Kai Möller 1

Dworkin’s Theory of Rights in the Age of Proportionality

ABSTRACT: There is probably no conceptualisation of rights more famous than Ronald Dworkin’s

claim that they are ‘trumps’. This seems to stand in stark contrast to the dominant, proportionality-

based strand of rights discourse, according to which rights, instead of trumping competing interests,

ultimately have to be balanced against them. The goal of this article is to reconcile Dworkin’s work

and proportionality and thereby make a contribution to our understanding of both. It offers a critical

reconstruction of Dworkin’s theory of rights which does away with the misleading label of rights as

‘trumps’ and shows that, far from being in conflict with proportionality, properly understood

Dworkin’s work supports and supplements that doctrine and provides a much-needed account of its

moral foundation as being about human dignity, freedom, and equality.

1Associate Professor of Law, Department of Law, London School of Economics and Political Science. I
would like to thank Jurriën Hamer, Ester Herlin Karnell, Mattias Kumm, Dimitrios Kyritsis, Iddo
Porat, Gila Stopler, and Samuel Tschorne for their valuable comments and suggestions. Various
versions of this paper were presented at the Law Faculty Seminar Series at Utrecht University and the
Free University of Amsterdam, the Public Law Theory Seminar Series at Humboldt University Berlin,
and the Berlin Colloquium on Global and Comparative Public Law at Humboldt University. I am
grateful to the organisers and the participants for their valuable feedback. Furthermore, I would to
acknowledge the support of the Grant Agency of the Czech Academy of Sciences as part of a project
on ‘The Role of the Principle of Proportionality in the Decision-Making of Courts’ (grant ID: 15-
23955S – Subsidy RVO: 68378122), awarded to the Institute of State and Law of the Academy of
Sciences of the Czech Republic.

1
There is probably no conceptualisation of rights more famous than Ronald Dworkin’s claim

that they are ‘trumps’. Yet, this is puzzling in at least two ways. First, it is not at all clear

what it means to say that rights are trumps; whatever the correct answer may be, it cannot

mean that rights such as the right to private life, the right to property, or the right to an

education – to name just a few rights that are widely recognised in contemporary rights

discourse – always or at least normally ‘trump’ all or most interests that clash with them.

Second and related, by singling out their ‘trumping’ quality as their central, defining feature,

Dworkin knowingly or unknowingly places his theory in sharp contrast to the dominant,

proportionality-based strand of rights discourse, according to which rights, far from

trumping conflicting interests, ultimately have to be balanced against them.

The goal of this article is to reconcile Dworkin’s theory of rights and proportionality

and thereby make a contribution to our understanding of both. 2 As will be shown, the

strength of Dworkin’s theory lies in the substantive moral foundation that it offers – in

particular its conceptions of human dignity, freedom, and equality –, but its weakness is the

structural account of rights as ‘trumps’, as evidenced by the fact that despite its fame this

idea has never really resonated in legal practice. With regard to proportionality, the opposite

picture presents itself: the doctrine does not give any indication as to its moral foundation –

which is illustrated by the fact that its most famous theoretical account, namely Robert

Alexy’s theory of rights as principles and optimisation requirements, is a formal theory –

while offering a structure that has proven to be so useful that it has become the globally

dominant tool of rights adjudication. It will become clear that it is possible and indeed

important to reconcile and combine the structural strength of proportionality with the

substantive moral appeal of Dworkin’s theory. In order to do so, the essay first offers a

2 Jacob Weinrib (‘When Trumps Clash: Dworkin and the Doctrine of Proportionality’, forthcoming in
Ratio Juris) also aims to reconcile Dworkin and proportionality, but he takes a route that is different
from mine. While this essay focuses more on the values that guide Dworkin’s theory and, as will be
shown, proportionality analysis (human dignity, equality, and freedom), Weinrib focuses on the
structure that reasoning with rights has for Dworkin under his rights-as-trumps approach, and
discovers in Dworkin’s discussions of specific rights issues a range of claims regarding the conditions
under which a right can justifiably be limited that, as he demonstrates, resemble the various stages of
the proportionality test as applied, in particular, by the Canadian Supreme Court. This shows that
even Dworkin himself was not able to maintain his own structural claim that rights are trumps
without a considerable number of qualifications which bring its structure closer to proportionality;
this weakens the plausibility of his claim that their nature as trumps is a general structural feature of
rights.

2
critical reconstruction of Dworkin’s theory which is faithful to his core commitments but

departs from his narrative and terminology where necessary; in other words, the article

aims, in truly Dworkinian spirit, to present his theory in its best light. 3 This leads to the

conclusion that labelling rights as trumps is misleading and fails to capture what is

important about them under Dworkin’s own intellectual commitments (I.). The essay then

demonstrates that interpreting proportionality in light of Dworkin’s theory leads to a richer

conception of the doctrine (II.). Finally, it shows that Dworkin’s theory properly understood

provides us with a much-needed moral foundation of proportionality as based on his

original and important conceptions of human dignity, freedom, and equality (III.); the paper

thus makes the case for the continuing relevance of Dworkin’s theory of rights in the current

‘age of proportionality’ 4.

I. Dworkin’s theory of rights

Dworkin’s theory of rights has evolved over the years and decades of his career. 5 In this

essay, I will work mostly with its last and, I believe, most appealing incarnation as presented

in his books Is Democracy Possible Here? 6 and Justice for Hedgehogs 7. Readers who are familiar

with Dworkin’s earlier writings may be surprised to find that many of the concepts and

ideas that drove his earlier work are of little or no relevance in his last statement on rights.

