Law & Ethics of Human Rights

Volume 4, Issue 1 2010 Article 2

RIGHTS, BALANCING & PROPORTIONALITY
The Construction of Constitutional Rights
Robert Alexy

Christian Albrechts University Kiel, lsalexy@law.uni-kiel.de

Copyright c 2010 The Berkeley Electronic Press. All rights reserved.
The Construction of Constitutional Rights

Robert Alexy
Abstract
This article calls for the construction of constitutional rights as principles, rather than as rules.
The rule construction conceives subsumption or classication as the appropriate form for the ap-
plication of constitutional rights. It attempts, in this way, to avoid the problems associated with
balancing. By contrast, the principles construction argues that balancing is inevitable and unavoid-
able. Balancing is at the very core of the proportionality test. The debate over the construction
of constitutional rights is, therefore, rst and foremost a debate over proportionality analysis. The
central objection to the principles construction is that balancing and, with it, the proportionality
test, is irrational. This irrationality objection is countered by analysis of the formal structure of
balancing; the analysis shows that balancing is a case of rational legal argument that is expli-
cated by means of an arithmetic formula: the Weight Formula. The Weight Formula provides a
demonstration of how and why balancing is possible as a form of rational legal argument. It also
makes it possible to show that proportionality analysis endangers neither the power nor the force
of constitutional rights.
KEYWORDS: Weight Formula, constitutional rights, rational legal argument, principles, rules,
balancing, proportionality test

Professor of Public Law and Legal Philosophy at the Christian Albrechts University Kiel. I would
like to thank Stanley L. Paulson for suggestions and advice on matters of English style.
21 Published by The Berkeley Electronic Press, 2010
The Construction oI Constitutional Rights
INTRODUCTION
There are two Iundamentally diIIerent constructions oI constitutional rights: the rule
construction and the principles construction. Neither oI these constructions is at
any point realized to the Iullest extent, but they nevertheless represent two opposing
ideas on which the solution oI nearly all questions oI the doctrine oI constitutional
rights turns. Questions pertaining to constitutional rights are not simply questions
in a particular area oI law. The answers given to such questions have consequences
Ior the structure oI the entire legal system. The spectrum extends Irom the third
party or horizontal eIIect, that is, the bearing oI constitutional rights on private law,
right up to the relation between the legislature and the practice oI constitutional
review, behind which the tension between constitutional rights and democracy is
Iound. The question oI which construction: the rule construction or the principles
construction is to be preIerred is, thereIore, by no means a problem oI theoretical
interest alone. It also has Iar-reaching practical import. For this reason, it is a basic
question oI constitutionalism.
I. THE RULE CONSTRUCTION
A. RULES AND PRINCIPLES
The basis oI both the rule and the principles construction is the norm-theoretic
distinction between rules and principles.
1
Rules are norms that require something
defnitively. They are dehnitive commands. Their Iorm oI application is
subsumption. II a rule is valid and applicable, it is defnitively required that exactly
what it demands be done. II this is done, the rule is complied with; iI this is not
done, the rule is not complied with. By contrast, principles are norms requiring
that something be realized to the greatest extent possible, given the Iactual and
legal possibilities at hand. Thus, principles are optimi:ation requirements. As such,
they are characterized by the Iact that they can be satisfed to varying degrees, and
that the appropriate degree oI satisIaction depends not only on what is Iactually
possible but also on what is legally possible. Rules aside, the legal possibilities
are determined essentially by opposing principles. For this reason, principles, each
taken alone, always comprise a merely prima Iacie requirement. The determination
oI the appropriate degree oI satisIaction oI one principle relative to the requirements
oI another principle is balancing. Thus, balancing is the specifc Iorm oI the
application oI principles.
1
See ROBERT ALEXY, A THEORY OF CONSTITUTIONAL RIGHTS 47-49 (Julian Rivers trans., OUP,
2002) (1985).
http://www.bepress.com/lehr/vol4/iss1/art2 22
Law & Ethics of Human Rights, Jol. 4 [2010], Iss. 1, Art. 2
The distinction between rules and principles is at the heart oI a theory that
might be called 'principles theory. The principles theory is the system drawn Irom
the implications oI the distinction between rules and principles. These implications
aIIect all areas oI the law. As Iar as constitutional rights are concerned, one
may speak oI the 'principles theory oI constitutional rights as well as oI their
'principles construction. The debate over the principles theory is, frst oI all, a
debate over weighing or balancingand, thereIore, since balancing is the core oI
the proportionality test, a debate over proportionality analysis.
