Nikiforos Panagis: Putting Balancing in the Balance


The importance of the principle of proportionality in assessing the lawfulness of a measure that interferes with individual rights is reflected in the fierce debates it has generated. The present essay will focus on the main issues that are at stake between proponents and critics of proportionality, in an attempt to assess the doctrine in comparison with other proposed methods of adjudication that reject it. Although there are probably as many conceptions of the proportionality analysis as are its defenders, it will be argued here that none of the accounts of the principle widely encountered in theory and in practice provides a method of adjudication worth pursuing. To begin with, it will be shown that if proportionality is viewed as a mere technical device, independent from any moral reasoning, then it is faced with the unresolvable issue of the incommensurability of the items that are to be balanced. If, further, it is accepted that moral reasoning does make part of the proportionality analysis, it will be argued that the premises on which the principle of proportionality is based do not allow for a solid account of human rights and point towards an implausible path of moral reasoning. Finally, it will be pointed out that even if, instead, proportionality is supported with a sound theory of human rights, its use by the judiciary for the review of legislation distorts the notion and purpose of judicial review and finds itself at odds with fundamental conceptions of the institutional balance needed in a representative democracy. For the purposes of this essay proportionality will generally be regarded as neatly defined by Judge Barak, namely as being based on two components: the first is the legality of the purpose pursued (‘legitimate aims’) and the second is the legitimacy of the means, which in turn is divided in three sub-components, demanding that the means be suitable, necessary and stricto sensu proportionate to the aim pursued[1].

According to its most ambitious proponents, proportionality is nothing but a mere technical device that allows judges to detach themselves from moral questions (which they are incapable and improper of answering) and to arrive to conclusions by a quasi-mechanical evaluation of the facts provided to them in each case: they feed the imaginary pair of scales with the facts describing the damage suffered to the individual by the contested public measure and the benefit to the community springing from the same measure and they come up to a verdict as to its validity[2]. The argument most commonly advocated against such a notion of proportionality concerns the incommensurability of the values put into the balancing scales. This argument attempts to torpedo the internal structure and architecture of the proportionality analysis and runs that, in the absence of a common measure, proportionality cannot lead to a rational and objective answer (as a technical device, a pair of scales, would), but merely to assist in making a subjective choice based on moral reasoning[3]. In other words proportionality, in its purest technical form as advocated by Beatty, demands that the judges weigh the immeasurable (the intensity of the interference with the principle protecting the right in question) with the unspecified (the importance of satisfying the competing principle, usually the public interest)[4]. Judge Barak, aiming to find such a common metric that will enable comparison, uses the notion of ‘the marginal social importance’ that can be attached to each of the conflicting principles or interest at the point of conflict, assessing the importance to society of the benefit gained by the realisation of the measure in question, as opposed to the importance to society of preventing the limitation of the opposing human right[5]. The assignment of such importance to each right is, in turn, dependent on the fundamental perceptions of the relevant society: not all rights are of equal social weight, despite the fact that they possibly enjoy equal constitutional status[6]. Similarly, the importance of the achievement of the intended purpose corresponds to the substance and the pressing nature of the social interest that will be satisfied by such an achievement.

Barak’s model, albeit detailed and intelligently structured, can easily be criticised as depending on considerations that are themselves incommensurable[7]. Indeed, Alexy acknowledges that the use of the weight formula[8] requires a judgment that will accurately determine the highly dispositional notions that form part of the equation; in other words a judgment that will allow us to determine, for example, how ‘great’ the health risks from smoking that underpin the measure of tobacco labelling are and, similarly, how ‘serious’ the interference with the right to conduct a business is[9]. As Alexy admits, this judgment raises a claim to correctness[10], and in this sense it derives from moral reasoning, from a moral argument as to which of the competing interests takes priority in the case at hand[11]. This is what Möller calls ‘balancing as reasoning’, as opposed to mere ‘interest balancing’[12]. First of all, this is arguably a misleading use of the term ‘balancing’: indeed, moral reasoning of any sort involves considerations in favour and against an argument for its evaluation and thus the conclusion reached might be regarded as the result of balancing[13] – but in the context of the proportionality analysis ‘balancing’ has the far narrower and stricter sense of the final step of the proportionality analysis (which is preceded by the equally technical steps of legitimate aims, suitability and necessity) and, in that sense, it is a term of art. What is more important for present purposes, however, is to point out that it is largely undisputed by both critics and defenders of proportionality that proportionality is not a mere technical device that automatically quantifies factual inputs and transforms them to legal outputs (as perhaps Beatty would want it to be[14]), but that it entails a moral analysis. Thus, incommensurability cannot be a vice in itself, given that in most legal systems judges are par excellence required to ‘balance the unbalanceable’[15], to reconcile incommensurable issues in many fields of law (such as criminal law, tort law), and they are expected to do so by a certain reasoning. The question, then, turns to whether proportionality provides the structured framework that would ideally guide a judge through her reasoning process, or if there may be a better framework for this purpose. Bearing in mind that proportionality deals with adjudication on issues of fundamental rights, the answer to such a question will inescapably depend on the account of the nature of human rights that underpins the overall structure of the principle of proportionality.