Generation after generation of jurisprudence students and scholars (including this author)

have pored over ideas such as the distinction between external and internal preferences,

utilitarianism as a background theory which necessitates the existence of rights, the famous

right to equal concern and respect, and the principle-policy distinction. One can still find

traces of them in Dworkin’s late work; however, they do not any longer occupy a prominent

place. In my view – which I just state but do not defend in this paper – his final account of

3 On the nature of interpretation in general and the necessity to present the materials that are
interpreted in their best light see Dworkin, Law’s Empire (Hart Publishing, 1986), ch. 2.
4This is a reference to the title of Alexander Aleinikoff’s influential article ‘Constitutional Law in the
Age of Balancing’, 96 (1987) Yale Law Journal 943.
5For an instructive overview of the development of Dworkin’s thought see Paul Yowell, ‘A Critical
Examination of Dworkin’s Theory of Rights’, 52 (2007) American Journal of Jurisprudence 93.
6Ronald Dworkin, Is Democracy Possible Here?: Principles for a New Political Debate (Princeton
University Press, 2008) (henceforth: Dworkin, IDPH).
7 Ronald Dworkin, Justice for Hedgehogs (Harvard University Press, 2011) (henceforth: Dworkin, JH).

3
rights is superior to the earlier ones in that it focuses squarely on the important issues

relating to the foundational values of human dignity, liberty, and equality; it thus overcomes

and leaves aside the unnecessary complications arising mainly as a consequence of the

unfortunate yet historically understandable entanglement of rights and utilitarianism which

in my judgment – and presumably Dworkin’s own, given that he revised the theory

accordingly – no longer resonate today.

As a preliminary matter, Dworkin makes a conceptual distinction between

constitutional and political rights. 8 Political rights are those (moral) rights which ought to be

protected as (legal) constitutional rights. So what Dworkin calls ‘political’ rights is what

most constitutional theorists call ‘constitutional’ rights: not the specific rights of any

particular constitution but rather those rights which a constitution ought to protect. With

this clarification in mind, let us turn to Dworkin’s account of political rights.

‘Most legitimate acts of any government involve trade-offs of different people’s interests;
these acts benefit some citizens and disadvantage others in order to improve the
community’s well-being on the whole ... But certain interests of particular people are so
important that it would be wrong – morally wrong – for the community to sacrifice those
interests just to secure an overall benefit. Political rights mark off and protect these
particularly important interests. A political right, we may say, is a trump over the kind of
trade-off argument that normally justifies political action.’ 9

As mentioned above, I am going to criticise the usefulness of the term ‘trumps’ when

characterising rights. Nevertheless, it is not difficult to see some appeal in the notion of

rights as trumps based on the above statement. For example, the government might claim,

rightly or wrongly, that closing down my communist newspaper would improve the

community’s well-being on the whole (because, say, workers would find less support for

communist ideas and would therefore not go on strike as often, which would benefit the

economy and thus enable the community to provide better education and hospitals to its

citizens). However, my right to freedom of the press would function as a trump over these

benefits to the community’s well-being and I would therefore continue to be entitled to

publish my newspaper.

How can we identify which rights we have? Dworkin explains:

8 Dworkin, IDPH, 30-1.

9 Ibid., 31.

4
‘Someone who claims a political right makes a very strong claim: that government cannot
properly do what might be in the community’s overall best interests. He must show why the
individual interests he cites are so important that they justify that strong claim. If we accept
the two principles of human dignity that I described in the last chapter, we can look to those
principles for that justification. We can insist that people have political rights to whatever
protection is necessary to respect the equal importance of their lives and their sovereign
responsibility to identify and create value in their own lives.’ 10

So we may say that for Dworkin, people have a right to be treated with dignity; it is the

importance of human dignity that justifies the extraordinary moral force of rights, namely to

block policies that might further the community’s overall best interests. Human dignity, for

Dworkin, consists of two principles: the principle of intrinsic value (an equality principle)

and the principle of personal responsibility (a liberty principle). The principle of intrinsic

value ‘declares the intrinsic and equal importance of every human life’ 11: ‘each human life

has a special kind of objective value. It has value as potentiality; once a human life has

begun, it matters how it goes.’ 12. Among the clearest cases of violations of this principle, and

therefore of rights, are discrimination and genocide 13 because such acts rest on the belief that

some human beings are worth less than others, or even worth nothing.

The principle of personal responsibility ‘holds that each person has a special

responsibility for realising the success of his own life, a responsibility that includes

exercising his judgment about what kind of life would be successful for him.’ 14 Forcing a

person to adopt the state’s official religion or punishing homosexual behaviour as being

against God’s or the majority’s will would usurp the individual person’s responsibility to

realise the success of his own life and would therefore violate the second principle of human

dignity.

Dworkin’s account of human dignity, equality, and liberty adds something

distinctive to the often-cited claim that ‘every person should be treated as free and equal’ by

specifying that equality is best understood as being about the equal and objective

importance of every life and that freedom is concerned with every person’s individual

10 Ibid., 32.
11 Ibid., 37.
12 Ibid., 9.
13 Ibid., 37.
14 Ibid., 10.

5
responsibility to identify value in his or her own life. As so often, the devil is in the detail: in

cases that are less clear than the examples mentioned above, how can we establish whether a

policy respects or violates dignity? Dworkin gives us some guidance. With regard to the

principle of personal responsibility, Dworkin argues that it necessitates two limitations on

the state. First, there are certain foundational decisions which people must be free to make,

such as ‘choices in religion and in personal commitments of intimacy and to ethical, moral,

and political ideals’. 15 Second, a law must not rely on a motive which assumes the

superiority of one ethical ideal over another. 16 Dworkin uses the terms ‘ethics’ and ‘morality’

in an idiosyncratic sense: for him, ethics refers to the duties a person has towards him- or

herself, whereas morality refers to the duties a person has to others. 17 For example, murder

is immoral because it violates a duty one has to others (in particular, the possible victims).