B. THE POSTULATE TO AVOID BALANCING
The counterpart oI the principles construction, the rule construction, can be seen as
an attempt to avoid the problems connected with balancing. Constitutional rights
norms are considered as rules that are applicable, in essence, without balancing.
This does not mean that the application oI constitutional rights is conceived, in
all instances, as an unproblematic subsumption. Subsumption, here as elsewhere
in the law can be rather diIfcult and may require intermediate steps as well as
Iurther arguments oI diIIerent kinds in order to justiIy these intermediate steps.
2

Thus, it can be very doubtIul whether a certain statement is an expression oI one`s
opinionprotected by the Ireedom oI speech, or whether a certain activity is an
exercise oI religion, or whether a certain valuable advantage counts as property
protected by the constitution. It is decisive Ior the rule construction that not only
such questions as these but, going beyond them, all other questions that may arise
in the application oI constitutional rights be resolved in essence without balancing.
This solution, Ireedom Irom balancing, may acquire either a positivistic
or a non-positivistic character. An example oI a positivistic construction, Iree
oI balancing, is Ernst ForsthoII`s postulate that all questions connected with the
application oI constitutional rights be resolved by means oI the traditional canons oI
interpretation
3
that is, by appealing, above all, to the wording oI the constitutional
rights provisions, the intentions oI those who Iramed the constitution, and the
systematic context oI the provision being interpreted. Today the best-known version
oI a non-positivistic construction, Iree oI balancing, is Iound in the work oI Ronald
Dworkin. According to him, the application oI constitutional rights is not a matter oI
striking a balance. Rather, it concerns 'the very diIIerent question oI what morality
2
See ROBERT ALEXY, A THEORY OF LEGAL ARGUMENTATION: THE THEORY OF RATIONAL DISCOURSE
AS THEORY OF LEGAL JUSTIFICATION 221-30 (Ruth Adler & Neil MacCormick trans., 1989) (1978).
3
ERNST FORSTHOFF, RECHTSSTAAT IM WANDEL: VERFASSUNGSRECHTLICHE ABHANDLUNGEN 1954-
1973, at 173 (2d ed. 1976).
23 Published by The Berkeley Electronic Press, 2010
The Construction oI Constitutional Rights
requires.
4
II the principles construction is defned as a proportionality construction
that includes balancing essentially, then this, too, is a rule construction, albeit one
oI a special kind.
C. PROBLEMS OF THE RULE CONSTRUCTION
The problems oI the rule construction emerge most clearly in connection with the
question oI the limits oI constitutional rights. Here only two constellations are
considered: that oI a constitutional right with a statutory reservation, and that oI a
constitutional right guaranteed without this reservation.
The simplest Iorm oI a statutory reservation is present, Ior example, when a
constitutional rights provision guarantees, in a frst step, constitutional rights such
as the right to liIe and to bodily integrity, and, in a second step, then empowers
the legislature by means oI a clause such as '|t|hese rights can only be interIered
with on a statutory basis
5
to limit these rights. II one Iollows the rule construction
and takes these provisions literally, the limitation clause makes possible any
interIerence with liIe and bodily integrity as long as the interIerence is based on a
statute. Thus, the constitutional right is reduced to a special statutory reservation,
and it loses the power it would otherwise have had to bind the legislature. From
the standpoint, then, oI substantive law, such a constitutional right simply has no
bearing on the legislator. This, however, stands in contradiction to the binding
Iorce oI constitutional rights on the legislature. One may attempt to avoid this
situation by adding more rules. The systematically most demanding attempt here
is the prohibition to inIringe on the core content oI a constitutional right, as Iound
in Article 19 (2) GG. Even here, however, the legislature remains completely Iree
at every point beneath the threshold oI the core content. Moreover, it is highly
unlikely that the core content can be determined at all in a manner that is Iree Irom
balancing.
