VITTORIO VILLA
LEGAL THEORY AND VALUE JUDGMENTS�
ABSTRACT. The aim of the paper is that of putting into question the dichotomy
between fact-judgments and value judgments in the legal domain, with its epis-
temological presuppositions (descriptivist image of knowledge) and its method-
ological implications for legal knowledge (value freedom principle and neutrality
thesis). The basic question that I will try to answer is whether and on what condi-
tions strong ethical value-judgments belong within legal knowledge.
I criticize the traditional positivist positions that have fully accepted the value-
freedom principle and value-neutrality thesis, but I also submit to critical scrutiny
the new post-positivist views, that, even if they show interesting conceptual
developments on the matter, end up, nevertheless, by presupposing the same
epistemological image, which I call descriptivism. I stress that only by giving up
descriptivism and accepting constructivism as a general image of knowledge is
it possible to see the problem of value-ladenness of legal knowledge in a new
light.
On the basis of the constructivist image, I present two theses: firstly, at a
much broader epistemological level, I advance the minimalist thesis on value-
judgments, which simply removes the general ban on treating values as present
within knowledge; secondly, I advance the strong thesis on legal value-judgments,
which consists in arguing for the necessary presence of ethical value-judgments in
legal knowledge. I draw, in the end, some important implications from acceptance
of the strong thesis. One of these implications is a new distinction that replaces the
traditional distinction between fact-judgments and value judgments, namely, the
distinction between value-judgments exhibiting a cognitive function and value-
judgments exhibiting a creative function.
KEY WORDS: value-judgments, legal knowledge, descriptivism value freedom
principle, constructivism
� I wrote this paper during a period of research spent in the Centre for Law
and Society, Edinburgh University, thanks to an invitation by Zenon Bankowski
and Neil MacCormick, supported by the Lindsay Bequest in the Faculty of Law
of Edinburgh University. I would like to thank the Faculty most warmly for this
support, and also to thank very much my friend Neil MacCormick both for the
great help that he gave me in linguistic improvements to my first draft, and for
having read previous versions of the paper giving valuable advice and criticism.
Law and Philosophy 16: 447–477, 1997.
c
1997 Kluwer Academic Publishers. Printed in the Netherlands.
448 VITTORIO VILLA
1. INTRODUCTION
It has always been an influential and widely shared idea, in the
research program of legal positivism,1 that there is a sort of dichoto-
mous opposition between two kinds of discourses about positive law:
the first one is ‘descriptive’ in character, because its aim is informa-
tive: it tries, as a matter of fact, to give either theoretical explanations
of certain general features of positive law (from the point of view of
legal theory), or semantical reconstructions of some specific part of
a given legal system (from the point of view of legal dogmatics); the
second one is ‘evaluative’ in character, because it expresses, albeit
in different ways, some sort of attitudes or commitments, ethical,
political, ideological, or the like, on the given object.2
This opposition is by no means a necessary feature of legal posi-
tivism; yet, if it is accepted, a more specific methodological thesis can
1 In my opinion the notion of legal positivism has a much broader scope than
what is currently assumed, because it includes all the conceptions that presuppose
a certain concept of positive law, a concept according to which: i) law is always and
integrally an human product; ii) this product is radically contingent, in the sense
that its connection with morals, which could even be considered as ‘necessary’
for certain legal systems (this is the case of our current western legal systems)
regards values that are always context-bound, that is, values which can never
be objectively justified. From this point of view, there is a mutually exclusive
opposition between natural law and legal positivism (opposition which amounts
to an opposition between ethical objectivism and ethical relativism): both could
be defined putting the sign of negation before the opposite one (positivist or
naturalist). It is important to remark that, according to this definition, even legal
realism (and Dworkin’s theory) is a form of positivism. For this kind of definition of
legal positivism, see my book Conoscenza giuridica e concetto di diritto positivo.
Lezioni di filosofia del diritto (Torino: Giappichelli, 1993), pp. 105–108.
2 A classical formulation of this thesis is given by Kelsen: according to Kelsen,
legal science’s function does not consist in attributing values or in expressing
evaluations, but rather, in a description of its own object, description which is
performed leaving values out of consideration. See H. Kelsen, Reine Rechtslehre,
Italian. transl. (Torino: Einaudi, 1975 (3rd edn)), p. 84. According to another
leading figure in legal positivism, Hart, the separation, internal to the domain of
discourses referring to positive law, between conceptual analyses and evaluations,
represents one of the five meanings usually given to the notion of legal positivism.