By way of contrast, determining and living one’s life in line with one’s religious convictions

is a duty people have to themselves. While it is acceptable for the state to force people to live

in accordance with collective decisions of moral principle (for example, to refrain from

murdering), its laws must remain neutral between competing ethical ideals. Thus,

moralism18 and (deep) paternalism are unacceptable. 19 Furthermore, even laws that are not

coercive may violate the second principle: for example, a display of the Ten Commandments

in courtrooms would violate the requirement of state neutrality in questions of ethics.

Dworkin is less clear about which rights can be derived from the first principle of

human dignity. He states that due process (fair trial) rights flow from it 20 and says, more

specifically, that treating every person’s life as objectively and equally important requires

refraining from inflicting the special and great harm of punishment on him unless his guilt is

15 Dworkin, JH, 368-9.

16 Ibid., 369.
17 Dworkin, IDPH, 20-21.
18While ‘moralism’ is a term widely used and understood, in Dworkin’s terminology, it would be
more accurate to speak of ‘ethicalism’: forcing people to live in line with the state’s or majority’s
ethical convictions.
19Dworkin, IDPH, 37-8, 73-4. (Impermissible) deep paternalism (where the state imposes its own
preferred set of ethical values on a person in order to improve that person’s life) has to be
distinguished from (permissible) superficial paternalism (which aims to help people achieve what
they actually want) which is found, for example, in seat-belt laws.
20 Dworkin, JH, 371-2.

6
proven. 21 Furthermore, as mentioned above, the right not to be discriminated against

obviously flows from the principle of equal importance. 22 Most of Dworkin’s writing on

equality is about distributive justice, and he has developed a theory – equality of resources –

that is intended to help us structure an argument about which distribution of resources is

fair. 23 But it is unclear to what extent he believed that this is a matter of rights. Given that,

for him, people have a right to be treated with dignity, and given that the first principle of

dignity requires certain kinds of resource distribution, it would seem to follow that people

ought to have constitutional rights to either a fair or at least non-arbitrary distribution of

resources. However, to my knowledge, Dworkin has never made such a claim (possibly

because his writing on constitutional rights is mostly concerned with U.S. constitutional law

and he may have thought that such a view, however appealing as a matter of political

morality, did not ‘fit’ well enough with U.S. constitutional history to qualify as the legally

right answer).

It is remarkable that Dworkin’s writing on rights oscillates between high level

abstraction – in particular his account of the two principles of human dignity – and specific

application – such as his well-known defence of abortion rights,24 or his writing on the

importance of a strong protection of free speech. 25 There is little in his scholarship relating to

the room between high abstraction and concrete application, and it is possible that he

thought that the best way to reason about questions of rights is to apply the abstract

principles directly, without any intermediation, to any given specific problem. It is not my

concern here whether or not this is adequate, but this starkly contrasts with accounts of

rights that place proportionality front and centre because proportionality and related

doctrines are located precisely in that middle space between foundational principles and

concrete application. As far as I can see, the only structural point that Dworkin makes about

rights is, precisely, that they function as ‘trumps’ over the community’s well-being. But

while we were able to see the intuitive appeal of that statement, if we look at it again in light

21 Ibid., 372.
22 See n 13 and accompanying text.
23 Dworkin, JH, 356-363.
24Dworkin, Freedom’s Law (Oxford University Press, 1996), ch. 1-4; Dworkin, Life’s Dominion: An
Argument about Abortion, Euthanasia, and Individual Freedom (Knopf, 1993).
25 Dworkin, Freedom’s Law (above n 24), ch. 7-10.

7
of a more nuanced understanding of his theory, the claim turns out to be misleading and

unhelpful.

It has some appeal in the case of the above-mentioned foundational decisions whose

protection flows from the principle of personal responsibility. For example, we may say that

the state must not prohibit the foundational decision of whether or not to have an abortion

even if by doing so, it could enhance population growth and therefore greatly further the

community’s well-being; in this sense the right to abortion functions as a trump. But in the

other categories that Dworkin mentions, talk of rights as trumping the community’s well-

being is problematic. Let us first consider the prohibition of (deep) paternalism (which flows

from the second principle of human dignity). Paternalism is motivated by the desire to

improve the agent’s life; for example, a prohibition of sodomy might be motivated by the

desire to lead gay people to lead less sinful lives. But it is unhelpful to frame such a policy as

being about the community’s well-being (which rights can then trump): paternalism makes a

claim to improve the affected person’s well-being; it is not necessarily concerned with any

effects on the community. Second, with regard to rights flowing from the first principle of

dignity, again, it is not obvious why these rights should be seen as trumping the

community’s well-being. For example, it might be the case that dignity requires the right to a

free education for all. Of course, someone who disputes this could try to argue that the

community’s well-being would suffer under such a policy (to which Dworkin could then

respond that the right to education trumps any such concerns). But an argument against free

education need not be about the community’s well-being; it could be, for example, about

justice, claiming that collecting the resources needed to fund education would for some

reason be unjust towards the rich.

