what is liberalism?

a pluralist realist account

Liberalism is a particularly contested term. I’m not referring to liberalism here as a contentious matter of party politics, but as one of the main theoretical frameworks for addressing important moral questions of a politico-philosophical nature. Of course, the substance of party politics can be mapped on to these theoretical frameworks, but that’s a downstream concern I’ll leave for another day. Rather, the aim of this piece is to address theoretical contestation head on, and propose a relatively simple answer to the difficult question ‘what is liberalism?’.

One place to start when addressing this question is to think about who does liberalism, of the theoretical kind. As the historian of ideas Alan Ryan points out, it’s “easy to list famous liberals; it is harder to say what they have in common”. And sure, if Locke, Mill, Kant, Hayek, and Rawls are not liberals of some kind, then the term seems lacking in value. Nonetheless, there are so many differences between the approaches and conclusions of the canonical liberal thinkers that it's hard to accept there could be a set of substantive conditions binding them and only them together. Moreover, any approach relying on the assumption that canonical liberals are liberals, in order to discern truths about what it is to be a liberal, is at risk of circularity!

Of course, it could be that it’s impossible to say what all liberals, and only all liberals, have in common: that liberalism either no longer coheres as a doctrine, or that it’s always been an example of one of W.B. Gallie’s ‘essentially contested concepts’. Rather, I’ll accept from the off that we can solve the problem of liberalism’s deep variety by accepting that liberalism is not one single theoretical framework, but a family of frameworks: a set of variants. Indeed, this approach can help us to see why liberalism’s deep variety isn’t a problem, but a crucial part of its value.

A tripartite focus on freedom

Before turning to the value of liberalism’s deep variety, however, I’m going to propose some ideas about what makes its variants cohere. My main contention, rather unsurprisingly, is that any variant of liberalism will be notable for its commitment to the value of freedom. Indeed, I contend that each liberal variant will have the following three freedom-related features:

• A central commitment to freedom as a foundational objective value, and a clear theory of freedom’s relation to other values

• A philosophical conception of equally-held individual freedom and societal freedom

• An adherence to a full moral worldview on which freedom is particularly significant, and on which freedom plays a central part in grounding a conception of legitimate political society

There's insufficient space in this short piece to explain fully why I’ve identified these particular features. One reason, however, is that these features fit neatly together: they aren’t just compatible; they're mutually supportive in a way that cashes out what it means for liberalism to be a variant-rich politico-philosophical theory focused on freedom. A second reason, albeit again at risk of circularity, is simply that this is what I’ve concluded having read many of the canonical works of liberalism. In that light, it’s hardly a controversial set of features I’ve described!

Finally, beyond those reasons, I quite like the limited defence that Jan Narveson offers in a short article in which he sets out his own conception of liberalism:

“Obviously I cannot undertake a full-scale defense of liberalism here. But many would, I think, accept these [Narveson's proposed] four components even if they wouldn't identify them as 'liberalism'; and those who disagree may still find the present argument of interest.“

So even if you don’t agree with my take on liberalism — which is quite different from Narveson’s — I hope you might find something of value or interest in some part of it.

Space for pluralism

I’ll now return to my contention that all liberal variants have the three freedom-related features I listed above. One of these features relates to central values, another to conceptions of individual and societal freedom, and another to adherent moral worldviews. What’s crucial to note, however, is that the substance of each of these features could be very different within different variants, and in many different ways: including differing sets of values, conceptions of freedom, treatments of legitimacy within their worldviews, and so on. Indeed, liberalism’s space for pluralism is typically taken as central to its appeal, and I think my definitional approach captures this.

Nonetheless, I contend that liberalism’s inherent pluralism is, and must be, underpinned by commitment — across all its variants — to the grounding significance of objective and irreducible foundational moral values. Minimally, these values must include freedom and equality, because these values are central to the core liberal principle that all human beings are, and should be treated by each other as, the bearers of equally-held freedom.

This, in turn, explains the special demands that liberal theories make in relation to the justification and scope of political society. In particular, the foundational liberal values of freedom and equality underpin further objective but reducible liberal societal values, which protect and promote the equal freedom of societal members, typically including autonomy, consent, competition, and tolerance.

On my account, therefore, liberalism is value-pluralist both in the sense of: 1) affording space for multiple foundational values; and 2) requiring space for the individual to determine and pursue their own conception of the good life, free from over-bearing interference.¹

Ruling out two kinds of theories

Now, there are some politico-philosophical theories, which are typically taken to be variants of modern liberalism, that adopt a postmodern, subjectivist, approach to truth.² Yet inherent in my account is the belief that such a move, whether it is to deny the existence of scientific facts or moral facts, is to sacrifice the grounding that the family of liberal theories depends upon. I’m happy to accept therefore that these subjectivist theories should not be treated as liberal variants.

