Skip to navigation – Site map

HomeNuméros33Discussion with Bruno CelanoDo jurists need pre-conventions?

Discussion with Bruno Celano

Do jurists need pre-conventions?

Pierluigi Chiassoni
p. 37-43


The paper offers a comparison between the legal theory of normative facts on the one hand, and Bruno Celano’s theory of pre-conventions on the other, suggesting two ways that the latter may be of use to well-meaning jurists.

Top of page

Full text

1 The problem

  • 1 Celano 2016.

1In the sophisticated, tightly argued essay Pre-Conventions. A Fragment of the Background,1 Bruno Celano makes the following claims:

  • 2 Celano 2016: 30.

1. The “antithesis between ‘nature’ and ‘convention’”, one of the fundamental commonplaces (“topoi”) in Western philosophy, is “a bit too naïve”.
2. The naïveté is due neither to the absence of phenomena that are “unmistakably so by nature” (think about the fact that fire burns), nor to the absence of phenomena that are “unmistakably so by convention” (think about Art. 138 of the Italian Constitution).
3. The naïveté is rather due to the fact that the “two terms
are not mutually exclusive”. There indeed exist ‘normative facts’: that is to say, “phenomena of great importance for human life which do not fall exclusively in one or in the other category, while participating in both”.
4. “Pre-conventions” – the “embodied conventions that have become ‘second nature’” either to human beings as such, or to humans as constituents of local societies – are ‘of this’ hybrid ‘kind’.
5. Pre-conventions constitute a fragment of the (Searlean) ‘Background’: they belong to the set of unintentional phenomena making up the basis for the very distinction between nature and convention; they stay in a ‘space’ where, in Celano’s own words, we “may find the conditions allowing us to frame one phenomenon as purely natural, or purely conventional (or cultural)”.

2Celano’s problem consists, apparently, in furthering our understanding of the distinction between ‘nature’ and ‘convention’ (‘culture’, man-made ‘rules’) by casting light on what lies behind it and the way it works in our daily lives as humans and tribesmen.

  • 3 Bobbio 1994: 46-47. Bobbio (1994: 46) observes that, generally speaking, ‘normative’ means ‘having (...)

3I must confess that I cannot see why the distinction between ‘nature’ and ‘convention’ (‘culture’, ‘rules’) should depend on the ‘Background’ rather than on a carefully built and thoughtfully argued for conceptual framework (which, of course, may take into account whatever we may find in the Background to be of use). I shall leave this issue aside, though. Celano, as we have seen, considers pre-conventions to be ‘normative facts’ of sorts. At least since the time of Georges Gurvitch, normative facts have been part of the stock and trade of jurisprudence.3 The question I shall consider, from a strictly juristic standpoint, runs as follows: does Celano’s theory of pre-conventions provide new, fruitful insights into normative facts which jurists may take advantage of in their reflections on the ‘nature’ of law and legal sources?

2 Jurists on normative facts

  • 4 Bobbio 1994: 46-57. Voluntary and legally qualified facts of normative production are also consider (...)

4In an essay published some years ago, Norberto Bobbio identifies, upon fine analysis, four kinds of ‘normative facts’: (1) normative facts à la Gurvitch, (2) legally unqualified facts of normative production, (3) non-voluntary facts of normative production, and, finally, (4) non-voluntary and legally unqualified facts of normative production.4

  • 5 Bobbio 1994: 46, quoting a passage from L’idée du droit social (1932), where Gurvitch characterises (...)

51. Normative facts à la Gurvitch (faits normatifs) are social facts “intrinsically endowed with a legal value”; because of their intrinsic legality, they provide the “foundation” for “the binding force” of positive legal norms.5 This occurs since (and insofar as) “formal legal sources” – statutes, customs, judicial precedents – draw their content from such facts.

  • 6 Savigny 1840: §§ 4–5; Marini 1978: 174–180; Chiassoni 2016: 185–188.