So I think that it is unfortunate under Dworkin’s own commitments to create an

opposition between rights and the community’s well-being. We might consider fixing this

problem by regarding rights as trumping not only the community’s well-being but any value

or consideration that conflicts with them: justice, well-being, utility, and so on. This,

however, would amount to saying that rights must prevail over a policy that is justified on

some other ground and would, in other words, simply point to the largely uncontroversial

fact that policies need to comply with rights in order to be legitimate.

8
 To conclude, I do not think that the claim that rights are trumps is fortunate. It had

its role to play in the earlier versions of Dworkin’s theory 26 but does not point to an

important or insightful feature of rights in his later work; it should therefore be dropped.

While less catchy, Dworkin should have claimed instead that the foundational right that

people have is to be treated with dignity, and that this requires, first, that every policy

respect the equal and objective importance of their lives and, second, that every policy

recognise everyone’s personal responsibility for realising the value of his or her life.

II. Proportionality and Dworkin’s theory of rights

Proportionality-based judicial review usually employs a two stage analysis in order to

answer the question of whether a given policy violates rights. The first question is whether

the policy under consideration interferes with (limits, restricts) a right. The second is

whether the interference is justified, and usually this is the case if it is proportionate. The

first stage is of little practical relevance because of the phenomenon that George Letsas has

called ‘rights inflation’: 27 the scope of (prima facie) rights has become so broad in legal

practice that it is difficult to find a case where the reviewing court concluded that the policy

at stake did not interfere with a right and therefore was not in need of a justification. The

consequence of this is that most if not all of the analytical work takes place at the second

stage, where the test is proportionality. 28 Proportionality has four steps. 29 First, the policy

26In his earlier work, he assumed – in line with the prevailing moral discourse at the time – that
policies were usually justified on utilitarian grounds; rights came in to correct the occasionally
impermissible aggregation of preferences that utilitarianism proposes. In that framework it made
sense to say that rights trump utilitarian justifications.
27George Letsas, A Theory of Interpretation of the European Convention on Human Rights (Oxford
University Press, 2007), 126.
28This has led Webber to claim that ‘the entire constitutional rights-project could be simplified by
replacing the catalogue of rights with a single proposition: The legislature shall comply with the
principle of proportionality.’ (The Negotiable Constitution: On the Limitation of Rights [Cambridge
University Press, 2009], 4). One might add that proportionality does not apply to all rights, though:
some rights – for example the right to freedom from torture and inhuman or degrading treatment or
punishment – are absolute, meaning that any interference with them amounts to a violation and
proportionality analysis does not apply. It is possible to reconcile absolute rights with proportionality
by arguing that an interference with an absolute right is always disproportionate because it always
imposes a burden on the right-holder that is more than he can be expected to bear. See Kai Möller,
The Global Model of Constitutional Rights (Oxford University Press, 2012), 139-140.
29Robert Alexy, A Theory of Constitutional Rights (Oxford University Press, 2002), 66; Mattias Kumm,
‘Political Liberalism and the Structure of Rights’, in Pavlakos (ed.), Law, Rights and Discourse: The Legal

9
that interferes with the right must pursue a legitimate goal (legitimate goal stage). Second, it

must be a suitable means of achieving the goal at least to some extent; put differently, there

must be a rational connection between the policy and the goal (suitability or rational

connection). Third, there must not be a less restrictive and equally effective alternative

(necessity; least restrictive means). Fourth and finally, the burden imposed on the right-holder

must not be disproportionate; establishing this requires balancing the seriousness of the

interference with the right against the importance of the competing interests (proportionality

in the strict sense; balancing).

On a superficial analysis, it seems clear that proportionality, by regarding rights as

balancing requirements, stands in stark contrast to a conception of rights that regards them

as trumps. 30 But I have already shown in the previous section that Dworkin’s famous label

does little to illuminate the structure of rights and that it is more adequate to summarise his

theory as requiring that every person’s right to human dignity has to be protected, which

means that every human life must be treated as having objective and equal importance and

that every person’s sovereign responsibility for the success of his or her life must be

respected. My argument in this section is that there is no conflict between his conception of

Philosophy of Robert Alexy (Hart, 2007), 131, 137; Kai Möller, ‘Proportionality: Challenging the Critics’,
10 (2012) International Journal of Constitutional Law 709, 711. There are variations as to the precise
formulations of the test which are however not relevant here.
30It is remarkable that Dworkin seems to never have written about proportionality as a doctrine of
constitutional law (maybe because it is not used in the U.S., which is the jurisdiction that he was most
familiar with and wrote about regularly). In Is Democracy Possible Here? [above n 6, 48-9] there is a
brief engagement with limitations clauses in the European Convention on Human Rights; Dworkin
simply states that ‘[t]hese qualifications were the result of political compromises that served to
reassure countries hesitating to embrace the convention’ – a statement that is rather simplistic and
inadequate as a general account of limitation clauses; this is followed, in the next paragraph, by a
claim to the effect that an abstract right needs to be refined before it can be considered absolute. I do
not think that a lot of insight about proportionality and limitations clauses can be gained from these
cursory claims; rather, their ad hoc nature indicates that Dworkin never took these concepts seriously.
Dworkin did occasionally discuss the idea of balancing; but usually he had in mind balancing in the
sense of a utilitarian cost-benefit calculation which he rejected as not doing justice to questions of
rights, a point with which I agree: balancing in constitutional rights law is concerned with a conflict of
interests which must be resolved through moral reasoning; thus, balancing refers not to utilitarianism
but must be used in the sense of ‘balancing as reasoning’ (see below fn 37 and accompanying text).
See, for example, his ‘It is absurd to calculate human rights according to a cost-benefit analysis’, an
opinion piece in the British Guardian, available at
https://www.theguardian.com/commentisfree/2006/may/24/comment.politics (last accessed 24 April
2017).