Summarising the conception of rights under the proportionality analysis, Gardbaum argues that they are ‘important prima facie claims (usually, but not necessarily, against the state) that can, in principle, be limited or overridden by non-constitutional rights claims, specifically by certain conflicting public policy objectives’[16]; they are ‘important prima facie, but nonetheless rebuttable’[17]. In this sense, proportionality is based on a principal of ‘definitional generosity’[18]: there is no reason to narrowly define the scope of interests protected as a right, and thus all acts by public authorities affecting individuals are eligible to be put under the scrutiny of the proportionality test[19]. At the same time, the normative implications of such a specification of a right are similarly narrow, for the constitutional right’s guarantee is merely a guarantee against disproportional state action, not a guarantee against unacceptable state action irrespective of what proportionality might otherwise suggest[20]. The two limbs of such a conception of rights – the ‘definitional generosity’ of rights and the fact that they are rebuttable – are both heavily criticised. Starting from the criticism on the latter limb (rights being rebuttable) will help better understand the difficulties in accepting the former (the broad definition of rights).

As proponents of proportionality advocate, application of the proportionality analysis would accept that nearly every public policy claim is an appropriate goal to be achieved, to the detriment of the conflicting human right – in other words, (almost) all goals pursued by policies are obviously legitimate, and thus meet the threshold of the first step of the proportionality test, namely the ‘legitimate aim pursued’[21]. Nevertheless, it may be argued that such a conception fails to secure one of the major commitments of most modern liberal conceptions of political justice, the claim to antiperfectionism, which demands that public authorities have no jurisdiction to prescribe what the ultimate orientations and commitments of an individual should be, and that they cannot impose any conception of a good life[22]. Under this argument, one of the basic functions of rights is to protect individuals from impositions relating to how they should live their lives, to shield them against any external determination of what the ‘happiness’ that they may pursue[23] should consist of. Thus, furthering a particular perfectionist ideal is not a reason to weigh when assessing the proportionality of a measure: it is an ‘excluded reason’, one that does not constitute a legitimate purpose that can justify any infringement of individual liberties. A good application of the proportionality test would arguably reject any illicit justifications as legitimate goals in the first step of proportionality analysis[24], although this caveat of ‘excluded reasons’ is considered by some defenders of proportionality as important enough to form part of a theory distinct from the doctrine of proportionality[25]. Indeed, what is missing from the analysis of the ‘excluded reasons’ doctrine is the method by which such excluded reasons are to be found: certainly the method cannot be that of balancing, since our process of reasoning is only in the first step of the proportionality analysis, unless by ‘balancing’ we do not consider the term of art used in the proportionality context, but the assessment of a moral argument[26].

In the same vein, the only way to avoid the slippery slope of consequentialism that the utilitarian premises of proportionality might lead us to and to prevent the political community from treating individuals as means for the accomplishment of collective ends is by admitting that ‘clearly [sic] … some ends [which emerge from the self-governing decision-making process] are just not reasonably or plausibly compelling ones’[27] for allowing us to interfere with the rights of an individual. Here, again, the plea for a ‘correct application’ of the proportionality analysis[28] points towards the right direction – that these ends be dismissed in the first step of proportionality analysis – but conceals the method by which the adjudicator should define the aims that are excluded from the ambit of the regulatory power of the state. This, however, is the most important part of the adjudication process in such cases, namely to determine which factors are relevant or important and, thus, which reasons deserve to be put in the balance in the first place. Proportionality analysis fails to provide a clear reasoning process in that respect, not only because it is in practice applied in a more gestalt rather than in a seriatim manner by many jurisdictions[29] but, more importantly, because it saves the method of ‘balancing’ for the last step of the test, logically long after the legitimacy of the aim pursued is upheld as a matter of principle.

