement is more efficient.

Perhaps the most astute and powerful “checking” argument for civil society is made by Ernest Gellner, in Conditions of Liberty: Civil Society and its Rivals (1994). Gellner points out that “civil society” is ambiguous. From one point of view, civil society can mean the “social residue left when the state is subtracted.”12 Consisting of strong bonds of solidarity in family, tribes, and religious institutions, this “residue” can prove to be very potent. The polycentric nature of traditional societies is very effective if we were only interested in checking the power of the modern state. In a modern, democratic culture, however, civil society must not only check state power, but must also liberate the individual from the suffocating obligations of common faith and kinship. The “miracle of civil society” requires loose associations which protect individual liberty from the solidaristic bonds of the state and of traditional communities. Once the strong solidarities from above and below are weakened, we enjoy the kind of society suitable to what he calls the “modular man.”

The [Western conception of civil society] has not committed itself either to a set of prescribed roles and relations, or to a set of practices.  The same goes for knowledge: conviction can change, without any stigma of apostasy. Yet these highly specific, unsanctified, instrumental, revocable links or bonds are effective. The associations of modular man can be effective without being rigid.13

Gellner’s work is very important because he delineates the full implications of the “counterweight” theory of civil society: devolution from above and dis-incorporation from below. His sociology is at once descriptive and normative—at least for us, who live in a market culture, and who prize the associational life of “modular man.” That human nature is sociable, and that sociability is capable of strong solidarities from above and from below are not in question. The question, rather, is whether such strong solidarities are useful and agreeable to the democratic culture. Gellner has emphasized what we have called partnerships rather that a plurality of societies possessing intrinsic common goods.

The uses of “modular man” and his passing partnerships notwithstanding, it’s still possible to imagine and to analyze social relationships that line up with the original principle of subsidiarity. Aristotle proposed that political society begins in necessity but (if things mature and go well) ends in excellence. We can extend his insight more generally to any true society with a two-fold common good (common end or ends, and common form of action). In the Catholic tradition, there are three such societies called “necessary”: marriage-family, polity, and church.14 The first two are by nature, the third by grace. Necessity here means necessary for eudaimonia, human flourishing.

Human beings are familial, political, and ecclesial animals. Rooted in necessities, these societies tend under the right conditions to distinct kinds of perfection. The original utility is transcended, precisely insofar as the social bond becomes the perfective principle. Married people want to procreate, but also to do so matrimonially, and as parents, and grandparents. To create and perform a Haydn symphony, it is necessary for there to be several players on different instruments, but the perfection is the performance, doing it together. To create and maintain a polity, it is necessary to institute orders of peace beyond the family, but its perfection is the political bond itself. The perfection in each case is meeting the necessity and allowing it to become an ongoing excellence.

The premise of the older doctrine of subsidiarity stands or falls on the fact human happiness is discovered in social bonds that are not merely transitive, but which persist not only as necessities but as perfections—worth doing for their own sake. Therefore, a comprehensive social order needs to keep them in play and harmonized with one another.  Consider ten people who are citizens of France, and members of families, and parishioners of St. Rita’s parish: we might hope that those memberships serve one another without prejudice, but this is easier said than done. We can take for granted that there are many other kinds of social relations that can be engaged temporarily, revised or discarded as prudence might suggest. An elite military unit can enjoy a profound bond, but with apologies to the Spartans, it is not a koinonia, a way of life that contains the rudiments of flourishing.

Work among the Necessary Societies

In initiating and developing the social encyclical tradition, popes such as Leo XIII and Pius XI appealed to the principle of subsidiarity to explore how the rapidly changing conditions of human work might be harmonized with the three necessary societies, and other social bodies. Human labor begins by seeking a change in material conditions, but—as Leo emphasized in Rerum Novarum (1891)—it always already transcends those material changes, for human work requires intelligence and providence, namely the provisions for the wellbeing of others, beginning in the family, and extending to corporations, unions and guilds, the polity, and so forth. Labor is not just the perfection of material things but also, and primarily of persons.

The perfection, the subsidium, brought about by work is not only some good or service, but includes a social bond, absent which it could only be deemed slavery. Here again is the circular notion of subsidiarity: the political community gives subsidium by protecting the social bonds, and the workers give their subsidium by maintaining the social cells of friendship.

That the older notion of subsidiarity is not totally obscure is confirmed by some recent social scientific findings. Social capital theorists, for example, suggest that resources of social participation should be evaluated in three dimensions:  the type of society (sports club versus a political party or a motorcycle club), the scope of memberships (few versus many affiliations), and the intensity of participation (very active vs. compliant).

By design, the social capital methodology is committed to a very weak typology regarding the kinds of social life, their respective ends, and internal modes of union intended by common action. For this method of analysis, social capital (trust and cooperation) is a surplus created by individual actions that are in a general way social, at least as measured by trust and cooperation. What matters, however, is not a ranking of social orders in an order of ontological or anthropological importance, but rather the individual’s skills and morale built up through diverse social engagements.15 At the end the theorist wishes to give quantitative picture: more or less sociality.

For instance, a widely-cited study of social capital concludes: “the most productive form of participation with regard to the formation of social capital seems to be not only participation in several associations, but multiple affiliations in associations with different purposes.” A motorcycle club, the Knights of Columbus, a bowling team, and the national guard: on this view, each is good, and as good as the others. It’s the accumulation and diversification and time given to the individual’s associations that matter.