10
rights properly understood and proportionality, and that, indeed, interpreting the four

stages of the proportionality test in line with the requirements of human dignity leads to a

richer understanding of the test.

1. The legitimate goal stage

At the first stage of the test, the legitimacy of the goal pursued by the policy in question

must be assessed. The abstract formulation of the proportionality test does not give any

guidance as to what makes a goal legitimate or illegitimate; it is therefore necessary to

interpret it in line with an attractive conception of rights. On the assumption that Dworkin’s

theory provides such a conception, it can directly be applied here. As explained above,

Dworkin claims that a law must not rely on a motive which assumes the superiority of one

ethical ideal over another; this makes moralism and (deep) paternalism impermissible. Thus,

a goal is illegitimate if it assumes the superiority of one ethical ideal over another. 31 The

textbook example of such a goal would be a law that prohibits homosexual sex on the

grounds that such sex is ethically worthless, against God’s will, or harmful to the person

engaging in it. Similarly, it would be impermissible to cite as the goal of a policy banning

headscarves in public universities the view that an Islamic lifestyle in general, or a particular

interpretation of Islam which makes the headscarf necessary, is ethically unappealing and

should therefore be prohibited or discouraged.

Of course, most cases are not as clear-cut as the examples I have just given; this raises

the question of whether Dworkin’s principle of personal responsibility is also helpful for the

resolution of more complex cases. I think that it is, and will use the European Court of

31 A parallel point was first made by Kumm (above n 29), 142-8. See also my discussion of the
legitimacy of goals in Möller (above n 28), 183-93. I do not mean to exclude the possibility that a goal
could also be illegitimate by virtue of being incompatible with the first principle of dignity, but it will
fortunately be very rare in legal practice for a state to put forward such an argument. For example, a
policy that disadvantages a minority ‘in order to give expression to their lesser value’ would be such
a case. In practice, even obviously discriminatory practices are usually defended by relying on a goal
that is in principle legitimate but that does not ultimately justify the policy at stake. For example, in
the ‘gays in the army’ case of Smith and Grady v. United Kingdom (29 (2000) E.H.R.R. 493), the U.K.
argued not that the goal of the policy was to express the view that homosexuals are ‘worth less’;
rather it argued that allowing homosexuals to serve in the army would lead to tensions within the
army and therefore to a threat to national security. Since protecting national security is obviously a
legitimate goal, the ECtHR was right to accept the goal put forward by the U.K. and to let the policy
fail at a later stage of the proportionality test.

11
Human Rights’ judgment in Stübing v. Germany, 32 which concerns incest, as an illustration.

While incest is an activity which many people will ethically disapprove of quite strongly, it

is not obvious whether its prohibition can be justified, given that the moralistic or

paternalistic goal of stopping people from living ethically worthless lives does not qualify as

legitimate under Dworkin’s principle of personal responsibility. The fact that Germany did

not even try to rely on this reason is an indicator of the appeal of Dworkin’s principle:

banning an activity because of ethical disapproval would widely be seen as simply

unacceptable and an unjustifiable intrusion into the personal life of the right-holder. The

problem, however, is that even though the state may not openly rely on ethical disapproval,

the motives that it cites instead may still violate the principle of personal responsibility for

the following reasons. First, the goal, while not moralistic or paternalistic, may nevertheless

and on closer inspection turn out to violate the principle of personal responsibility for other

reasons; and second, the goal cited may serve as a pretext for a moralistic or paternalistic

goal.

Stübing is an example in point. In the case, Germany had argued that its ban on incest

was justified because it pursued the goals of, first, protecting the family (the risk being an

‘inversion of social roles’ within the family structure); second, the protection of the weaker

partner in a relationship (the risk being a structural imbalance which posed a threat to the

weaker partner’s general and/or sexual self-determination); third, preventing genetic

damage to children; and fourth, maintaining a societal taboo against incest. 33 The assumed

need to maintain a societal taboo against incest is an instance of a goal that, while not openly

moralistic or paternalistic, still (arguably) violates the second principle of dignity: a societal

taboo against a particular behaviour means, broadly, that the respective behaviour (in this

case, incest) is not seriously considered as part of the menu of options available to a person;

a coercive measure designed to maintain an existing taboo can therefore be considered as an

indirect form of thought control which, in turn, can be seen as violating the principle of

personal responsibility by denying the individual the ability to reach his own judgment

about the value of a particular activity.

32 Stübing v. Germany, (2012) 55 E.H.R.R. 24.

33 Ibid., paras 46-50.

12
 The goal of preventing genetic damage to children can be seen as an example of a

goal which is (arguably) legitimate but serves as a pretext for an illegitimate goal. As a

preliminary point, it is not clear that the goal should be considered legitimate because it is

arguable that it is better for children with a genetic defect to be alive than not to be alive at

all. But assuming that the goal is indeed legitimate, the possibility exists that it serves as a

pretext for the moralistic motives of the state because in other contexts, the risk of genetic

damage to the child is never considered to be a valid reason to prohibit sex and/or

procreation; 34 for example, it is uncontroversial that disabled people have the right to

procreate even if their children are likely to be disabled, too.

