Coherence
Coherence
Coherence
Abstract. The aim of this article is to propose a theoretical theme to explain coher-
ence in legal reasoning. The main argument that this paper wants to put forward is
that theories of coherence in the legal system should be differentiated from theories
of coherence in legal reasoning. These focus on arguments, and on how the given
arguments are connected. In particular, the notion of coherence in legal reasoning
proposed here is a modest one. The article applies this theme to the case-law of the
European Court of Justice in environmental matters. This provides an example of
how to deal with conflicts between incommensurable goods, and how to promote
coherence by justifying decisions.
Simple folk have an advantage over the learned, who often get lost in the search
for general laws. They often have the intuition of the individual. But this
intuition by itself is insufficient. (Eco, The Name of the Rose)1
Introduction
The aim of this article is twofold. First, it proposes a theoretical theme to
explain coherence in legal reasoning; second, it applies this theme to the
case-law of the European Court of Justice in environmental matters.
The main argument that this paper intends to put forward is that theories
of coherence in the legal system should be differentiated from theories of
coherence in legal reasoning. Theories of coherence in the legal system focus
* Most sincere thanks to Zenon Bankowski and Kevin Walton for extremely useful comments
and corrections. All remaining mistakes are my own responsibility.
** This paper was written during my stay at the Centre for Law and Society, University of
Edinburgh, as Marie Curie Research Fellow between 1998 and 2000.
1
Translated from the Italian original: “I semplici hanno qualcosa di più dei dottori, che spesso
si perdono alla ricerca di leggi generalissime. Essi hanno l’intuizione dell’individuale. Ma
questa intuizione da sola non basta.”
© Blackwell Publishing Ltd 2003, 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden 02148, USA.
Coherence in Legal Reasoning 297
on fitting a decision into the legal system, and on the fitting together of all
components of the legal system.2 Theories of coherence in legal reasoning
focus on the arguments, and on how the given arguments are connected.
This paper argues that prior to asking whether the rule contained in a deci-
sion coheres with the legal system, one should ask whether the reasoning
itself coheres. Legal reasoning, and in particular legal adjudication, is a
matter of supporting judicial decisions by arguments or reasons which
cumulate,3 form chains,4 or form cumulation-nets;5 theories of coherence in
adjudication focus on how to form coherent cumulations, chains, or nets of
reasons.
In particular, the notion of coherence in legal reasoning proposed here is
a modest one. It is modest in different senses. It does not aim (like global
notions of coherence) at identifying what the law is according to a princi-
ple, or set of principles. It rather aims at recognizing the value pluralism of
a legal system, that is, the idea that different and incommensurable values
belong to and collide within the legal system. The way of dealing with value
pluralism is modest: arguments; and so is the notion of coherence, for it aims
at providing tools to elaborate coherent supportive structures of arguments,
rather than at providing a definitive answer.
This paper also analyses several European Court of Justice (the Court)
decisions on environmental matters, which concern the conflict between two
incommensurable goods, namely, environmental protection and economic
freedoms. Incommensurable goods such as the environment and economic
freedoms cannot be measured, that is, they “cannot be aligned along a single
metric” (Sunstein 1997, 238). However, the absence of a metric does not
imply that conflicts between incommensurable goods cannot be treated in a
fair and rational way. Choices can be made; the question is how can incom-
mensurable goods be evaluated and choices be made without their being
irrational or arbitrary (for no metric is supplied). Arguments, i.e., reasons
are the main tool to justify rationally choices between incommensurable
goods. Here, the modest notion of coherence is needed to evaluate how far
the Court complies with the theoretical theme this paper proposes, and to
evaluate how coherent its legal reasoning is. The conclusions show that
attempts to generate coherence by making all decisions fit into a single line,
namely integration, and the criticisms of judicial activism of the Court are
2
Coherence in the legal system focuses on fitting together all components of the legal system.
Narrative notions of coherence in the legal system (Esser, Dworkin) are currently very popular,
partly because narrativeness is taking over logical inferences (deriving norms from a superior
rule) as the way to both describing and/or justifying legal systems. In addition, telling (coher-
ent) stories, rather than demonstrating truths, is a dominant feature of our current scientific
paradigm, especially since we understand telling stories as an activity of both reproducing
(past rules) and creating (new rules).
3
The reasons which justify ORx are r1 and r2 and r3 and rn.
4
The reasons which justify ORx are r1 because of p1.
5
The reasons which justify ORx are r1 because p1 and r2 because p1 and p2.
ing coherence in this sense is provided by the case-law of the Court on waste
oils management: Principles governing this branch of law collide quite often.
