Liberal Neutrality: A Rejection

By: Paul Butterfield , MSc Philosophy 2015, University of Glasgow

This article will argue that a commitment to neutrality cannot be the defining feature of the liberal state, at least under any currently common definition of neutrality. It will consider some desiderata for a definition of the liberal state, and conclude that existing interpretations of neutrality fail to fulfil these criteria. In doing so, it will agree with the arguments of Alan Patten in ‘Liberal Neutrality: A Reinterpretation and Defence’ about the implausibility of previous attempts to define liberalism via neutrality, but ultimately conclude that his own interpretation of neutrality cannot be understood to define the liberal state, either.

State neutrality in political philosophy is typically related to ‘conceptions of the good’: if a state is neutral, it will avoid promoting one, or some, of its citizens’ conceptions of the good over others. A conception of the good is a belief about what makes a human life valuable, or worth living. A belief that the most valuable life is one spent perfecting one’s basketball performance, for instance, constitutes a conception of the good.

This is the general idea of political neutrality, but there are various ways of interpreting the idea more specifically. Patten, in ‘Liberal Neutrality’, claims that traditional conceptions of state neutrality can be categorised as either “neutrality of intentions” or “neutrality of effects” (p.254). A state violates neutrality of intentions if it has, as its aim or reason for adopting a policy, the idea that certain conceptions of the good are preferable to others and should, therefore, be promoted. A state violates neutrality of effects, on the other hand, if its policies result in the promotion of certain conceptions of the good over others. Patten rejects both, and ultimately proposes his own version of neutrality which, he believes, can be the defining feature of a liberal state: “neutrality of treatment”. A state violates Patten’s neutrality of treatment if its policies themselves promote some conceptions of the good over others.

Patten explains the difference between the concepts of neutrality of treatment and neutrality of effects by way of an analogy involving a philanthropist (p.258). He supposes that our philanthropist is approached by two good causes in need of funding. To neutralise her effects in relation to each cause, the philanthropist would need to donate the amounts of money she calculates would make each cause equally successful. (We can imagine these sums coming apart, for instance, in the case of a charity dedicated to saving lives from a disease with an inexpensive vaccine, in comparison with a charity dedicated to saving lives from a disease whose eradication requires expensive research.) To neutralise her treatment of each cause, however, she simply needs to donate an equal amount of money to each. Neutrality of treatment requires equal accommodation for various conceptions of the good, then, but not necessarily an equal effect on the popularity or success of various conceptions of the good.

Patten rejects neutrality of intentions because it will generate false positives, considering states as ‘neutral’ when they seem, in fact, to be no such thing. He observes that a state which establishes a state religion could still act neutrally, on an intentions view, if it did so in order to increase respect for the state, and not because the conception of good professed by that religion is preferable to that of other religions. But establishing a state religion is one of the most clearly non-neutral policies a state can carry out. We can frame Patten’s concerns with neutrality of intentions in a more general way: no policy is now inherently neutral or non-neutral, as neutrality is now dependent upon the motivations of policy-makers. But there are various policies for which it seems bizarre to claim that they could ever be neutral (or non-neutral), no matter what the motivations of policy-makers are. So neutrality of intentions tracks the wrong factors.

A related worry, unmentioned by Patten, is that a state could qualify for neutrality on this view through a combination of ignorance and optimism. Consider a state which enacts a policy and falsely believes that this policy will not promote any particular conception of the good. Even though they are wrong, the deluded policymakers of this state have guaranteed the state’s neutrality, according to neutrality of intentions. After all, they certainly could not intend for their policy to promote any conception of the good – they don’t even realise that it is likely to do so. But, whatever standard we want to use to distinguish between neutral and non-neutral states, it must surely relate to something more substantial than the ignorance and optimism of policymakers. For more than one reason, then, neutrality of intentions cannot be the correct interpretation of neutrality, and cannot define liberalism.

Neutrality of effects will not be able to do so, either. This is because every possible action (or inaction) a state can perform will have some effect on the popularity of various conceptions of the good. Patten writes that even “legal protections of the basic liberties make it relatively more difficult for boring and unpopular ways of life to flourish than would be the case in a legal system where people are locked into particular ways of life” (p.257). The upshot of a ‘neutrality of effects’ view, then, is that essentially no policy a state could introduce would be considered neutral.

Even if neutrality of effects could separate neutral policies from non-neutral policies, it would still produce the wrong results in important cases. Consider an inept dictatorship. It would like nothing more than to impose one particular conception of the good upon its citizens, but all of its policies get bogged down in bureaucratic red tape. As a result, its efforts to promote one conception of the good end up having no material effect upon its citizens’ day-to-day lives. On a neutrality of effects view, this dictatorship would be as neutral – and as authentically neutral – as the state which manages, through genius political manoeuvring, to ensure that it does not make any conception of the good more popular or successful than any other. And a definition of neutrality which suggests that these states are equally neutral cannot be accurate.