Compare that with membership in what could be called a non-substitutional society. A non-substitutional society has the following properties: (1) There is no social equivalent [a bowling team does not substitute for a parish, though a parish might have a bowling team]. (2) Inclusion means full inclusion or no inclusion at all—the baptized person is a member of the Body of Christ. (3) Membership is irreducibly social, which means that the social relation(s) do not come into existence simply by virtue of exchanges and distributions; rather, exchanges and distributions presuppose the social union—you can’t buy in or cash out—not even in purely civil marriage and divorce can one take 35 percent of the marriage. (4) Members intend for the society to be more than merely transient—the union of a family cannot be intended to happen only on Tuesdays and Thursdays for the next twenty-four months.

In the case of societies having these properties, trust and cooperation measured by social capital theorists are a good start, but do not adequately capture the social virtues needed to sustain a common life. Some social scientists do look deeper, albeit behind the veil of aggregate correlations. For example, in a recent groundbreaking study of the breakdown of middle and lower middle class, the Princeton economists Anne Case and Angus Deaton suggest that recent patterns of mortality and morbidity “move in tandem” with withdrawal (exclusion) from marriage, children, religious congregations, and political society.

The study points to the importance of participating in and sustaining a rather traditional typology of social unions. Case and Deaton do not defend such unions; they are content rather to point toward them via their data sets. To fail at marriage, or family, or civic participation, or religious membership is far more serious than failing at bowling. This was the logic of describing the three necessary societies as “necessary”: a person who cannot sustain the bond with spouse or family, who is disillusioned and angry with every political authority, from mayors, to governors, to national parties, and who has dropped participation in churches, is not a prime candidate for happiness. Of course this does not prove that the older notion of social bonds is the key to human eudaimonia, but it is an impressive correlation nonetheless.

Our polarizations today thrive on deep suspicions of social pluralism, by which I mean the subsidiary relations within a more comprehensive community. It is not just a crisis of political authority or of economic efficiency but one of social realities. The social bonds are almost entirely reactive. We are back to what Aristotle would have deemed mere necessity without the prospect of a perfection.

This article is an American Affairs online exclusive, published December 20, 2021.


1 Sententia Politic., Lib. 2 1.1. n. 11; Lib. 1. 1 n. 3.  Thomas’s argument is not only that a political common good is superior to the good of an individual, but also (and chiefly) that the political community “includes all  the other communities,” all of the bonds necessary for human flourishing.

2 Charles Taylor, Modern Social Imaginaries (2004), 10.

3 Prevalence of the top-down situation is due to the standard problem of protecting “lower” societies against the heavy hand of the state.

4 Digest 3.4.1 at 137.  Note the three important verbs in this dictum: concedere, permittere, confirmare.

5 Montesquieu, Spirit of the Laws, 11.4.

6 Roberts v. Jaycees, 468 U.S. 609 (1984), at 619.

7 Id., at 617f.  But expressive associations have intrinsic value because they express and safeguard the person’s “ability independently to define one’s identity.”  Even so, Brennan is very careful not to attribute intrinsic value to the association as such.  The value belongs to the individual.

8 Baehr v. Lewin, 852 P. 2d 44, 58 (citing Gussin v. Gussin 73 Hawaii 470, 483, 836 P. 2nd 484, 491 (1992).

9 Goodridge v. Dept. of Public Health, 440 Mass. 309 (1993).

10 Compare to Justice Joseph Story, a jurist from Massachusetts and Chief Justice of the Supreme Court.  “[Marriage] may exist between two individuals of different sexes, although no third person existed in the world, as happened in the case of the common ancestors of mankind.  It is the parent and not the child of society.”  Joseph Story, Commentaries on the Conflict of Laws (1834), 100.

11 Whether the European Community treaties (and amendments) mean by subsidiarity something more than a rule of social efficiency is difficult to determine because the original language, informed by Catholic social thought (mostly through the German thinkers) was transported to a more lawyerly emphasis on constitutional allocation of powers. The difference between the two is intelligently surveyed by Christoph Henkel, “The Allocation of Powers in the European Union: A Closer Look at the Principle of Subsidiary,” 20 Berkeley J. Int’l L. 359 (2002), 359-386.

12 Ernest Gellner, Conditions of Liberty: Civil Society and its Rivals (New York: The Penguin Press, 1994), 212.

13 Ibid., 100.  Gellner’s work has a strong resemblance to Justice Brennan’s position that associational life should not be ranked according to some antecedent hierarchy, but rather according to the value of the individual.  In a free society, the law should refrain from locking-in, as it were, a traditional scheme of social values.

14 The expression “three necessary societies” comes from Pope Pius XI’s Divini Illius Magistri (1929), §11. For further discussion, cf. Russell Hittinger, “The Three Necessary Societies,” First Things (June 2017).

15 Dag Wollebæk and Per Selle, “Participation and Social Capital Formation:  Norway in a Comparative Perspective.”  Scandinavian Political Studies, Vol. 26 – No. 1, 2003, 67-91. The authors conclude that “the most productive form of participation with regard to the formation of social capital seems to be not only participation in several associations, but multiple affiliations in associations with different purposes.”

Sorry, PDF downloads are available
to subscribers only.


Already subscribed?
Sign In With Your AAJ Account | Sign In with Blink