For example, on the one hand, the principles of self-sufficiency, proximity,
and “polluter pay” support limits on intra-Community shipments of waste;
on the other hand, free movement of goods and priority for recovery ban
these limits. Solving such a conflict is a matter of generating coherence: “As
points come out”—as Edward says—the Court tries to solve these antino-
mies by laying down the rules which aim to make sense of the legislation
on waste oil management.7
Third, the notion of coherence that Edward has in mind is not global: It
aims at making sense of a particular area of the legal system in which incon-
sistencies appear. Such an account of coherence is very close to the area-
referring or local coherence defended by Levenbook. She says: “Coherence
[. . .] is not always to be achieved to the whole system of established law in
the jurisdiction in question [. . .] Sometimes coherence is to be achieved only
to a subset, to a group of legal standards and decisions constituting a branch
of law. This is a much more manageable task than achieving global coher-
ence, and one that judges with limitations of knowledge and time can be
expected to achieve” (Levenbook 1984, 371).
To summarize, Edward’s quotation seems to provide some clues con-
cerning the notion of coherence the Court might have in mind. First, a coher-
ent account of the law has to be differentiated from a coherent account of
adjudication; second, coherence is not a foundationalist idea, that is, neither
the law nor the justification of a judicial decision can be exclusively deter-
mined by reference to its coherence; and third, local coherence is preferred
to global coherence.
This paper shares most of the above-mentioned findings, but some of them
need to be expanded. In particular, this paper holds that (i) the coherence of
the legal system is not the same as the coherence of the legal reasoning but
both notions are interrelated; (ii) coherence itself collapses when it is thought
of as a foundationalist criterion of the legal system and legal reasoning, and
yet it contains a claim to (axiological) validity; (iii) global accounts of coher-
ence are rejected, although the alternative notion is not a local one.
7
In this sense, the Court established in the Walloon waste case (Case C-2/90, Commission of the
European Communities v. Kingdom of Belgium [1992] I-4431) that according to the principles of
self-sufficiency and proximity which govern in environmental matters (Article 174 EC Treaty
[ex Article 130r]), shipments of waste, no matter whether recyclable or not, might be limited
for environmental purposes. In a similar case, Chemishe Afvalstoffen Dusseldorp, (C-203/96 [1998]
ECR I-4075), the Court delivered a different ruling, for intra-Community shipments of waste
for recovery cannot be limited according to the principle of priority for recovery introduced in
the Regulation on the supervision and control of shipments of waste (259/93 of 1 February).
The Court generates coherence by introducing a distinction: that between waste for disposal
and waste for recovery—principles of self-sufficiency and proximity govern the legislation on
waste for disposal, whereas the principle of priority for recovery governs the legislation on
waste for recovery.
the principle of effet utile (in European law) guarantees the predictability of
judicial decisions and the unity of legal reasoning. Indeed, in EC law, one
could hold that legal conflicts have to be solved in such a way that the effec-
tiveness in practice of European law is always guaranteed. In addition, using
the principle of effet utile as a criterion of coherence has the further benefit
of making all judicial decisions fit into a single line, and in this sense, it pro-
motes the unity of legal reasoning. However, both benefits (predictability
and unity) are fictitious. The effet utile doctrine neither makes the outcome
of legal conflicts more predictable, nor does it promote the unity of legal
reasoning. The case Commission v Germany (C-102/97 [1999] ECR I-5051) pro-
vides a good example in both senses. In this case, one of the last of the waste
saga, the Commission brought an action against Germany for failure to give
priority to the recycling, rather than the thermal processing of waste oils.