The importance of this step and the difficulty in determining the relevant factors that are to be assessed in any situation that demands decision-making should not be underestimated, simply because in practice courts tend to accept almost all purposes sought by the government as legitimate[30]. On the contrary, it is in this step where the government will be called to provide sound reason-based justifications for the claim that its action fell within its regulatory and discretionary power, and to confront the counter-claim that of the individual that the government was obliged (not) to act in the case in question. Thus, the argument that the principle of proportionality institutionalises the accountability of public authorities before independent tribunals that hold public hearings on the application of every individual[31] holds with equally effective results for every form of judicial review.

Space does not permit the elaboration of a sound theory of adjudication that would ideally play the role of guiding the moral reasoning of a court regarding the legitimate actions of the state, although a lot of merit may be found in the Kant-inspired idea that persons are moral agents to whom organised society owes unconditional respect, and thus that the liberties whose compromise should be excluded from the outset are the ones whose abridgment entails a denial of one’s dignity as a moral agent[32] or, in Dworkinian terms, a denial of ‘equal concern and respect’[33]. It suffices for present purposes to ascertain that proportionality does not explicitly articulate a sound answer as to how to exclude such unacceptable aims from the range of the political authorities, but instead accepts them in principle, only to balance them out in the final step, without explicit reasons as to why the achievement of these aims cannot justify such infringements to individual rights.

Still, despite these important caveats – which arguably challenge both the internal architecture and the external overall structure of proportionality[34] –, it could be argued that proportionality is needed after the exclusion of a great part of the justification on the part of the public authorities in the first step, for the assessment of the measure in question. Under this analysis, one would accept that some ‘clearly important’ rights are non-rebuttable, maintaining at the same time ‘a prima facie right to everything’, similar to the notion of total freedom that Hobbes attributed to individuals (not as members of an organised society, but) in the state of nature[35]. In that sense, every piece of legislation would give rise to an issue (not of human rights, which are protected in the first step of the proportionality analysis, but) of individuals’ interests, and thus the judiciary would be called to decide on virtually every question of public policy. Such a scheme, however, would tilt the institutional balance between the legislature and the judiciary excessively in favour of the latter: the courts would not effectively act as fora of review of the acts of public authorities, but as bodies of appeal of policy choices made by the representative and accountable officials – a task which courts lack the resources, the know-how and (most importantly) the legitimacy for[36]. The dangerous implications of such a conception of judicial review are acknowledged by certain proponents of proportionality, who themselves admit that ‘judicial review of a legislature’s balancing should not require it to prove that every conceivable alternative policy would have been less effective, for this drags a court too far into the realm of policy analysis and evaluation. … [c]ourts should ensure that the legislature has in fact made a judgment that pursuing objectives that conflict with constitutional rights is justified under the constitutional criteria’[37]. But this scrutiny – ensuring that the political authorities have in fact taken all relevant factors into consideration before making a judgment – is fairly distanced from the ‘Idealtypus’ of proportionality analysis and practically approaches the method of review advocated here.