See H.L.A. Hart, ‘Positivism and the Separation of Law and Morals’, Harvard
Law Review I (1958), pp. 593 ff. A very influential statement of this thesis,
in Italian analytical positivism, is that given by Bobbio. According to Bobbio,
compliance with the value-freedom principle constitutes one of legal positivism’s
three aspects. See N. Bobbio, Giusnaturalismo e positivismo giuridico (Milano:
Ed. di Comunita´, 1972), pp. 105–107.
LEGAL THEORY AND VALUE JUDGMENTS 449
be derived from it: the thesis according to which it is at least possible,
if not highly recommended, for legal theorists and jurists, to produce
purely descriptive discourses on positive law, that is discourses from
which every kind of value-judgment or of commitment, whether
political, ethical, or the like, is radically excluded. This thesis can be
– and is in fact – held by positivists independently of the different
ways in which these descriptive discourses might be characterized
(for instance, as scientific discourses, in a strong or in a weak sense,3
etc.). Hereinafter, this thesis will be called the ‘neutrality thesis’.4
It does not matter, for the tasks of this paper, to take into account
the different ways in which this thesis might be coordinated with –
or sometimes logically deduced from – other theses. In my opinion,
this thesis has anyway gained a quite canonical formulation, in con-
temporary legal positivism, in so far as it has represented one of the
ways of stating the separability thesis (its methodological version)
and has become, according to prescriptivist positions in analytical
legal philosophy, one of the (arguably logical) outcomes of the great
division5 in the legal domain.
In the space of this paper, however, I will not deal with these quite
complicated matters, because I think in any event that the neutrality
thesis can be analysed in its pure form, that is, through examin-
ing its proper epistemological and methodological justifications and
leaving out of consideration its possible logical or conceptual con-
nections. This thesis, in short, can stand on its own feet, as far as it
can be conceived as the legal version of the highly influential and
historically long-enduring methodological principle usually called
the principle of ‘value-freedom’, in its connection with the episte-
mological foundation given to it by empiricist images of science.
Seen in its pure form, therefore, the neutrality thesis incorporates,
as its basic epistemological foundation, the traditional empiricist
3 I have studied the problem of the scientific character of legal theorists’ and
jurists’ discourses in my book Teorie delle scienze naturali e teorie della scienza
giuridica. Modelli e analogie (Milano: Giuffre´, 1984).
4 I borrow this expression from K. Fu¨sser, ‘Farewell to ‘Legal Positivism’:
the Separation Thesis Unravelling’, in R.P. George (ed.) The Autonomy of Law:
Essays on Legal Positivism (Oxford: Clarendon Press, 1996), pp. 119–62.
5 See on this point Celano, who thinks that to accept the ‘great division’
logically implies refusing any place for value-judgments inside social sciences.
See B. Celano, Dialettica della giustificazione pratica. Saggio sulla legge di Hume
(Torino: Giappichelli, 1994), p. 49.
450 VITTORIO VILLA
opposition between fact-judgments and value-judgments, and, as its
methodological corollary, that set of rules which go under the head-
ing of the value-freedom principle. This will be the version of the
thesis that I will put under critical scrutiny in this paper.
Those who side with the neutrality thesis are ready to acknowl-
edge, of course, that descriptions and evaluations can easily occur
together in the practices of theorists and – more often – of jurists;
but they stress, nevertheless, that it is just one of the most significant
goals of legal philosophy to dispel this practical (perhaps unavoid-
able) inconsistency, establishing, with the maximum of clarity
and exactness, the proper conceptual boundaries between the two
domains.
It is important to note that, in this kind of perspective, legal value-
judgments are almost always equated with judgments expressing a
– positive or negative – stance on a given legal system, as a whole
or in one of its parts: in other words, according to this opinion, to
give a value-judgment implies almost always to take a stance from
the inside of the system, to express a practical attitude towards it.6
I am of the opinion that this is a big misunderstanding which it is
important to dispel; and I will try to do so in the paper.