The disputed provision was Article 3 of Directive 75/439 on waste oils,
which establishes such a priority “where technical, economic, and organi-
sational constraints so allow.” Since this provision has derogating effects, the
Commission claimed that it should be interpreted restrictively, whereas
Germany argued that it should not. The former interpretation focuses on the
obligations of Member States under EC law, whereas the latter focuses on
the level of discretion of Member States under EC law. Indeed, the inter-
pretation of derogating provisions in EC law is a matter very much related
to the tension between Member States and European institutions.9 The Court
referred to the effet utile doctrine in deciding which interpretation to
embrace: If the technical, economic, and organisational circumstances
obtaining in a Member State were automatically to constitute constraints
which make it impossible to adopt measures for the recycling of waste oils,
that provision would be deprived of all practical effect, and Article 3 would
impose no genuine obligation on Member States. However, to see the deci-
sion of the Court in terms of the effectiveness of EC law is to see only part
of the picture, for the Court referred to other principles of EC law, namely
9
A similar example is provided by the well-known Lappel Bank case (C-44/95 [1996] ECR I-
3843). Here, Article 2 of Directive 79/409 on the protection of wild birds, establishes that
Member States are to take all necessary measures to maintain the population of all those species
of birds at a level which corresponds in particular to ecological, scientific, and cultural require-
ments, while taking account of economic and recreational requirements. Whereas the British
government held that Member States had discretion in determining the boundaries of special
protected areas (a measure devoted to the protection of wild birds) according to ecological and
economic interests, the Royal Society for the Protection of Birds held that economic and recre-
ational interests could be taken into account for the management of the special protected area,
not for its designation. The Court upheld the latter position; it was keen to stress that Article
2 of the Directive on the protection of birds could not amount to an autonomous derogation
clause. Such a conclusion can be interpreted as reinforcing the effectiveness of EC law; however,
it would be more appropriate to interpret it as the equilibrium between the discretion of
Member States and their obligations under EC law: Member States do not have discretion to
take account of economic and recreational considerations when establishing the boundaries of
special protected areas, but they do when implementing the measure.
10
In hard cases, this article argues, either the law does not provide a solution, or it provides
contradictory answers. It should be added that hard cases are also those in which the law pro-
vides a clear answer but either it is unreasonable, or the consequences are not acceptable.
11
It seems that Dworkin interprets judicial discretion as an open door for arbitrariness in law.
12
As Raz says, it would be absurd to elaborate a theory of coherence of law and legal adjudi-
cation which ignores its object, namely the law. Indeed, the test of fit proposed by Dworkin
expresses the requirement of taking account of previous legal material (statutory provisions
and precedents). However, rather than a requirement of coherence, the test of fit expresses a
crucial requirement of any account of legal adjudication.
13
Dworkin seems to argue that it is possible to achieve perfect moral solutions according to a
coherent set of principles, and do so in real conditions of practical reasoning (limits on time
and knowledge, coercion, etc.). If that is the case, there is no need to justify the authority of the
law; moreover, there is no need of law.
14
Dworkin’s concern for unity goes further, since in order to interpret the law as a unified
(coherent) set of principles, the judge must represent the community. Günther is right when he
says that “Dworkin projects the ideal of a coherent interpretation from the analysis of legal
adjudication in hard cases into the ideal of a principled community, and then takes it back into
legal adjudication. In consequence, it is the judge who inherits integrity as a device of repre-
sentation of a fair and just practice in a community of principle; it is the judge who becomes
the representative of the ‘single author’ of a coherent interpretation” (Günther 1995, 46).
Ultimately, Dworkin’s conception of coherence to legal adjudication seriously clashes with the
principle of democracy and democratic representation.
of coherence, the local approach accepts value pluralism, that is, it accepts
that there is a great variety of values which are irreducible, and conflicts
between them are an inescapable aspect of moral theory. Judges achieve
particular compromises between colliding values which should not be
tied together as if one single compromise covered all cases. Coherence, in
this sense, should not be an instrument to eliminate conflicts of values; it is
rather “a mere by-product of the consistent application of a sound moral
doctrine” (Raz 1994, 299). The “sound moral doctrine” to which Raz refers,
does not instruct the judge to take the morally best ruling; rather it instructs
the judge to take that decision which does not undermine the authority of
the law, for comprehensive reforms of the law cannot be attained by the judi-
ciary. Indeed, if one treats law, as Raz does, as authority, then, those coher-
ent compromises achieved in legal reasoning are mere “by-products.” It
seems that Raz is arguing for some form of literal interpretation (the most
loyal to the authority of the law) and against coherence.
However, both global and local accounts of coherence share the same
drawback: They do not differentiate between coherence in law and coher-
ence in legal reasoning. Both local and global notions focus on coherence of
the law, so a legal decision is coherent if it fits into a set of principles
(Dworkin), or principles and norms of a particular branch of law (Raz and
Levenbook). For theories of coherence of legal system, coherence is a quality
of the ruling, rather than of the reasoning.
16
Bankowski uses the metaphor of “holding the curtain ajar” to bridge the gap between the
domain of the universal (rules) and the particular. He argues that rules are abstractions of the
particularity of the case which take us away from the particular itself. Indeed, when applying
a rule, one has to abstract the particularity of the case and make out of it a legal category, so
applying the rule to the case is the activity of creating categories which can be subsumed into
the norm. Bankowski proposes a different way of dealing with the universal and the particu-
lar: That of balancing without folding this balancing into universalistic criteria. He calls that
“balancing when holding the curtain ajar.” This allows us to take account of the particular
without exhausting it in general categories.