6
No less serious are the problems oI the rule construction in cases oI
constitutional rights guaranteed without a statutory reservation, that is, constitutional
rights with respect to which the constitution explicitly provides Ior no limits. In the
GG this is the case, Ior instance, with Ireedom oI religion and oI science. II one
were to subsume, as an isolated instance, religious suppressionwhen required by
a particular religious Iaithand medical experiments with human beingswhen
4
RONALD DWORKIN, IS DEMOCRACY POSSIBLE HERE? PRINCIPLES FOR A NEW POLITICAL DEBATE
27 (2006); see Kai Mller, Balancing and the Structure of Constitutional Rights, 5 I. CON: INT`L J.
CONST. L. 453, 458-61 (2007).
5
See Grundgesetz Ir die Bundesrepublik Deutschland (Basic Law) art. 2 (2) (3) |hereinaIter GG|.
6
See ALEXY, supra note 1, at 192-96.
http://www.bepress.com/lehr/vol4/iss1/art2 24
Law & Ethics of Human Rights, Jol. 4 [2010], Iss. 1, Art. 2
advancing scientifc progressthis type oI isolated instances oI subsumption
would be allowed. The alternative oI not classiIying these measures as religious
acts or scientifc research would contradict the wording oI the constitutional
rights provisions. II, however, the rights to Ireedom, liIe, and bodily integrity oI
those concerned are employed as reasons Ior imposing limitations, one inevitably
conIronts the need to balance. In this way, the postulate oI systematic interpretation
warrants, indeed, requires that one move beyond the scope oI the rule model.
II. PRINCIPLES CONSTRUCTION AND PROPORTIONALITY ANALYSIS
The principles construction attempts to resolve these and a great many additional
problems by conceiving oI constitutional rights as principles, that is, as optimization
requirements.
The signifcance oI constitutional rights as principles stems Irom their
connection to the principle oI proportionality. This connection is as close as
it could possibly be. It consists in a relation oI mutual implication. The
principle oI proportionality with its three sub-principles oI suitability, necessity,
and proportionality in the narrower sense Iollows logically Irom the defnition
oI principle, just as the defnition oI principle Iollows Irom the principle oI
proportionality with its three sub-principles.
7
This means that the principle oI
proportionality is valid iI constitutional rights have the character oI principles
and that constitutional rights have the character oI principles iI the principle
oI proportionality determines the application oI constitutional rights. The core
oI the principles construction consists in this necessary connection between
constitutional rights and proportionality.
III. OBJECTIONS TO THE PRINCIPLES CONSTRUCTION
To the principles construction as well as to the principles theory in general a great
many objections oI various sorts have been raised, and it is impossible to discuss
all oI them here. But because the objections, despite their diversity, have a certain
systematic relation to each other, a short general account oI them is oIIered beIore
taking up the most central objection. A classifcation into seven groups seems
to be advisable. The frst group comprises the norm-theoretic objections. Here
the concern is over such questions as whether there exist legal principles at all,
8

7
Id. at 66-69.
8
Larry Alexander, There Are No Legal Principles, in INSTITUTIONAL REASON: THE
JURISPRUDENCE OF ROBERT ALEXY (Matthias Klatt ed., OUP, Iorthcoming, 2010).
25 Published by The Berkeley Electronic Press, 2010
The Construction oI Constitutional Rights
whether and how principles can be distinguished Irom rules,
9
whether principles
are norms,
10
and whether a basic opposition oI rules to principles is not misguided,
in light oI the extraordinary variety oI norms.
11
The objections oI the second group
may be designated as argumentation-theoretic. Here the central issue is whether
balancing is properly seen as rational argumentation or whether it must not be
classifed as a non-rational or irrational method.
12
The objections oI the third group
turn on the question oI whether the principles construction counts as a danger to
constitutional rights, eliminating as it does their strict validity as rules.
13
Objections
oI this kind concern the doctrine oI constitutional rights. They may, thereIore, be
termed doctrinal objections. Whereas the doctrinal objections appeal to the danger
oI too little Iorce in constitutional rights, the opposite, the danger oI too much
Iorce, stands at the center oI the objections oI the Iourth group. These objections
claim that the optimization thesis leads to a proliIeration oI constitutional rights,
leading in turn to an overconstitutionalization oI the legal system. The institutional
consequence is said to be a 'shiIt Irom the parliamentary legislative state to a
constitutional adjudicative state.