Structurally, there are two ways of dealing with this problem. The first is to focus on

the subjective motivation of the law-makers and inquire into their true motives. This,

however, creates unanswerable questions because those motivations will often not be

accessible; it also sets unfortunate incentives for lawmakers to hide and possibly lie about

their true motives. The second and preferable way is to focus on the objective justifiability of

the policy. Thus, in the Stübing example, preventing genetic damage to children would be

considered to be a legitimate goal; however, the relevance of the fact that other forms of

procreation leading to children with genetic damage are not prohibited would be considered

at a later stage of the test. 35

The above remarks are not meant as a comprehensive discussion of the intricacies of

Stübing; rather, their point is to show how Dworkin’s second principle of human dignity can

successfully be integrated into the interpretation and application of the proportionality test

and enrich the proportionality analysis.

2. The balancing stage

If a policy passes the legitimate goal stage, the next stages of the proportionality assessment

concern its suitability, necessity, and proportionality in the strict sense. For reasons of

explication of my argument, this section will focus on the balancing stage. It is the final and,

usually, decisive stage of the proportionality analysis and requires a balancing exercise

34 This argument was made by the applicant; see ibid., para. 36.
35This question could be considered at the balancing stage or arguably at a separate stage assessing
the coherence of the policy. I have argued in favour of adding a coherence stage to the proportionality
assessment in Möller (above n 31), 125-6.

13
between the right-holder’s interests and the competing public or individual interests; the

overall question is whether the policy places a burden on the right-holder that is

disproportionate, that is, more than he or she can legitimately be expected to bear.

One of the issues pointed out by authors who are critical of balancing relates to the

lack of clarity as to what the concept means in constitutional rights law. 36 It seems obvious

that properly understood it neither points to a mechanical exercise of quantification and

comparison nor a utilitarian or consequentialist kind of reasoning; neither form of balancing

would do justice to questions of rights. 37 Rather, the starting point for any theory of

balancing must be that balancing simply refers to the need to make a moral argument as to

which of the two (or more) competing interests takes priority in a given case. 38 This

interpretation is preferable not only from a moral perspective; it is also compatible with the

semantic meaning of the term: for example, when we speak of the need to find a ‘balanced’

solution to a moral problem, this means that we are looking for a solution which attaches the

correct weight to all relevant considerations and is therefore morally defensible. But if

balancing simply refers to the need to construct a moral argument about the competing

values, which value or values should guide it?

I wish to suggest that Dworkin’s principles of human dignity can be interpreted to

provide us with a theoretical starting point if not a comprehensive theory of balancing. In a

nutshell, I will argue that the balancing stage of the proportionality test is concerned with

ensuring that the policy at stake respects the equal importance of the right-holder’s life.

Thus, balancing is shorthand for assessing the compliance of a policy with the first principle

of human dignity.

This can be explained in the following way. To say that we have to ‘balance’ the

burden on the right-holder against the importance of the competing right or public interest

refers to the imperative of designing policies in a way that is ‘balanced’ – that is, that

distributes burdens and benefits in a way that is compatible with the principle of equality.

36 Grégoire Webber (above n 28), 88-9; Stavros Tsakyrakis, ‘Proportionality: An assault on human
rights?’, (2009) 7 International Journal of Constitutional Law 468; Jürgen Habermas, Between Facts and
Norms: Contributions to a Discourse Theory of Law and Democracy (MIT Press, 1998), 259.
37 George Letsas, ‘Rescuing Proportionality’, in Cruft, Liao, and Renzo (eds.), Philosophical Foundations
of Human Rights (Oxford University Press, 2015), 316, 325-6. See also above n 30.
38 Möller (above n 29), 720-1; Möller (above n 31), 140.

14
Conversely, to say that a policy is disproportionate (in the strict sense) means that the state

has attached too little weight to the interests of the right-holder: if it had attached the correct

weight, it would have noticed that the burden on him or her was too high and therefore

disproportionate. But to attach too little weight to someone’s interests means to fail to treat

his interests as equally important as everyone else’s interests, and therefore fails to

acknowledge the equal importance of his or her life.

Let me give as an example the recent case of S.A.S. v. France, 39 where the European

Court of Human Rights decided that the French ban on the full-face veil was compatible

with the European Convention on Human Rights. The goal pursued by the ban, which the

Court found to be legitimate, was the protection of ‘the ground rules of social

communication’ and the conditions of ‘living together’. 40 One of the issues at stake was

whether a ban on the full-face veil was a disproportionate burden on those women wishing

to conceal their faces for religious reasons. If we assume (with the dissent) that it was, then

we are saying that France asked for ‘too much’ from the affected women and that,

correspondingly, it attached too little weight to the legitimate interests of those women.

Thus, it failed to treat them as someone whose lives have an importance that is equal to

everyone else’s.