6Bobbio, by the way, considers Gurvitch’s idea of normative fact to be ill conceived. He sees it as an attempt, laden with confusion, to solve the problem of the transition from fact to law, which consists in claiming that there are ‘places’ where, paraphrasing Vico, “ius et factum convertuntur” or where law and fact convert into each other or, in other words, come to coincide. A further comment on such a mysterious notion is, perhaps, in order: Gurvitch’s normative fact is, apparently, nothing but a sociological avatar of Friedrich Carl von Savigny’s idea of legal relationship (Rechtsverhältnis), a basic social phenomenon, a living element, which is the key to legal institutes, rules, and concepts.6

  • 7 Bobbio 1994: 49.

72. Legally unqualified facts of normative production are social facts that work, paradoxically, as legally un-authorised facts of legal production. They produce legal norms (they “have as an immediate consequence […] the production of a legal norm”) without being previously authorised to do so (“independently from a normative qualification”).7 Normative facts of this kind, it must be noticed, are sometimes qualified as extra ordinem legal sources: that is to say, sources that produce law, but are outside a positive legal order’s set of formal sources. The paradox in the former sentences disappears, however, as soon as one considers the clearest examples of a legally unqualified normative fact: revolution (which Bobbio expressly focuses on) and coup d’état (which Bobbio does not take into account). A successful revolution brings into effect a change in the constitution of a country, and this is (usually) not a legitimate constitutional revision method allowed by the pre-existing legal order (unless, of course, that legal order belongs to the club of suicide forms of government).

  • 8 Bobbio 1994: 51.
  • 9 Bobbio 1994: 48, 51–52. A ‘normative fact’ in this sense is a fact which “the legal order regards a (...)
  • 10 Bobbio 1994: 51-53.

83. Non-voluntary facts of normative production (“improperly called natural”, as Bobbio makes clear)8 are ways, recognised by legal orders, of producing “valid norms of the system”, which do not involve discrete intentional (voluntary) acts of production by determinate organs, occurring in a determinate time and place.9 Customary law – as the set of valid legal customs in a certain legal order – represents perhaps the clearest example of laws produced by means of non-voluntary normative facts.10

  • 11 “As a normative fact, necessity is, all at the same time, a non-voluntary fact and an unqualified f (...)

94. Finally, there are normative facts that are, at the same time, legally unqualified and non-voluntary facts of normative production. According to Bobbio, laws produced out of “necessity” can be regarded as laws produced by an unqualified and non-voluntary normative fact.11 We may notice, however, though in passing, that ‘necessity’ may be considered a law-making factor by a corresponding positive law principle of necessity (“Necessity makes law”). In that case, necessity becomes a legally qualified fact of legal production. When that is the case, producing law ‘out of necessity’ may be an extraordinary event for a society, but by no means something that is legally extra ordinem – think about the Roman law institute of dictatorship, or about the power of certain magistrates to issue decrees and orders for the purpose of ‘public necessity’, or in case of a ‘state of necessity’.

  • 12 From the standpoint of the international law principle of effectiveness, as considered, e.g., by Ha (...)

10By way of commentary on the views that Bobbio considers in his analysis of normative facts, it is worthwhile noticing that talking of revolution and necessity as ‘normative facts’, as legally unqualified facts that ‘produce legal norms’, is talking loosely. Indeed, on closer examination, such facts as revolution, usurpation (in the form of a coup d’état or in other forms), and necessity do not immediately produce norms. Rather, they are, properly speaking, facts that represent the condition for somebody – the revolution leaders, the usurper, the chiefs of the Golpista party, the sovereign – to become competent, if only by self-authorisation, to produce legal norms: regardless of whether they are valid norms of a newly established legal order, as is the case with revolution, usurpation, and coup d’état, or valid norms of the old legal order, as is usually the case with necessity.12

11The preceding remarks suggest that there are two different meanings of ‘normative fact’ at stake here. On the one hand, a ‘normative fact’ is a norm-producing fact proper: namely, a fact, the occurrence of which, according to some norm on the legal production of an existing legal system, has the effect of creating a legal norm. Think about the long-time repetition of a certain pattern of behaviour (accompanied by the normative attitude of it being required or permitted by law), which, according to the standard theory of custom, produces a norm of customary law. On the other hand, a ‘normative fact’ is a fact that triggers a legal competence to produce new law, that is the factual condition upon which a normative competence, a competence to produce norms, arises – irrespective of whether it is awarded by the pre-existing legal order or self-awarded.