14
One may reIer here to institutional objections.
The fIth group consists oI interpretation-theoretic objections. They concern
the question oI whether and how the principles construction can be established
as a correct interpretation oI a catalogue oI constitutional rights as understood in
positive law.
15
Is it possible to justiIy the universal validity oI the principles or
proportionality construction,
16
or is it to be applied, at best, only occasionally and
9
RalI Poscher, Einsichten, Irrtmer und Selbstmiverstndnisse der Prin:ipientheorie, in
DIE PRINZIPIENTHEORIE DER GRUNDRECHTE: STUDIEN ZUR GRUNDRECHTSTHEORIE ROBERT ALEXYS 59, 65,
70 (Jan-R. Sieckmann ed., 2007).
10
Jan Henrik Klement, Jom Nut:en einer Theorie, die alles erklrt, 63 JURISTENZEITUNG
756, 760 (2008).
11
Poscher, supra note 9, at 73-74.
12
See JRGEN HABERMAS, BETWEEN FACTS AND NORMS: CONTRIBUTIONS TO A DISCOURSE THEORY
OF LAW AND DEMOCRACY 259 (William Rehg trans., Polity Press, 1996) (1992); Bernhard Schlink,
Der Grundsat: der Jerhltnismigkeit, in FESTSCHRIFT 50 JAHRE BUNDESVERFASSUNGSGERICHT 445,
460 (Peter Badura & Horst Dreier eds., 2001).
13
See HABERMAS, supra note 12, at 258-59 (maintaining that conceptions oI rights as
principles might result in the collapse oI rights that are properly understood, i.e., rights act as a 'fre
wall.)
14
ERNST-WOLFGANG BCKENFRDE, STAAT, VERFASSUNG, DEMOKRATIE 190 (1991) (trans. R. A.).
15
Matthias Jestaedt, Die Abwgungslehre ihre Strken und ihre Schwchen, in STAAT IM
WORT: FESTSCHRIFT FR JOSEF ISENSEE 253, 260, 262-63, 275 (Otto Depenheuer, Markus Heintzen,
Matthias Jestaedt, & Peter Axer eds., 2007); Poscher, supra note 9, at 79; Klement, supra note 10,
at 761, 763.
16
See, e.g., DAVID M. BEATTY, THE ULTIMATE RULE OF LAW 162 (2004): 'Proportionality is
a universal criterion oI constitutionality.
http://www.bepress.com/lehr/vol4/iss1/art2 26
Law & Ethics of Human Rights, Jol. 4 [2010], Iss. 1, Art. 2
on an ad hoc basis? The objections oI the sixth group concern validitv-theoretic
questions. These objections reproach the principles theory Ior jeopardizing the
superior rank oI the constitution and the subjection oI the executive and judiciary to
statutory law.
17
The hierarchical structure oI the legal system is said to collapse in
the whirl oI balancing.
18
The seventh group, fnally, is Iormed by science-theoretic
objections. The principles theory is said to consist oI 'statements that Ior reason
oI their abstractness do not say anything at all,
19
statements that are capable oI
explaining each and every decision taken, but without 'any directive power Ior
decisions to be taken in the Iuture.
20
For this reason, so it is argued, the principles
theory cannot suIfce as a doctrine oI constitutional rights.
21
IV. THE RATIONALITY OF BALANCING
A. THE CENTRAL ROLE OF THE RATIONALITY PROBLEM
The argumentation-theoretic objections that concern the rationality oI balancing
count as the most important group. II balancing were by its nature irrational, it
would have to be rejected, and with it, principles qua norms that require something
irrational.
22
The norm-theoretic debate would have lost its point. Moreover, iI
balancing were irrational, it would not make sense to argue on behalI oI balancing
as the applicable criterion on the question oI the admissibility oI limitations on
constitutional rights. The doctrinal objections would win hands down. The same
would apply to the institutional critique. To be sure, the situation is more complex
in the case oI the interpretation-theoretic arguments. II balancing should prove to be
irrational, one would adhere to it at most in instances where a constitution explicitly
required a proportionality test, as, Ior instance, in the case oI Article 52 (1) oI the
Charter oI Fundamental Rights oI the European Union
23
. Even here, however, one
may ask whether a norm that requires something irrational might not be corrected
17
Jestaedt, supra note 15, at 268, 275; Poscher, supra note 9, at 76; Klement, supra note
10, at 759.