Our fundamental conceptions of democracy should lead us into assuming that the political authorities of a democratic society have the primary say in questions of what the common good is[38], thus accepting that they may use coercive force for all sorts of purposes and restrict freedom in all sorts of ways, as long as they do not deny a liberty that would compromise our dignity – but such an action, as argued before, would be considered unacceptable by a court in the primary threshold, as an illicit aim that the public authorities are not entitled to pursue in the first place. In other words, once the protection of rights is secured by determining which factors are lawfully put into the balancing scales and, thus, once the ambit of the coercive power of the state is delimited, the actual process of ‘balancing’ interests in order to come up with the appropriate means that will lead to the appropriate ends should arguably be left to the initial decision-maker. Within this process of decision-making by the political authorities there may be good arguments in favour of adopting a version of a proportionality analysis[39], but this is not to mean that the actual assessment of (or, in the proportionality parlance, the weight afforded to) the interests put into scales is to be performed again by a second decision-maker (in our case, the court), because the principal role of the political bodies is precisely to make such political assessments of interests[40]. As noted above, under the form of judicial review advocated here the government will be called, upon the application of any individual, to justify in public whether or not its action fell in the ambit of its regulatory and discretionary power. Similarly, the impartial court will be asked to assess these arguments – balancing their plausibility, one might illustratively say – in order to determine whether the government took into account all the important and relevant factors, excluding at the same time all the irrelevant considerations, before making its choice – but no more. The court cannot assess the choice made; it can only (and certainly must) ensure that the political decision-maker made a decision with full cognisance of what is at stake – and this is undoubtedly a crucial and arduous task which should not be underestimated.

It is often noted in this context that the courts are frequently called, by virtue of their role, to balance things under similar circumstances, such as in tort law or criminal law[41]. Nonetheless, in these cases it is the courts that assume the role, by virtue of their constitutional competences, of the initial decision-maker: just like the public authorities are externally constrained by the constitution (which demarkates their acts, as explained above) when deciding upon questions of public policy, the courts are constrained by the law when deciding upon a criminal sentence that they will impose; and, as argued above, the public authorities might be strongly encouraged to decide in a proportional manner, in a similar way as the court is called to decide proportionately on the sentence to be imposed[42]. What might be confusing in this context is that the court performs two functions consequently. First, it reviews the (criminal, for example) law enacted by the legislature which it is called to apply: if, say, the law provides capital punishment for the most severe crimes, it might arguably rule that this upper limit trespasses upon the liberties of individuals (in other words that the legislature denied through this law the individuals’ dignity as moral agents) and thus refuse to apply it, in the same way as it would prohibit the political authorities from maintaining in their arsenal during their decision-making process a similar measure that denied individuals their liberty. After this step, however, the court does not perform any form of review, but instead it assumes the role of the initial decision-maker as to the exact punishment to be imposed, and thus it enjoys relative discretion within the limits of reasonableness, similarly to the political authorities enjoying a – far larger, given the political, rather than merely legal, considerations that come into play – discretion as to the policy they will pursue. Here, as always, the principle of proportionality can only be rationally antecedent to choice – and, depending to the organ performing the decision-making process, this can be considered a virtue or a vice – but it cannot be thought at the same time as a rationally determinant of choice[43]: it can assist one in choosing, but it does not contain the answer itself.

In conclusion, it seems that, on balance, the use of the principle of proportionality does not facilitate but rather hinders the adjudicating process: clearly, the question of incommensurability is not a problem in itself, but proves that proportionality needs support from a sound theory of adjudication and particularly a theory of adjudication that mitigates the perfectionist and consequentialist underpinnings of the principle of proportionality in its crudest form. For that purpose, what is needed is a moral theory that will crystallise what claims can lawfully be put into the balancing scales and, thus, the whole process of justification (both on the part of the authorities and on the part of the individual) turns to the question whether the initial decision-maker has taken into consideration all the relevant factors, and excluded all the irrelevant factors, for her assessment: this is the difficult question to be answered by the court. Beyond this benchmark, proportionality may be a plausible method of decision-making, for it structures the reasoning of the decision-maker, but its outcomes are not something to be re-assessed by a court, unless one wants it to become the ultimate arbitrator of any question of the political community.

[1] Barak, ‘Proportionality and Principled Balancing’ (2010) 4 Law & Ethics of Human Rights 1, at pp.5-6.

[2] Beatty, The Ultimate Rule of Law (OUP 2004), pp.168-169.

[3] Webber, The Negotiable Constitution (Cambridge University Press, 2009), p.98.

[4] Endicott, ‘Proportionality and Incommensurability’ Oxford Legal Studies Research Paper No. 40/2012, p.5.

[5] Barak, supra (n.1), at pp.7-8.

[6] Barak, supra (n.1), at pp.9-10.

[7] Endicott, supra (n.3), at p.3.