Admittedly there is little direct support in Dworkin’s writing for my claim that the

first principle of dignity requires what constitutional rights lawyers usually refer to as

balancing. Most of his writing about specific rights issues is about the second principle of

dignity: think of his well-known defences of abortion, hate speech, or euthanasia. When he

writes about the first principle, he is mostly concerned not with rights but with theories of

distributive justice (in particular, taxation). 41 However, there is no reason why the first

principle should be entirely or even predominantly about fair taxation, and Dworkin himself

certainly did not intend it to be; for example, he claims in Justice for Hedgehogs that the right

to due process flows from the principle of equal importance, 42 and he states in Is Democracy

Possible Here? that certain forms of torture, discrimination, and genocide violate the first

39 S.A.S. v. France, (2015) 60 E.H.R.R. 11.

40 Ibid, paras 121-2.
41 See above n 20 - 23 and accompanying text.
42 Dworkin, JH, 371-2.

15
principle. 43 As mentioned before, my goal in this article is not to engage in Dworkin

exegesis, but rather to reconstruct his theory in a way that is coherent and shows it in its best

light. To regard the reach of the first principle of dignity as restricted to questions of

resource distribution and/or blatant denials of equal worth strikes me as arbitrary. 44

Therefore, it is not only possible but indeed necessary to hold that the principle of equal and

objective value demands a proportionate, ‘balanced’ distribution of burdens and benefits;

and this is precisely what under proportionality is assessed at the balancing stage.

In a slightly different context, Letsas has considered an argument which points to an

important objection to my approach here. He writes about the famous Hatton case, 45 which

was about a policy allowing night flights at Heathrow airport which affected the sleep of

some of the residents in the area. Letsas is critical of protecting the interest in good sleep as

part of Article 8 ECHR and balancing it against the competing interest in economic

development. He accepts that the interest in good sleep is an important interest and that

justice and equality (for our purposes: Dworkin’s first principle of human dignity) might

have something to say about its adequate protection. But:

‘Perhaps the Heathrow night flights scheme did not allocate the various resources in play
(sleep, economic benefits, employment) in a way that treats people as equals under some
egalitarian theory of distribution. Perhaps the applicants have a solid claim based on
distributive justice to request the government to restrict or abolish night flights, relocate the
airport, or compensate the residents. But these are not matters falling within the law of
human rights. We cannot inflate the concept of human rights so much that it covers the
whole realm of justice. Human rights would then lose their distinctive moral force.’ 46

Applied to the issue that is my concern at this point, the problem can be presented in the

following way. If we accept, with Dworkin, that (1) rights are about dignity, (2) dignity is,

among other things, about treating people’s lives as equally important, and (3) this means

that burdens and benefits must be distributed in a just way, then it would seem to follow

43 Dworkin, IDPH, 37.

44 To avoid misunderstandings: I agree with Dworkin that the first principle of human dignity
prohibits the acts Dworkin mentions. Under contemporary human and constitutional rights law,
these acts would usually be regarded as absolutely prohibited, meaning that proportionality analysis
would not apply or (alternatively) that any interference with the respective right would be regarded
as disproportionate; on this see above n 28; my point is only that the first principle should have a
wider application than Dworkin seems to imply.
45 Hatton v. U.K., (2003) 37 EHRR 28.
46 Letsas (above n 27), 129.

16
that the scope of rights becomes so broad as to cover the whole domain of justice. This

would be problematic because it is widely and, to my mind, correctly accepted that not

every policy which is unjust also violates rights.

I believe that there is indeed a structural weakness in Dworkin’s writing about rights

and human dignity because he does not address, let alone resolve, this important point; as

mentioned above, as far as I can see, he does not make any systematic attempt to work out a

theory of the relationship between the first principle of human dignity and the scope of

rights. However, the correct answer to this problem is not to radically ban all considerations

relating to balancing from the scope of rights, as Letsas seems to suggest. 47 Rather, another

and more appealing solution is to distinguish between justice and rights in a different way,

namely by claiming that rights require only that a policy be based on a reasonable, as

opposed to the one correct, conception of justice, and that therefore, a distribution of

burdens and benefits is proportionate as long as it is reasonable. 48 This approach finds

considerable support in the case law of courts that rely on proportionality, which regularly

make statements to the effect that their task is not to establish whether the legislature

designed the best possible (or ‘just’) policy but rather only to review whether the burden on

the right-holder is justifiable or reasonable. 49

3. The suitability and necessity stages

The suitability and necessity stages, too, can be explained as flowing from Dworkin’s first

principle of dignity. Let us first consider the case of suitability. An unsuitable policy places a

burden on the right-holder but does not contribute to the achievement of its goal. This can

easily be shown to violate Dworkin’s first principle of human dignity because treating a

person’s life as (equally) important surely requires abstaining from burdening him for no

reason. A similar consideration applies in the case of necessity; an unnecessary policy

47 Letsas concludes his chapter by stating: ‘If we want to be accurate we have to say that in cases
like Hatton, states’ margin of appreciation is infinite and in cases like Dudgeon states have no margin
of appreciation whatsoever.’ (ibid., 130). I interpret this as meaning that any balance of interests
would be acceptable in Hatton, which is just a different way of saying that no balancing should take
place at all.
Mattias Kumm, ‘The Idea of Socratic Contestation and the Right to Justification: The Point of Rights-
48

Based Proportionality Review’ 4 (2010) Law & Ethics of Human Rights 141; Möller (above n 31), 117-123.
49 Möller (above n 31), 200-204.

17
burdens the right-holder to a greater extent than necessary because the legitimate goal could

be achieved to the same extent through a less burdensome policy. Again, respect for human

dignity requires choosing the less burdensome policy because placing an unnecessary

burden on the right-holder fails to treat his life as (equally) important.

4. Conclusion

Dworkin’s theory of rights properly understood is compatible with proportionality, and

furthermore, it illuminates the meaning and interpretation of the different stages of the test.