18
Jestaedt, supra note 15, at 269-70.
19
Id. at 269 (trans. R. A.).
20
Klement, supra note 10, at 756 (trans. R. A.); RALF POSCHER, GRUNDRECHTE ALS ABWEHRRECHTE:
REFLEXIVE REGELUNG RECHTLICH GEORDNETER FREIHEIT 76 (2003); Jestaedt, supra note 15, at 269.
21
POSCHER, supra note 20, at 77-78.
22
This applies in any case to optimization relative to the legal possibilities. Optimization
relative to the Iactual possibilities might be retained even in the case oI irrationality oI balancing.
23
Charter oI Fundamental Rights oI the European Union 2000 O.J. (C 364) 1 (Dec. 7,
2000).
27 Published by The Berkeley Electronic Press, 2010
The Construction oI Constitutional Rights
more appropriately by way oI interpretation. The validity-theoretic objection
would carry the day, too. Who would care to assent to an irrational restriction
on the validity oI a norm? The irrationality oI balancing would, fnally, be grist
Ior the mills oI all science-theoretic objections. For these reasons, the rationality
problem shall take the centre stage in what Iollows. To be sure, a reIutation oI the
irrationality objection does not imply the reIutation oI any oI the other objections.
The rationality oI balancing is not a suIfcient condition Ior the rebuttal oI the other
objections, but it is, in any case, a necessary condition.
B. THE IRRATIONALITY OBJECTION
Many authors have contested the rationality and, with it, the objectivity oI
balancingespecially emphatically by Habermas and Schlink. Habermas` central
point is that there exist 'no rational standards Ior balancing: 'Because there are no
rational standards Ior this, weighing takes place either arbitrarily or unrefectively,
according to customary standards and hierarchies.
24
Where Habermas speaks
about arbitrariness and un-refected customs, Schlink employs the concepts oI
subjectivity and decision:
In the test oI proportionality in the narrower sense, it is only the subjectivity
oI the tester that can, in the end, be eIIective . The operations oI evaluation
and balancing required by the proportionality test in the narrower sense can
be achieved . in the end, only in a decisionistic manner.
25
Is this true? Is balancing really non-rational or irrational, arbitrary, subjective,
and decisionistic? Does balancing indeed mean that rationality, correctness, and
objectivity are abandoned? It is diIfcult to answer these questions without knowing
what balancing is. And to fnd out what balancing is, some insight into its structure
is presupposed.
C. PARETO-OPTIMALITY
It has already been noted that balancing is the subject oI the third sub-principle
oI the principle oI proportionality. This is the principle oI proportionality in the
narrower sense, which concerns optimization relative to the legal possibilities at
hand. By contrast, the subject oI the frst two sub-principles, the sub-principle oI
24
HABERMAS, supra note 12, at 259.
25
Bernhard Schlink, Freiheit durch EingriffsabwehrRekonstruktion der klassischen
Grundrechtsfunktion, 11 EUROPISCHE GRUNDRECHTE-ZEITSCHRIFT (EUGRZ) 457, 462 (1984) (trans.
R. A.); see also Schlink, supra note 12, at 460.
http://www.bepress.com/lehr/vol4/iss1/art2 28
Law & Ethics of Human Rights, Jol. 4 [2010], Iss. 1, Art. 2
suitability and necessity, is optimization relative to the Iactual possibilities. In this
respect, they are concerned with the question oI whether the Iactual possibilities
allow Ior the avoidance oI costs to constitutional rights without bringing about
costs contrary to the aims oI the legislator. The issue, in other words, is Pareto-
optimality.
26
This is Iar less problematic than balancing, which is concerned with
the question oI which side has to bear the costs. For this reason, the principles oI
suitability and necessity shall not be discussed here.
27
In any case, the mere Iact
that the principles construction is able to grasp this aspect, too, is a strong argument
on behalI oI its correctness.
D. THE LAW OF BALANCING
The basic idea oI optimization relative to the legal possibilities at hand can be
expressed by a rule that might be called the 'Law oI Balancing. A statement oI this
rule runs as Iollows: 'The greater the degree oI non-satisIaction oI, or detriment
to, one principle, the greater must be the importance oI satisIying the other.