[8]In Alexy’s theory, the weight formula calls us to establish, in turn, the degree of non-satisfaction of the first conflicting principle and of the importance of satisfying the competing principle before assessing whether the latter justifies the former. (see Alexy, ‘Balancing, Constitutional Review, and Representation’ (2005) 3 International Journal of Constitutional Law 572 at 575-576).

[9] Example given by Alexy, supra (n.8), at pp.574-575.

[10] Alexy, supra (n.8), at p.577.

[11] Möller, ‘Proportionality: Challenging the critics’ (2012) 10 International Journal of Constitutional Law 709, at p.715.

[12] Möller supra (n.11), at p.715.

[13] Tsakyrakis, ‘Proportionality: An assault on human rights’ (2009) 7 International Journal of Constitutional Law 468, at p.473, citing Waldron, ‘Fake Incommensurability: A Response to Professor Schauer’ (1994) 45 Hastings Law Journal 813, at p.821.

[14] See Beatty, supra (n.2), at p.168. Cf. Webber, supra (n.3), at pp.107-108.

[15] Endicott, supra (n.4), at pp.14-15.

[16] Gardbaum, ‘Democratic Defense of Constitutional Balancing’ (2010) 4 Law & Ethics of Human Rights 78, at. p.82.

[17] Gardbaum, supra (n.16), at p.82.

[18] Tsakyrakis, supra (n.13), at p.480.

[19] Kumm, ‘Political Liberalism and the Structure of Rights: On the Place and Limits of the Proportionality Requirement’ in Pavlakos (ed.), Law, Rights, Discourse: Themes of the Work of Robert Alexy (Hart, 2007), at p.140.

[20] Webber, supra (n.3), at p.105.

[21] Möller, supra (n.11), at p.712.

[22] Kumm, supra (n.19), at p.142.

[23] According to the US Declaration of Independence (1776).

[24] Möller, supra (n.11), at p.718.

[25] Gardbaum, supra (n.16), at p.86.

[26] This, as noted before, is the case for any type of moral reasoning.

[27] Gardbaum, supra (n.16), at p.100.

[28] Khosla, ‘Proportionality: An Assault on Human Rights? A Reply’ (2010) 8 International Journal of Constitutional Law 298, at p.302.

[29] As defenders of proportionality themselves acknowledge – see Gardbaum, supra (n.16), at p.99.

[30] Möller supra (n.11), at 724.

[31] As argued by Kumm, ‘The Idea of Socratic Contestation and the Right to Justification: The Point of Rights-Based Proportionality Review’ 4 Law & Ethics of Human Rights 141, at 152-154.

[32] Tsakyrakis, ‘Total Freedom: The Morality of Proportionality’ (available at:, at p.7.

[33] See Tsakyrakis, supra (n.32), at p.2: “‘Human worth’, ‘human dignity’, ‘human inviolability’, and ‘equal concern and respect’ are just variations of the same basic idea: a just and well-ordered society is one that recognizes persons as moral agents and treats them as such”.

[34] Webber, supra (n.3), at pp.100-101.

[35] Tsakyrakis,  supra (n.32), at p.4.

[36] Tsakyrakis, supra (n.32), at p.3.

[37] Gardbaum, supra (n.16), at p.104.

[38] Cf. Kyritsis, ‘Constitutional Review in Representative Democracy’ (2012) 32 Oxford Journal of Legal Studies 297, at p.315.

[39] For example, this method will enable us to clearly follow the reasoning process and evaluate whether the conclusion reached is reasonable. See Ekins, ‘Legislating Proportionately’ (on file with the author), at pp.3-4, who nevertheless provides good arguments against following the principle of proportionality as a general rule in the legislative decision-making process.

[40] See Dworkin, ‘You cannot calculate human rights based on cost’ The Guardian (24.05.2006).

[41] Endicott, supra (n.4), at pp.13-14.

[42] Despite the fact that the constraints of the court are much stricter, because it is supposed to take into account only legal considerations (as opposed to the government when deciding on a public policy), the fact remains that the court enjoys relative discretion as to the punishment it will impose.

[43] Webber, supra (n.3), p.97.

Nikiforos Panagis is an LLM student at Oxford University

This entry was posted in Papers and tagged . Bookmark the permalink.

Leave a Reply

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