There are other theories, which are typically taken to be variants of classical liberalism — and sometimes of libertarianism — that seek to depend on a single objective moral value. Nozick’s libertarianism, as presented in Anarchy, State, and Utopia, is sometimes considered a single-value theory in this sense. Now, it's true that Nozick takes a thin approach to morality. But just because he attempts to advance an account of economic justice that depends solely on entitlement for normative purposes, doesn’t mean that his kind of libertarianism is value monist. And I think it definitely doesn't mean that his libertarianism is a value-monist theory on which the relevant single value is ‘absolute property rights’, as contended, for example, by Jo Wolff.³

This is not least because Nozick’s approach to wider morality — and indeed to economic justice, alone — quite obviously depends on valuing freedom. Indeed, it's because of Nozick’s deep moral commitment to freedom that his focus on property rights arises. But it’s also easy to find Nozick paying serious attention to moral values beyond freedom — particularly utility, but also equality, for instance — and which aren’t reducible to freedom, never mind to entitlement or property rights. As it happens, my view is that single-value moral theories aren’t just overly morally thin, but are also incoherent. But I’ll save arguing that for another day.

The classical/modern divide

As intimated above, one important distinction within the family of liberal theories is the classical/modern divide. Now, my definitional account of liberalism is compatible with classical liberalism and modern liberalism having variants of their own. But these terms serve largely to distinguish between a traditional style of liberalism dating back to Early Modern times, albeit with much earlier roots, and a newer style of liberalism, which emerged, largely in America, in the second half of the twentieth century.

Classical liberalism is perhaps most associated with the writings of Locke, Smith, and Mill, and at its heart, is focused on the importance of freedom in the context of the relationship between the individual and the state. The canonical classical liberals held freedom-focused commitments on matters of both social relevance (e.g., to free speech and religious tolerance) and economic relevance (e.g., to the value of competition and free trade), although the social/economic distinction is never neat.

As such, classical liberalism has many overlaps with libertarianism, and both have become increasingly associated with the prioritisation of the economic, and particularly a focus on property rights. The Rawls scholar Samuel Freeman claims that “[c]lassical liberals characteristically consider economic liberties and robust rights of property to be as important as, if not more important than, basic personal liberties such as freedom of conscience, expression, association, and freedom of the person” — offering Hayek as a key example of such a thinker.

I won’t go into detail here, but I think Freeman's description is generally better suited to variants of libertarianism than classical liberalism. Moreover, on my account, the extent to which any variant of libertarianism or indeed classical liberalism diverges from a broad commitment to individual freedom will help to determine its suitability for consideration as a liberal variant. I’ll return to classical liberalism below, when I sketch out a Lockean type of liberalism, which I’ll claim overlaps extensively with a dominant conception of classical liberalism.

In contrast, ‘modern’ liberalism, which is often associated with political activism, is typified by its focus on inequalities arising in the relationships that groups of people have with the state and other sources of power. Here, an interest in freedom typically tracks what is generally conceived as a ‘leftwards’ trend towards increased redistribution of holdings, state intervention in private concerns, and a move away from the theoretical lens of the individual.

Comprehensiveness

Further contestation about liberalism’s identity is reflected in contemporary debate about the extent to which liberalism is totalising, or morally comprehensive: that is, whether liberalism suffices, alone, to provide a moral worldview, and as such can extend beyond addressing politico-philosophical matters. This debate involves questions about whether other, potentially standalone, rival theories can be subsumed within liberalism.

For instance, if you are liberal and democratic, are you both a liberal and a democrat? Can you be one, but not the other? Do these questions make sense? Josh Ober is unusual in arguing that a commitment to democracy does not entail a commitment to liberalism. Samuel Freeman’s contention that Locke is a liberal but not a democrat is open to criticism on various fronts, including potentially incoherence.

I’ll respond to this particular kind of contestation largely by emphasising that liberalism, even as a family of relatively diverse variants, clearly cannot provide answers to every important philosophical question. No liberal variant will give you a full answer to why you should believe in a mind-independent external world. Although, in the same way I’ve contended liberalism must depend on the existence of moral facts, I’ve also contended it must depend on the existence of scientific facts about the world about us.

But what is liberalism’s full remit regarding matters of morality? Should those 'full moral world views' on which freedom is particularly significant — the kinds of worldview that I’ve contended liberals must hold — each be described as something particular, other than ‘liberal’?

Contemporary discussion of these matters has been influenced by Rawls’ idea of a ‘comprehensive doctrine’. A comprehensive doctrine, on Rawls’ account, is a moral doctrine that’s both general and comprehensive. Rawls gives utilitarianism as an example of this kind of doctrine. Utilitarianism is ‘general’, because its principled direction applies to a wide range of societal focuses, from individual conduct to institutional organisation and beyond. And it’s ‘comprehensive’ because it provides similarly wide-ranging conceptions of value and virtue: informing all kinds of personal and interpersonal conduct.

In his second book, Political Liberalism, Rawls tells us that two of his best-known doctrines are not comprehensive doctrines. One of these is ‘justice as fairness’, which, in Political Liberalism, he presents rather as a political conception of justice. The other is ‘political liberalism’, itself. Rawlsian political liberalism is not a comprehensive doctrine because, Rawls tells us, it is aimed at the peaceful co-existence of people who commit to different comprehensive doctrines, as enabled through their ‘overlapping consensus’.