  • 13 “In other words, a normative fact is a fact which brings into existence (or gives life to), and hen (...)

12The analysis of ‘normative fact’ offered by Bobbio turns, then, upon these two possible meanings of the expression. To be sure, a fact is ‘normative’ when it produces norms: when it is a norm-producing or norm-creating fact.13 Furthermore, a fact is ‘normative’ when, as we have just emphasised, it triggers a normative competence. A fact, however, can be ‘normative’ in (at least) two other different senses.

13To begin with, a fact can be ‘normative’ not because it produces norms, but rather because it is qualified by norms. In that case, a ‘normative fact’ is ‘normative’ because it is given sense, meaning or value, by some existing legal norm: more precisely, because it is an instance of the class of facts to which some valid legal norm connects a certain legal consequence in a legal system. In this sense, an individual act of depriving somebody else of her life or the fact that a certain building falls apart hurting some people are ‘normative facts’ from the standpoint of certain norms of criminal law and tort law. They are also facts that, typically, trigger normative competences awarded by the pre-existing norms of a system.

14Furthermore, a fact can be ‘normative’ when it is the index, or evidence, or representation of some existing norm. For instance, the pattern of behaviours usually performed by most members of a community by way of custom, with its typical, recurrent interplay of certain actions and reactions, is a ‘normative fact’ precisely in a sense in which it is the index, evidence, and representation of a certain customary norm which may be identified from it.

15The so-far legal analysis of ‘normative facts’ apparently leads to a conclusion which may seem bleak, but which, to my mind, shines with golden good sense. The conclusion may be phrased as follows: in order to account for the many different relationships between facts and law, facts and legal norms, we do not need to surrender to the confusing mysteries of romantic and post-romantic jurisprudence. We do not need to assume the existence of mysterious entities, which would simultaneously partake in both the factual and the normative. We have only to consider that there are facts – human behaviours and situations that are usually the effect of human behaviours – which either produce norms, or trigger competence norms (norms about norm production), or are qualified by norms, or are indexes, evidence, or representations of norms. We can always and profitably keep apart, for any individual event whatsoever, the factual and the normative. This keeping apart is, apparently, the key to understanding legal phenomena clearly.

3 Celano on pre-conventions

16Celano, as I have said at the outset, develops his search for pre-conventions as a fragment of the Background while investigating a fundamental of Western culture: the antithesis between ‘nature’ and ‘convention’. On closer and hopefully not erroneous examination, Celano characterises a pre-convention as follows. A pre-convention is a normative fact consisting in the convergent behaviour of some set of people, such that:

  • 14 Celano 2016: 13.
  • 15 Celano 2016: 15.
  • 16 Celano 2016: 13.
  • 17 Celano 2016: 19.
  • 18 Celano 2016: 22.

1. it is not the effect of an explicit agreement;
2. it is not liable to rational explanation, that is to say, it does not correspond to the reasoned pursuit by each of the people involved of their own goals;
3. it is rather the effect of the practical learning of some system of rules (like in disciplinary power situations, crawl training, natural language training, but also in the assimilation of the impalpable canons of (good) taste and style),
14 in such a way that “the body takes over”:15 i.e., the behaviour corresponding to that system of rules is performed by each of the agents involved in an automatic, unreflective, unintentional, rapid, fluid, effortless way, as if it were a biological regularity, so that people are (usually) not able to articulate the system of rules that their convergent behaviours put into practice;16
4. it is part (“a fragment”) of the normative unintentional background on which intentional conducts – leading to explicit or tacit rational conventions – take place: such a convention is part of a wide set of pre-conventions, which “are mostly in the background of our activities and thoughts, and […], passing usually unnoticed, delimit their spaces”;
17 and, finally,
5. it guides “action, fixing the
correct way to behave” in certain kinds of situations.18

17As I have stated at the outset, Celano regards pre-conventions as a species of normative facts; namely, in his view, a species of facts that partake in both nature and convention at the same time. It is worthwhile asking, by way of conclusion, whether there is any correspondence between Celano’s pre-conventions on the one hand, and jurists’ normative facts on the other.