28
The
Law oI Balancing shows that balancing can be broken down into three stages. The
frst stage is a matter oI establishing the degree oI non-satisIaction oI, or detriment
to, the frst principle. This is Iollowed by a second stage, in which the importance
oI satisIying the competing principle is established. Finally, the third stage answers
the question oI whether or not the importance oI satisIying the competing principle
justifes the detriment to, or non-satisIaction oI, the frst principle. II it were not
possible to make rational judgments about, frst, intensity oI interIerence, second,
degrees oI importance, and, third, their relationship to each other, the objections
raised by Habermas and Schlink would be justifed.
E. THE WEIGHT FORMULA
In order to show that rational judgments about intensity oI interIerence and
degrees oI importance are possible, I turn frst to a decision oI the German Federal
Constitutional Court on health warnings.
29
The Court characterizes the obligation
oI tobacco producers to place health warnings respecting the dangers oI smoking on
their products as a relatively minor or light interIerence with the Ireedom to pursue
one`s proIession. By contrast, a total ban on all tobacco products would count as
26
ALEXY, supra note 1, at 67-69.
27
See LAURA CLERICO, DIE STRUKTUR DER VERHLTNISMSSIGKEIT 26-39 (2001).
28
ALEXY, supra note 1, at 102.
29
Entscheidungen des BundesverIassungsgerichts (Decisions oI the Federal Constitutional
Court) 95, 173 |hereinaIter BVerIGE|.
29 Published by The Berkeley Electronic Press, 2010
The Construction oI Constitutional Rights
a serious interIerence. Between such minor and serious cases, others oI moderate
intensity oI interIerence can be Iound. In this way, a scale can be developed with
the stages 'light, 'moderate, and 'serious. Our example shows that the intensity
oI interIerence can be determined by means oI this scale.
The same is possible on the side oI the competing reasons. The health risks
resulting Irom smoking are high. Thus, the reasons justiIying interIerence weigh
heavily. II in this way the intensity oI interIerence is established as minor, and the
degree oI importance oI the reasons Ior the interIerence as high, then the outcome
oI examining proportionality in the narrower sense can well be describedas the
German Federal Constitutional Court has in Iact described itas 'obvious.
30
The teachings drawn Irom the tobacco judgment are corroborated iI one
looks at other cases. A rather diIIerent case is represented by the Titanic judgment.
The widely-published satirical magazine, Titanic, described a paraplegic reserve
oIfcer, frst, as a 'born murderer and then, in a later edition, as a 'cripple. A
German Court ruled against Titanic and ordered the magazine to pay damages to
the oIfcer in the amount oI 12,000 DM. Titanic brought a constitutional complaint.
The Federal Constitutional Court undertook a 'case-specifc balancing
31
between
the Ireedom oI expression oI the magazine (Article 5 (1) (1) GG) and the oIfcer`s
general right oI personality (Article 2 (1), in connection with Article 1 (1) GG).
This case, too, can be reconstructed by means oI the triadic scale: light, moderate,
and serious.
However, the triadic scale is not by itselI enough to demonstrate that
balancing is rational. A demonstration requires that it is possible to embed such
classifcations in an inIerential system that is understood as implicit in balancing
and that, in turn, is intrinsically connected with the concept oI correctness. In the
case oI subsumption under a rule, such an inIerential system can be expressed by
means oI a deductive scheme, called 'internal justifcation, that is constructed with
the help oI propositional, predicate, and deontic logic and integrated into the theory
oI legal discourse.
32
It is oI great importance, both Ior the theory oI legal discourse
and the theory oI constitutional rights, that in the case oI balancing a counterpart to
this deductive scheme exists.
33
This is the Weight Formula.
The core and, at the same time, the simplest Iorm oI the Weight Formula
runs as Iollows:
W
i, f

30
BVerIGE 95, 173 (187).
31
BVerIGE 86, 1 (11).
32
ALEXY, supra note 2, at 273-83.
33
Robert Alexy, On Balancing and Subsumption. A Structural Comparison, 16 RATIO JURIS
433, 448 (2003).