17
A decision which makes evident the violation of the first aspect of the claim to correctness
goes: “I sentence Mr. N to ten years of imprisonment, according to an invalid law”; whereas
the violation of the second aspect of the claim to correctness is made evident in the following
ruling: “I sentence Mr. N to ten years of imprisonment, according to an unfair law.” This aspect
of the claim to correctness is missing in the case Daniels and Daniels v. R. White & Sons and Tarbard
(1938) 4 All E.R. 258. The plaintiffs sued the manufacturer of lemonade and the publican who
sold a bottle of lemonade containing carbolic acid. The legal question was whether the publi-
can was liable for damages, even though she could not examine the content of the bottle of
lemonade she sold. Justice Lewis said: “I therefore find that this was a sale by description, and
therefore hold—with some regret, because it is rather hard on Mrs Tarbard, who is a perfectly
innocent person in the matter—that she is liable for the injury sustained by Mrs Daniels through
drinking this bottle of lemonade.”
decisions are like the signposts which guide our journey: They point to
something which has sense—which is coherent—but we never get there for
all they do is point us on the way.20 Yet there is a minimal relationship
between the signpost (arguments) and the path (legal system) in the sense
that the signposts help us to make sense as a path.21
To travel towards the ideal called coherence, two basic tools have to be
packed: a comprehensive account of reasons and supportive structures. Is
the Court prepared to travel to coherence or are the necessary tools avail-
able to travel towards coherence alien to the Court or to European law? The
answer to this question has an immediate implication: To test whether
or not the Court complies with the notion of coherence proposed here.
However, the most interesting implication of this analysis is to escape from
a monistic reconstruction of the case-law of the Court, making all decisions
fit into a single principle, such as European integration, and thinking of it
as a criterion of coherence. This can only be achieved at the high price of
deforming and caricaturising the compromise between different (colliding)
principles and values that each decision contains. Moreover, because the
modest notion of coherence focuses the analysis of the Court’s case-law on
reasons and connecting reasons, the criticism of judicial activism of the
Court can be rejected. More precisely, the modest notion of coherence argues
that policy arguments have to be taken into account and connected to other
arguments (rights and values) in order to promote the coherence of the legal
system.
20
Bankowski, applying his metaphor to coherence in legal adjudication, points out that there
is a minimal relationship between the signposts—arguments—and the path—the legal system:
Signposts make sense of the path, in the same way as arguments make sense of the legal system.
The journey metaphor seems to provide valuable insights to analyse the relationship between
coherence of the law and coherence of the legal system.
21
I wish to thank Bankowski for this further elaboration of his metaphor.
29
In the case 155/91 Commission v Council the Court referred to the distinction between prin-
cipal and ancillary objectives of legislative measures.
30
Adding independent chains of reasons does not promote coherence. In Common Law,
however, the distinction between ratio decidendi and obiter dicta can be perceived as an addition
of reasons, although not fully independent ones. However, this distinction rather points to the
fact that chains of reasons can be differentiated (and isolated), for one has binding force and
the rest do not. The fact that several chains of reasons can be identified in a judicial decision
does not invalidate the requirement that these chains have to be connected in order to promote
the coherence of the decision.
31
There are conflicts between legal norms which claim validity, and conflicts of legal norms
which claim to be applicable to a particular situation. Conflicts concerning the validity of legal
norms are solved by a number of criteria: by source-oriented or process-oriented criteria in the
case of systemic validity (has the legal norm been enacted by an authority with competence
and following the procedure?); acceptance-oriented criteria (custom) and effectiveness-oriented
criteria in the case of factual validity; and content-oriented criteria in the case of axiological
validity (acceptance of the norm). The outcome of applying these criteria is the expulsion from
the legal system of the non-valid legal norm. Turning to the second category of conflicts, namely
conflicts concerning the applicability of valid norms, these are solved by taking full account of
the relevant aspects of a situation. The outcome is not the dropping of a norm from the legal
system, but rather establishing the priority of a norm in a particular situation.
32
Principles can be defined as optimized commands or as commands to optimize. The first cat-
egory involves a passive evaluation: What is the given weight of a particular reason; the second
category—principles as commands to be optimized—involves an active evaluation: What is the
weight of principles in a particular situation all-things-considered.