One of the by-products of integrating Dworkin’s theory with proportionality is a challenge

to conventional wisdom according to which there exists a contrast between liberty rights

(such as the rights to life, freedom of religion, association, expression, and the right to

private life) and equality rights (in particular the right to non-discrimination). The previous

sections have shown that reasoning with so-called liberty rights requires an appreciation of

liberty as well as equality: the legitimate goal stage of the proportionality analysis is

concerned with liberty in that it ensures that the personal responsibility of the right-holder is

respected; the following three stages (suitability, necessity and balancing/proportionality in

the strict sense), however, are, on the analysis presented in this paper, concerned with

equality because a ‘balanced’ or proportionate policy distributes burdens and benefits in a

way that respects every affected person’s equal importance; conversely, a disproportionate

policy, by imposing a burden on the right-holder that is unsuitable, unnecessary or

disproportionate, fails to treat the right-holder’s life as equally important.

III. The importance of Dworkin’s contribution in the age of proportionality

So far I have offered a critical reconstruction of Dworkin’s theory of rights, and I have

shown that when viewed in its best light it is compatible with proportionality and can

indeed inspire a richer and more appealing interpretation of the four stages of the test. This

section will offer some additional reflections on the importance of Dworkin’s theory for the

current debates about proportionality and rights.

18
 Proportionality has been immensely successful as a legal doctrine: it was invented not

by scholars or theorists of rights but by judges, 50 and it had become a globally successful tool

of rights adjudication long before the first comprehensive theory of it was published. 51 Thus,

its success is primarily based on its practical usefulness for judges deciding cases about

constitutional rights: it offers them a helpful structure by splitting up the question of

whether a policy violates rights into four sub-questions that can be analysed separately. This

is no small achievement. But while the role of proportionality as a tool and a legal doctrine

has certainly been extremely useful, for a long time there was no agreement on, in fact not

even a conversation about, the values on which proportionality-based judicial review is

based. The fact that the first major and most famous account of proportionality-based

conceptions of constitutional rights, namely Robert Alexy’s theory of constitutional rights as

principles and optimisation requirements, is a formal theory, that is, one which explicitly

rejects the possibility of a substantive – value-based – theory of constitutional rights, 52

underlines this feature of the proportionality discourse.

In recent years a couple of substantive theories have been proposed. Mattias Kumm

has argued that proportionality-based judicial review protects every person’s right to

justification by instructing judges to engage in Socratic contestation in order to determine

whether an act by a public authority is justifiable in terms of public reason. 53 I have

contributed to the debate by arguing that the point of constitutional rights is to ensure that

any act which limits a person’s ability to live her life according to her self-conception is

justifiable to her; that this justification is successful if the policy at stake rests on a reasonable

specification of the spheres of autonomy of equal citizens; and that proportionality properly

understood assesses precisely this question. 54 We ought to add Dworkin’s conception of

rights and dignity to these existing attempts to make sense of proportionality. The rough

outline of a conception of constitutional rights that emerges from my reconstruction of his

50 On the history of proportionality, see Alec Stone Sweet and Jud Mathews, ‘Proportionality
Balancing and Global Constitutionalism’, (47) 2008-9 Columbia Journal of Transnational Law 72, sections
III and IV.
51 Namely Robert Alexy’s (above n 29).
52 Ibid., 11.
53 Kumm (above n 48).
54 Möller (above n 31), passim.

19
theory is the following. The foundational right that everyone has is to be treated with

human dignity. Thus, any law or other act by a public authority must respect his equality

(more precisely: the equal and objective importance of his life) and his freedom (his personal

responsibility for the success of his life). The justification of the principle of proportionality,

then, lies in its ability, if properly interpreted, to ensure the protection of human dignity and

its two sub-principles. At the legitimate goal stage, any justifications that are incompatible

with the value of personal responsibility are discounted. The following three stages

(suitability, necessity and balancing) assess whether the policy distributes burdens and

benefits in a way that can reasonably be regarded as compatible with the person’s equal

importance.

The existing theories of the point and purpose of constitutional rights are not

mutually exclusive and not even necessarily in competition with each other; rather, they

highlight different facets of the values underlying constitutional rights. Kumm’s emphasis is

on the fact that all state action must, as a matter of rights, be justifiable in terms of public

reason; thus, his focus is on the structural and institutional conditions of legitimate

authority; and while he believes that a conception of public reason as well as a more fully

worked-out account of the right to justification will have to say something about freedom

and equality, he does not provide a conception of these values. In my previous work, I have

attempted to spell out more concretely the meaning of equality and, in particular, autonomy

as structural building blocks of a theory of rights, but I have stopped short of providing a

moral grounding of those values. Dworkin’s emphasis, by way of contrast, is directly on the

substantive values that ground rights and legitimacy, that is, the two dimensions of human

dignity. Furthermore, there are links between the right to justification (that grounds rights in

Kumm’s and my accounts) and human dignity (which grounds rights for Dworkin): Rainer

Forst (who first proposed the idea of a right to justification as the basis of human rights) has

claimed that human dignity implies first and foremost that every person be treated as

possessing a right to justification. 55

Thus, Dworkin’s theory of rights should not be sidelined as one that is in

competition with proportionality and therefore, at least for the time being, on the losing side

Rainer Forst, ‘The Justification of Human Rights and the Basic Right to Justification: A Reflexive
55

Approach’, 120 (2010) Ethics 711, 717.

20
of history and of limited interest and relevance. Rather, his insights should be integrated

with the currently dominant, proportionality-oriented discourse about rights. The

importance and relevance of his work lies in its ability to, first, guide the interpretation of

the different stages of the proportionality test and, second, provide a distinctive account of

the moral foundation of proportionality-based approaches to rights as being about human

dignity, liberty, and equality.

21