I
i
I
f
http://www.bepress.com/lehr/vol4/iss1/art2 30
Law & Ethics of Human Rights, Jol. 4 [2010], Iss. 1, Art. 2
In this Iormula, one fnds that the variables Ior the abstract weights oI the
competing principles (W
i
, W
f
) and Ior the reliability oI the empirical assumptions
concerning what the measure in question means in the concrete case Ior the non-
realization oI the one principle and the realization oI the other principle (R
i
, R
f
) are
still lacking. But this can be leIt out oI consideration here, so that the simplest Iorm
just mentioned may stand Ior the complete Iorm:
34
W
i, f

In the simplest as well as in the complete Iorm, I
i
stands Ior the intensity oI
interIerence with the principle P
i
in our case, the principle granting the Ireedom
oI expression oI Titanic. I
f
stands Ior the importance oI satisIying the competing
principle P
f
, which, in our case, is the principle granting the personality right oI
the paraplegic oIfcer. Finally, W
i, f
stands Ior the concrete weight oI the principle
whose violation is being examinedin our case, that oI P
i
. The Weight Formula
gives expression to the point that the concrete weight oI a principle is a relative
weight. It does this, in the simplest Iorm, by defning the concrete weight as the
quotient oI the intensity oI interIerence with this principle (P
i
) and the concrete
importance oI the competing principle (P
f
).
Now, the objection is clear that one can only talk about quotients in the
presence oI numbers, and that numbers are not used in the balancing carried out in
constitutional law. The reply to this objection might begin with the observation that
the vocabulary drawn Irom logic, which we are using here in order to express the
structure oI subsumption, is not used in judicial reasoning, but that this vocabulary
is nevertheless the best means available to make explicit the inIerential structure
oI rules. The same would apply to an expression oI the inIerential structure oI
principles by means oI numbers that are substituted Ior the variables oI the Weight
Formula.
The three values oI the triadic modellight, moderate, and seriouscan be
represented by l, m, and s. To be sure, the triadic model by no means exhausts the
possibilities oI gradation. Balancing would be possible once one had a scale with
two values, say, l and s. Balancing is impossible only iI everything has the same
value.
35
Moreover, there are numerous possibilities Ior refning the scale. Adouble-
triadic scale is oI special interest. It works with nine steps or values: (1) ll, (2) lm,
34
See generallv Robert Alexy, The Weight Formula, in FRONTIERS OF THE ECONOMIC ANALYSIS
OF LAW 9, at 20-26 (Jerzy Stelmach, Bartosz Broek, & Wojciech Zaluski eds., 2007).
35
See AHARON BARAK, THE JUDGE IN A DEMOCRACY 166 (2006): 'One cannot balance without
a scale.
I
i
W
i
R
i
I
f
W
f
R
f
31 Published by The Berkeley Electronic Press, 2010
The Construction oI Constitutional Rights
(3) ls, (4) ml, (5) mm, (6) ms, (7) sl, (8) sm, (9) ss. It is oI utmost importance that the
possibilities oI refnement are limited. All classifcations are judgments. Everybody
understands a statement such as 'The inIringement is light (l) or a statement such
as 'The inIringement is a serious moderate inIringement. But how should the
statement 'The inIringement is a serious light inIringement oI moderate kind (lsm)
be understood, which is made possible by adding a third triad? A justifcation can
only be given Ior what one understands, and the justifability oI statements about
intensities is a condition oI the rationality oI balancing. This implies that graduation
in the province oI constitutional rights can work only with relatively crude scales.
In the end, it is the nature oI law, here, the nature oI constitutional law that sets limits
to the degree to which the graduation can discriminate and altogether excludes the
applicability oI any infnitesimal scale.
36
Calculable measurements by way oI a
continuum oI points between 0 and 1 cannot have any application.
Even crude, discrete scales, however, cannot dispense with the allocation
oI numbers iI the inIerential structure oI balancing is going to be expressed by the
Weight Formula. There are various possibilities Ior allocating numbers to the three
values oI the triadic model. A rather simple and at the same time highly instructive
possibility consists in taking the geometric sequence 2
0
, 2
1
, and 2
2
, that is, 1, 2,
and 4.