33
Aarnio (1997, 182) distinguishes between principles (e.g., free movement of goods);
principle-like rules (e.g., proportionality); and rule-like principles (e.g., the rule of reason).
34
Concerning the distinction between prima facie and pro tanto reasons, see Rabinowicz 1998,
and Kagan 1989.
reasons, principles apply in the same either-or fashion as rules: One princi-
ple (or principles) applies, and another (or others) does not. Those princi-
ples which have not been defeated are pro tanto reasons. Only as pro tanto
reasons can principles be understood as commands to optimize. Whenever
certain principles apply to a situation, a balance has to be found among
them. This means that some principles carry more weight and others less
weight. In other words, the weight of a principle will not undercut a com-
peting principle.
But what does it mean to say that some principles have more weight than
others? Indeed, the choice of the metaphor of balancing and weighing pre-
supposes the existence of a metric to measure the weight. This is not the
case, for principles do not contain a pre-established weight which is known
beforehand. Rather, the evaluation of the weight of principles35 depends on
the particular case. This proposal is extremely weak, but it is not irrational,
because it does not challenge the requirement of giving reasons to justify
why a certain aspect of a particular case is relevant.
The Court has included this theory of principles as commands to optimize
its model of legal reasoning. Indeed, in cases of conflict between rights (free
movement of goods) and goals (environmental protection), the Court
applies the rule of reason introduced in the Cassis de Dijon (case C-120/79
[1979] ECR 649) case. This rule contains a command to establish priority
orders between colliding Community principles (or interests); the outcome
should be neither disproportional nor discriminatory. Initially, the Court
referred to proportionality and discrimination as clear-cut criteria, which
determine the priority of colliding reasons in an either-or fashion. In this
sense, a proportional measure is that which least restricts the free movement
of goods, and a discriminatory measure is that which establishes a different
treatment between domestic and imported products. In recent judgments,
however, the rule of reason is thought of as a command to find a compro-
mise between colliding principles, that is, it is a command to establish pri-
ority orders which are the result of balancing colliding principles. This has
been possible because proportionality and non-discrimination are not
understood as definitive criteria, but rather as balancing criteria, that is, as
criteria which rationalize the process of evaluating the weight of reasons in
particular cases without imposing or determining the outcome. Balancing
criteria do not instruct the judge to find the less restrictive compromise, or
the non-discriminatory one in terms of distinction between domestic and
imported products. They rather instruct the judge to take account of the par-
ticularities of the case, to balance them, and to find a compromise between
them in such a way that “the more intensive the interference in one princi-
ple is, the more important must be the realization of the other principle”
35
Weight is the metric to use in the sense that one could balance principles against each other
to identify the one which is the heaviest, but there is no metric for weight, i.e., one would not
know how to measure heaviness, just that this one is heavier.
36
Priority orders are used to solve future similar collisions; they are thought of as rules because
otherwise the notion of balancing would be very weak (Aarnio 1997, 182).
particular, and the particular without closing the door to the universal.
Indeed, priority orders are the outcome of trying to encapsulate particulari-
ties into general rules, for the world is mediated by the concepts and cate-
gories of language and thought (Veitch 1998, 230). That is, priority orders are
the outcome of a process of abstraction of the particular in order to produce
categories which can be used to approach future cases. However, these pri-
ority orders—these categorisations of particularities—do not prevent the
judge from appreciating new features, and in this sense, creating a new pri-
ority order (as the former examples of the Court’s case-law show).
5. Reciprocal justification. This final property of supportive structures is
connected to the idea of plausible connections between reasons.37 Logic
support exists when p1 derives p2. This logic connection can also work in the
other direction, so that p2 supports p1. Both p1 and p2 are reciprocally justi-
fied. If that is true, then p1 and p2 are equivalent (Alexy and Peczenik 1990,
137); however, if all premises of a system are equivalent (all support each
other), then the system will contain one single premise. This conclusion is
absurd. To overcome this problem Alexy and Peczenik refer to the notion of
set of premises: The reason p1 together with r1 and s1 form a set of premises
from which p2 can be logically deduced. At the same time, say Alexy and
Peczenik, p2 might belong to another set of premises from which p1 derives
(Alexy and Peczenik 1990, 138).