Two questions from Rawls

Two questions arise from Rawls’ approach for my account of liberalism. First, whether any of my liberal variants could be sufficiently general and comprehensive to count as a comprehensive doctrine. Second, whether Rawlsian political liberalism counts as a variant of liberalism, on my account.

The answer to the first of these questions is surely no. This is because, on my account, one condition of counting as a liberal variant is that a doctrine must have "an adherence to a full moral worldview on which freedom is particularly significant, and on which freedom plays a central part in grounding a conception of legitimate political society”. Clearly, if anything, it is one of these ‘full moral worldviews’ that would count as a Rawlsian comprehensive doctrine, rather than any appended liberal variant.

The second question is trickier. But it seems clear that Rawlsian political liberalism does not count as a variant of liberalism, on my account, for at least the following three reasons. First, because the “clear theory of freedom’s relation to other values” that I require ranges beyond Rawlsian political liberalism’s tight political remit. Second, because Rawlsian political liberalism is again, by its nature, capable of accommodating competing comprehensive doctrines, rather than adhering to just one sufficiently freedom-focused worldview. And third, because the substantive reliance of Rawlsian political liberalism on consensus-making seems in tension with the commitment to objective moral values my account of liberalism demands.⁴

Lockean-type liberalism

I’ll finish by briefly sketching the particular variant of liberalism that is broadly my personal preference. By this, I mean it's the theoretical framework that I find most convincing for addressing important moral questions of a politico-philosophical nature. This is a variant of liberalism that's heavily inspired by the ideas of John Locke, and which overlaps extensively with a more general dominant variant of classical liberalism. My aim here, however, isn't to describe a particular theoretical trend over time — although many theorists have been inspired by Locke’s approach to liberalism — but rather to posit a coherent liberal variant, built from Locke’s notable relevant ideas.

Here, I’ll return to my initial contention that any liberal theory will have all of the following three features:

• A central commitment to freedom as a foundational value, and a clear theory of freedom’s relation to other values

• A philosophical conception of equally-held individual freedom and societal freedom

• An adherence to a full moral worldview on which freedom is particularly significant, and on which freedom plays a central part in grounding a conception of legitimate political society

I’ll now use this framework to identify what makes Lockean-type liberalism a distinct variant:

• Its central commitment to freedom and equality as foundational objective values, most clearly manifest in the core principle that no human being is naturally subordinate to another. From this derive significant commitments to reducible objective values including consent, self-ownership, and tolerance.

• Its theory of individual freedom, on which being free means being capable of making reasoned decisions about how to act or not to act, and following through on those decisions. Being a ‘free agent’, in this sense, is presented as something that sets human beings apart, morally, from creatures without this capacity.

• Its theory of societal freedom, on which human beings should be recognised as equal free agents, as a matter of fact, and who should therefore be free to exercise their natural capacity for making and acting on reasoned decisions. But who can and should also therefore be held morally responsible for those actions.

• Its adherence to a full moral worldview on which natural law reflects God’s wishes and the natural freedom and equality of all human beings, and acts as a necessary moral restraint on political power and interpersonal behaviour. Further features of this worldview include commitments to the following ideas: that rights, including a core set of natural rights, are central to matters of justice and wider morality; that an unequal distribution of holdings is not inherently problematic on equality grounds; that the consent of a political society’s members is necessary for its establishment; and that the members of such a society give the ongoing consent that’s required to maintain its legitimacy by partaking in democratic deliberation, on a representative model.

This is only a brief sketch of some relevant Lockean-liberal ideas.⁵ But I hope it goes some way to showing the richness possible in any particular variant of liberalism, on my pluralist realist definitional account.

1. If you’re interested in discussion of this kind of distinction, you can read Elinor Mason’s SEP entry: https://plato.stanford.edu/entries/value-pluralism/

2. See, e.g., the denial of biological truth that’s common to a particular strand of what’s thought of as modern liberalism. This is influenced not least by Judith Butler, who has stated that “[t]he notion that there might be a ‘truth’ of sex, as Foucault ironically terms it, is produced precisely through the regulatory practices that generate coherent identities through the matrix of coherent gender norms”: Butler, Gender Trouble, p.17.

3. See e.g., pp.3-4 in Wolff’s Robert Nozick, a book I otherwise like.

4. I’ll go into the details of why I think this, some other time — perhaps in the next of my weekly ‘5 top things I’ve been reading’ pieces.

5. I won’t bother to link to where to you can find each of these Lockean ideas, as they are all well-known and easily searchable. But comprehensive discussion of most of them can be found in the Second Treatise, except for the theory of individual freedom, which is laid out most clearly in the chapter entitled ‘Of power’ in An Essay Concerning Human Understanding. Some places to find good discussion of Locke as a liberal include: Richard Ashcraft’s Revolutionary Politics and Locke's Two Treatises of Government, William Kerstetter’s John Locke's Conception of Freedom, Daniel Layman’s Locke Among The Radicals, E.J. Lowe’s Locke, Eric Mack’s John Locke on Property, John Simmons’ The Lockean Theory of Rights, Gopal Sreenivasan’s The Limits of Lockean Rights in Property, and Jeremy Waldron’s God, Locke, and Equality.