37
In the Titanic judgment, the Federal Constitutional Court considered
the intensity oI inIringement (I
i
) with the Ireedom oI expression (P
i
) as serious
(s); it considered the importance oI satisIying the right to personality (P
f
) oI the
oIfcer (I
f
), in the matter oI describing him as born murderer`given its highly
satirical contextas only moderate (m), perhaps even as light (l). II we insert the
corresponding values oI our geometric sequence Ior s and m, then the concrete
weight oI P
i
(W
i, f
) is, in this case, 4/2, that is, 2. II, conversely, I
i
were m and I
f
were
s, the value would be 2/4, that is, 1/2. The priority oI P
i
is expressed by a concrete
weight greater than 1; the priority oI P
f
by a concrete weight less than 1. In all
stalemate cases the value is 1.
The description oI the oIfcer as a 'cripple, however, was considered by
the Court as a severe (s) interIerence with the right oI personality. This gave rise
to a stalemate, with the consequence that Titanic`s constitutional complaint was
not successIul in so Iar as it related to damages stemming Irom the description
36
Robert Alexy, Jerfassungsrecht und einfaches RechtJerfassungsgerichtsbarkeit und
Fachgerichtsbarkeit, 61 VERFFENTLICHUNGEN DER VEREINIGUNG DER DEUTSCHEN STAATSRECHTSLEHRER
7, 25-26 (2002).
37
The greatest advantage oI the geometric sequence consists in its providing Ior the best
representation oI the overproportional increase oI the power oI rights as correlated with an increasing
intensity oI interIerence, a Iact that serves as the basis Ior the reIutation oI the objection concerning
the dissolution oI the power oI constitutional rights. See Alexy, supra note 34, at 22-23.
http://www.bepress.com/lehr/vol4/iss1/art2 32
Law & Ethics of Human Rights, Jol. 4 [2010], Iss. 1, Art. 2
'cripple. By contrast, Ireedom oI expression enjoys priority in the case oI the
description 'born murderer with the consequence that the damages were in so Iar
as they related thereto disproportional, hence unconstitutional. The constitutional
complaint was to this extent successIul.
The rationality oI an inIerential structure essentially depends on the question
oI whether it connects premises that, again, can be justifed. In the Weight Formula
the premises are represented by numbers which stand Ior judgments. An example is
the judgment that the public description oI a severely disabled person as a 'cripple
is a serious interIerence with that person`s personality right. This judgment raises a
claim to correctness, which can be justifed in a discourse. The commensurability oI
the assessments on both sides oI the balance is granted iI the discourse is conducted
on the basis oI a common point oI view: the point oI view oI the constitution.
38

The Federal Constitutional Court justifes its assessment with the argument that the
description as a 'cripple is nowadays considered as an expression oI disrespect
and humiliation.
39
Naturally, one can carry on a dispute about this, as about so
many questions in the law. But contestability does not imply irrationality. II this
were the case, not only balancing, but legal reasoning as such would be Ior the most
part irrational. Precisely the opposite, however, is the case. Justifability, despite
the Iact that it cannot be identifed with provability, implies rationality, and, with it,
objectivity, understood as lying between certainty and arbitrariness.
The end is attained. Balancing turns out to be an argument Iorm
40
oI rational
legal discourse.
41
This suIfces to reIute the irrationalism objection as an objection
specifcally directed against balancing. OI course, one might enquire quite generally
into the possibility oI rational legal argument, and one could also think about
undertaking a reply to the objections oI the other six groups on the basis oI what
has been elaborated here. This, however, will have to await another occasion. Here
it suIfces to say that the irrationalism objection, on which everything else depends,
can be dismissed. Having done so, we might well be in a position to say that an
important step along the way toward a Iull deIense oI the principles construction oI
constitutional rights has been taken.
38
Id. at 18.
39
BVerIGE 86, 1 (13).
40
See ALEXY, supra note 2, at 93. This, perhaps, is the point oI Barak`s thesis that 'balancing
introduces order into legal thought. See BARAK, supra note 35, at 173; see also id. at 164.
41
As an inIerential scheme expressed by the Weight Formula, balancing is a Iormal
structure that contains, as such, no substance whatever. The application oI the Weight Formula,
however, requires that contentmade explicit by judgments about the intensity oI interIerence,
abstract weight, and the reliability oI empirical assumptionsis substituted Ior the variables oI the
Weight Formula. For this reason, one can say that balancing is procedurally substantive.