The property of reciprocal justification, as proposed by Alexy and
Peczenik, highlights two ideas already mentioned. First, logical support is
not enough to justify: Plausible connections between reasons (belonging to
the same set of premises) promote coherence. Second, the best metaphor to
explain the structure of legal justification is the net, for, in a net of reasons,
reasons reciprocally support each other. The metaphor of the net of reasons,
although very convenient to connect reasons in a coherent manner, does not
solve the problem of circularity of a coherent justification (if p1 supports p2,
then p2 supports somehow p1). That is because circularity is accepted. Or
better, what is accepted is the fact that our knowledge of what amounts to
a good reason does not have unshakable foundations (Peczenik 1998, 11)
such as empirical data. It is also accepting that the demand for reasons is
infinite, especially since the demand for reasons does not aim at finding out
the ultimate basis of knowledge (since monistic notions of coherence have
been rejected, see ibid., 3). Circularity is accepted: The more reciprocal the
justifying structure, the more robust is the justification.
This is the requirement least fulfilled by the Court. Indeed, its decisions
follow very much a logic-deductive pattern which excludes circularity: The
focus is on the justification of the conclusion (the verdict), rather than on the
justification of the premises themselves; it also focuses on deductive con-
nections rather than plausible supportive structures. The lack of reciprocal
37
The notion of plausible connections is from Taylor (1971, 7ff.).
6 Conclusions
This paper argued that coherence is a key component of a non-foundation-
alist theory of law (in opposition to Dworkin), without reducing it to a mere
by-product (in opposition to Raz and Levenbook). This has been possible by
elaborating a modest notion of coherence in legal reasoning which instead
of dealing with the question “Is the ruling R coherent within the legal
system?” deals with the question “Does the argumentation of the ruling R
38
Does it mean that the judge at time t1 was incorrect? Or as MacCormick poses the question,
“Can judges make mistakes?” (MacCormick 1999, 76).
39
This statement is backed by the idea that the notion of validity is axiological.
of the reasoning of the Court, including the use of policy arguments. This is
so, since the Court takes a comprehensive account of reasons, which is nec-
essary to promote coherence in legal reasoning. The Court rightly refers to
authority reasons and substantive reasons: values and principles. Rather
than conceiving principles as rights conferrers and different from policies,
the Court consider principles as both arguments of rights and arguments of
policy. The Court does not approach rights and policies as distinct and
mutually opposed spheres, but rather as interlocking spheres, for both can
be connected in order to elaborate coherent supportive structures.
References
Aarnio, Aulis. 1987. The Rational as Reasonable. Dordrecht: Reidel.
——— . 1997. Reason and Authority. Cambridge, Mass: Harvard University Press.
Alexy, Robert. 1992. Rights, Legal Reasoning and Rational Discourse. Ratio Juris 5:
143–52.
——— . 1996. Theorie der juristischen Argumentation. Frankfurt: Suhrkamp. (1st ed.
1978.)
——— . 1998. Coherence and Argumentation. In On Coherence Theory of Law. Ed.
Aulis Aarnio, Robert Alexy, Aleksander Peczenik, Wlodek Rabinowicz and Jan
Wolenski, 41–9. Lund: Skrifter utgivna vid Juridiska Fakulteten i Lund n. 131.
Alexy, Robert and Aleksander Peczenik 1990. The Concept of Coherence and Its
Significance for Discursive Rationality. Ratio Juris 3: 130–47.
Bankowski, Zenon. 2001. Living Lawfully. Love in Law and Law in Love. Dordrecht:
Kluwer.
Bankowski, Zenon and Emilios Christodoulidis. 1998. The European Union as an
Essentially Contested Concept. European Law Journal 4: 341–54.
Dworkin, Ronald. 1986. Law’s Empire. London: Fontana.
Edward, David. 1996. Judicial Activism: Myth or Reality? In Legal Reasoning and
Judicial Interpretation of European Law. Ed. Angus I. L. Campbell and Meropi
Voyatzi, 29–67. London: Trenton.
Günther, Klaus. 1995. Legal Adjudication and Democracy. European Journal of
Philosophy 3: 36–54.
Habermas, Jürgen. 1996. Between Facts and Norms. Cambridge: Polity.
Hartley, Trevor C. 1999. Constitutional Problems of the European Union. Oxford: Hart.
Kagan, Shelly. 1989. The Limits of Morality. Oxford: Oxford University Press.
Levenbook, Barbara. 1984. The Role of Coherence in Legal Reasoning. Law and
Philosophy 3: 355–74.
MacCormick, Neil. 1984. Coherence in Legal Justification. In Theory of Legal Science.
Ed. Aleksander Peczenik et al., 235–51. Dordrecht: Reidel.
——— . 1994. Legal Reasoning and Legal Theory. Oxford: Oxford University Press.
(1st ed. 1978.)