The neutrality thesis presupposes, as can be easily shown, the
methodological monism that is so typical of neopositivism (‘the
same method, that of natural sciences, must be applied to all scien-
tific or quasi-scientific disciplines’). But it cannot be ignored that the
legal domain is, in some aspects, profoundly different from that of the
natural sciences; in the legal domain, to take an example that touches
directly our point of interest, it could happen (and certainly it happens
today, in western legal systems) that ethical values penetrate deeply
‘inside the law’ and so become part of the object of study, through the
intervention of the legislator (as contents of explicitly issued legal
principles), or of legal doctrine (as implicit legal principles recon-
structed by jurists), or, in the end, of the loyalist lay members of the
given legal community (who express their value-laden acceptance
of the system). But this situation doesn’t necessarily pose a threat
6 This kind of conviction is expressed very clearly by Ross, when he says
that the ‘descriptive terminology’ that is used in the doctrinal study of law “has
nothing to do with moral approval or condemnation” of a given legal system’s
rules. See A. Ross, On Law and Justice (London: Stevens & Sons Limited, 1958),
pp. 31–32.
LEGAL THEORY AND VALUE JUDGMENTS 451
for the neutrality thesis: here the methodological rule to be followed
by jurists should be, according to Bobbio:7 to treat these values and
value-judgments ‘as if they were facts’, with the same neutral atti-
tude as that supposedly adopted by scientists towards facts. And this
is considered by traditional legal positivism a viable strategy in the
legal domain, as far as jurists succeed in limiting themselves to con-
centrating their attention on the social and empirically accountable
sources through which these values and evaluations become part of
positive law or contribute to its effectiveness.
If we enlarge for a while the scope of the analysis, inserting
this thesis into the much broader context of the most general prob-
lems faced by legal theories, we can soon notice that the dichotomy
between fact-judgments and value-judgments does not stand alone,
in complete isolation; on the contrary, it is strictly connected (through
logical and conceptual relations) with other dichotomies, form-
ing a sort of an ‘interactive net’. These dichotomies refer to all
the spectrum of legal practices performed by theorists, jurists and
public officials (judges, above all), and divide invariably these
practices into two opposite classes, marked by contrasting labels
such as ‘interpreting existing law’ and ‘creating new law’ (in legal
interpretation), ‘describing law as it is’ and ‘prescribing law as it
should be’, ‘making discourses upon law’ (and so using interpretive
arguments) and ‘making discourses inside law’ (and so using produc-
tive arguments8). In all these dichotomies it is more or less presup-
posed that legal practices can be distinguished with the help of two
opposed qualifications, ‘objective’ and ‘subjective’, both interpreted
in a strong sense. From this point of view, all the descriptive dis-
courses have to be qualified as ‘objective’, and all the value-laden
discourses have to be qualified as ‘subjective’. It is worth noting,
by the way, that this kind of demarcationism produces the undesir-
able outcome of putting all the value-judgments (from ‘reports of
taste’ to ‘ethical judgments’) into the same class, characterized by
the possession of these supposedly subjective features.
7 N. Bobbio, Giusnaturalismo e positivismo giuridico (supra n. 2), pp. 125–26.
8 This distinction is drawn by R. Guastini, ‘Produzione di norme a mezzo
di norme’, in Etica e diritto, ed. by L. Gianformaggio and E. Lecaldano (Bari:
Laterza, 1986), pp. 175, 186–87. I have criticized this distinction in my ‘Legal
Analogy Between Interpretive Arguments and Productive Arguments’, in Legal
Knowledge and Analogy, ed. by P. Nerhot (Dordrecht: Kluwer, 1991), pp. 165–82.
452 VITTORIO VILLA
One of the claims that I will make in the paper is that this demar-
cation is not only epistemologically flawed, but that it also produces
unsatisfactory outcomes at political and practical levels. It goes with-
out saying that Dworkin is obviously a good ally in the battle against
this highly influential conceptual framework, even if he is not, unfor-
tunately, sufficiently clear and analytically rigorous in developing his
ideas, given that among other things he leaves the epistemological
and semantical presuppositions of many of his ideas too obscure.9
But it is better to go on in an orderly way, and to mention, before all,
the most general claims advanced in the paper.