What, then, should we make of Patten’s favoured neutrality of treatment view? Certainly, it avoids the problems we have identified with neutrality of intentions and neutrality of effects. It can, after all, identify distinctly neutral and non-neutral states, and proscribe certain actions regardless of how well-meaning the states that perform them are. Nevertheless, neutrality of treatment cannot be the defining feature of liberalism. This is because it will not allow a state to tax its citizens – a policy that the liberal state must surely be allowed (if not required) to carry out.

To see why neutrality of treatment forbids taxation, consider citizens with wealth-based conceptions of the good – conception of the good summed up by a phrase like “I want to be wealthy”. I take it that such people exist: the pursuit of wealth seems like a common mid- to long-term goal, and we can call to mind entrepreneurs who would certainly consider their lives to have lost value, or meaning, if they were unable to accumulate wealth.

An income tax would infringe upon the success and popularity of wealth-based conceptions of the good, for two reasons. Firstly, an income tax places a larger burden upon those who are successful on a wealth-based conception of the good than those who are unsuccessful, because it requires the wealthy to contribute more than the poor. This is analogous to a policy which dictates that, because you are twice as good at basketball as I am, the state will take twice as much from your income as it takes from mine. Secondly, an income tax confiscates the direct means by which one becomes successful on a wealth-based conception of the good. This is analogous to the state-wide confiscation of some proportion of all basketballs. Inasmuch as the basketball-related policies above would fail to be neutral with regards to budding basketballers, taxation fails to be neutral with regards to the money-motivated.

But of course, the liberal state must be allowed to tax its citizens. If a version of neutrality prohibits the neutral state from taxing its citizens, it cannot define liberalism. For neutrality of treatment to do so, then, it will have to be shown that wealth-based conceptions of the good are not legitimate ‘conceptions of the good’ at all.

One way we might attempt to do this is to hold that it is impossible for a state to avoid infringing on the popularity and success of wealth-based conceptions of the good in this way. This would be the case if taxation was the only way to secure public goods. After all, for a state to exist at all, it seems that it must provide public goods of some sort. Even those political theorists with the most restrictive visions of the state believe that, to qualify, a government must perform some functions for its citizens – ensuring some level of security, for instance, or enabling the ownership of private property. If even these minimal functions of the state required funding through taxation, then equal consideration for those with wealth-based conceptions of the good would require the state not to exist in the first instance.

But the state can provide public goods without taxation. To see how, imagine a state which secures public goods – for vividness, let’s suppose it secures similar levels of public goods to modern European and North American democracies – through a policy of universal community service. That is to say, the state is able to provide the services it does not by confiscating some portion of our money, but by confiscating some portion of our free time. Rather than giving the state a percentage of our income, then, we donate (say) four hours a week of our time to (say) volunteering for the police force. I do not claim that this state would be preferable to a taxing state. I do not even claim that it would be more neutral than a taxing state, since we can imagine citizens with conceptions of the good which require lots of free time (think of those for whom the ‘good life’ consists in poolside sun-loungers and waiter-brought cocktails). My point is simply that this state is conceivable, and if it is conceivable, we cannot say that the taxing state only infringes upon wealth-based conceptions of the good through necessity. The money-motivated can quite legitimately point to those who value extensive leisure time, and ask, “Why infringe upon our chances of success in our chosen way of life, rather than theirs?” So we have foreclosed on the idea that taxation is a necessary part of political organisation.

Another attempt to disqualify the motivation to pursue wealth from being a conception of the good might involve the claim that conceptions of the good must be self-regarding. The phrase “I want to be wealthy” is other-regarding, so this argument would go, and thus it does not count.

It is certainly true that wealth-based conceptions of the good are, to some extent, other-regarding: being wealthy doesn’t require one to have a certain absolute sum of money, but to have more than most. But I don’t believe that this is other-regarding in any special or unusual way. Any conception of the good that involves a talent, or a career goal, will ultimately require one to live a life that is in some respect exceptional: to be more skilful, or more fortunate, than most. To disqualify wealth-based conceptions of the good for being other-regarding, then, we would also have to disqualify many or most other candidates for conceptions of the good.

There is, then, no way to avoid the conclusion that, on a neutrality of treatment view, taxation violates neutrality. Thus, neutrality of treatment cannot be the defining feature of the liberal state, since the liberal state must surely be permitted to tax its citizens. Furthermore, because they generate the false positive and false negatives that we have seen, neutrality of intentions and neutrality of effects are also unable to define the liberal state. It is conceivable that a future interpretation of neutrality might be more sophisticated, and avoid all of the problems identified in this article. An assessment of that theory of neutrality will have to be undertaken if and when it arises. For the moment, though, three interpretations of the term have been proposed, and all three fail to define liberalism. We should conclude, therefore, that liberalism is not defined by neutrality, regardless of how the latter term is currently understood.



Patten, Alan (2012): ‘Liberal Neutrality: A Reinterpretation and Defence’ in Journal of Practical Philosophy, Volume 20, Issue 3, pp.249-272

Share your thoughts on this!

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s