First, I will make some epistemological claims that represent a
criticism of the general presuppositions of the neutrality thesis, and I
will proceed to suggest a provisional statement of a new perspective
on the matter. What I will try to show is, in particular, that the neutral-
ity thesis invariably presupposes a very influential and historically
long-enduring epistemological position, that I have already called, in
other works of mine, ‘descriptivism’,10 using an expression which
has close connections with Putnam’s ‘metaphysical realism’.11 It
is interesting to note that this position, which will be examined in
due course, is still taken for granted by most of the contemporary
positivistic legal theories, even if its implications for legal practice
are put into question by some of these theories. For example, even
when legal realists reject the supposedly objective elements of legal
discourses (rejecting, for instance, the idea that interpretation is to
be regarded as a semantical discovery of the meaning of rules),
their own approach is still dependent on that presupposed episte-
mological position. These skeptical theories, that is, do still rely on
this epistemological presupposition; what changes is only the legal
philosophers’ opinion about the applicability of this epistemological
9 Dworkin, anyway, points outs very clearly the need for sharply distinguishing
‘ethical claims’ and ‘mere reports of taste’ in legal domain. See R. Dworkin, Law’s
Empire (London: Fontana Press, 1986), p. 81. For a much broader distinction, epis-
temologically grounded, between objective and subjective value-judgments, see
H. Putnam, Reason, Truth and History (Cambridge Mass.: Cambridge University
Press, 1981), p. 156.
10 See again my book Conoscenza giuridica e concetto di diritto positivo (supra
n. 1), pp. 173–85.
11 Putnam exposes and criticizes this epistemological conception in many
places; but see particularly H. Putnam, Reason, Truth and History (supra n. 9),
pp. 49–50.
LEGAL THEORY AND VALUE JUDGMENTS 453
framework as an interpretive scheme for understanding legal prac-
tices. From this point of view, therefore, it is a workable hypothesis
to assume that legal skepticism, in most of its versions, is parasitic
on legal objectivism: with the difference that legal skepticism thinks
that objectivism is no longer a live option in the legal domain.
It is important to add, nevertheless, that this situation has quite
dramatically changed in recent years. Analytical legal positivism has
tried very hard, in this more recent period, to put into question one or
the other of these dichotomies.12 Here the solutions to be advanced
cannot be, of course, those of ceasing to draw distinctions among
different aspects of legal practices, but, rather, those of making them
more adequate, using more sophisticated tools (in my opinion we
should adopt, first of all, a different image of knowledge).
Anyway, the broad discussion on the sources and flaws of these
dichotomies cannot be the concern of this paper. Neither can the
general epistemological framework be put in question as the main
issue of the paper. It has been a main topic of previous work, and I
am currently re-engaging with it as a long-term research program.13
Coming back again to the changes occurring in contemporary analyt-
ical legal philosophy, perhaps it is worth anticipating what I will say
in a more detailed way later on: these changes of opinion, or at least
these new ways of approaching the matter of legal value-judgments,
do not succeed, in most cases, in avoiding confusions, inadequacies
and misunderstandings.
In this broader context, the more modest task of the paper is that of
making some observations, of epistemological and methodological
character, on the supposed opposition between descriptive or infor-
mative legal discourses on one side, and evaluative discourses on the
12 In recent years, for instance, there has developed a very interesting debate on
which is the most feasible way to work out a concept of objectivity for legal prac-
tices that could represent a viable alternative to both strong objectivist positions,
too much compromised with metaphysical realism, and subjectivist positions,
which share relativistic or conventionalist premises. Among these attempts, see
particularly J. Coleman & B. Leiter, ‘Determinacy, Objectivity, and Authority’, in
Law and Interpretation: Essays in Legal Philosophy, ed. by A. Marmor (Oxford:
Clarendon Press, 1995), pp. 203–78, and N. Stavropoulos, Objectivity in Law
(Oxford: Clarendon Press, 1996).
13 I can only make reference here to my recent book Conoscenza giuridica e
concetto di diritto positivo (supra n. 1) and to my new book Costruttivismo e teorie
del diritto (Torino: Giappichelli), which will be published later in this year.
454 VITTORIO VILLA
other side. For brevity’s sake I will assume, without discussion, that
these descriptive discourses can be considered as instances of what
may be suitably called ‘legal knowledge’. The centre of our concern
in the paper will therefore be that of the relationship between legal
knowledge and value-judgments.
I have said before that traditional legal positivism conceives this
relationship as a radical opposition.