18Apparently, pre-conventions are not the homologues of juristic normative facts either in the sense of being norm-producing facts or in the sense of being facts that trigger the competence to produce norms.

19From the standpoint of a normative conception of law and morality, the one that regards them as made of norms (rules, principles, standards) and considers norms to be linguistic entities, Celano’s pre-conventions are, contrariwise, of the same kind of juristic normative facts in the two latter meanings of the expression I have considered earlier (at the end of § 2). They can indeed be considered to be ‘normative facts’, both in the sense of being facts qualified by (social) norms and in the sense of being facts that are indexes, evidence, or representations of (social) norms. On the one hand, a pre-convention is a convergent behaviour of some group of people, and such behaviour gets its sense, meaning, or value from a set of background social rules. It may be, as Celano suggests, that the people involved in the convergent behaviour (the practice) do what they do as a ‘second nature’: as if it were a piece of biological behaviour, with them being unable to articulate the rules that they are applying. Nonetheless, from a genealogical standpoint, there are rules in the background: I would not be performing a crawl stroke if a set of rules defining crawl swimming were non-existent. On the other hand, a pre-convention is a convergent social behaviour that is an index, evidence, and representation of some set of background social rules.

20I may now provide an answer, however only a tentative one, to the question I have posed at the beginning. I think jurists can learn a lot from Celano’s theory of pre-conventions. In my view, the learning mainly consists in getting hold of two points.

21First, legal experience surely does contain legal pre-conventions, which are worth discovering and bringing to the fore. Legal sociologists, psychologists and anthropologists should set to that valuable task.

22Second, the practice of law, in its everyday workings, takes place on a background of extra-legal pre-conventions, above all and insofar as language is involved. Theorists of legal interpretation, who are familiar with Searle’s theory of literal meaning, do know that. Nonetheless, Celano’s essay is a powerful reminder and a stimulus inspiring us to get to a more detailed map of where and how such pre-conventions interfere.

Top of page


Norberto BOBBIO, 1994: Consuetudine e fatto normativo. In: Contributi ad un dizionario giuridico. Torino: Giappichelli. 17–57.

Bruno CELANO, 2016: Pre-conventions. A fragment of the Background. Revus – Journal for Constitutional Theory and Philosophy of Law (2016) 30: 9–32. Available at:
Pierluigi CHIASSONI, 2016:
Da Bentham a Kelsen. Sei capitoli per una storia della filosofia analitica del diritto. Torino: Giappichelli.

Hans KELSEN, 1949 (1945): General Theory of Law and State. Cambridge (Mass.): Harvard University Press.

Giuliano MARINI, 1978: Friedrich Carl von Savigny. Napoli: Guida.

Top of page


1 Celano 2016.

2 Celano 2016: 30.

3 Bobbio 1994: 46-47. Bobbio (1994: 46) observes that, generally speaking, ‘normative’ means ‘having some reference to norms’ and, more specifically, depending on context, ‘having norms as content’, ‘producing norms’, but also ‘made of norms’ and ‘being the product of norms’.

4 Bobbio 1994: 46-57. Voluntary and legally qualified facts of normative production are also considered. As Bobbio makes clear, however, they are, properly speaking, acts of normative production: such as legislation and any other piece of written-law making.

5 Bobbio 1994: 46, quoting a passage from L’idée du droit social (1932), where Gurvitch characterises normative facts as “certain real facts of social life that, due to their special structure /…/ can serve as foundation, as impersonal authority, for the binding force of the law”.

6 Savigny 1840: §§ 4–5; Marini 1978: 174–180; Chiassoni 2016: 185–188.

7 Bobbio 1994: 49.

8 Bobbio 1994: 51.

9 Bobbio 1994: 48, 51–52. A ‘normative fact’ in this sense is a fact which “the legal order regards as fit /…/ to produce valid norms of the system”.

10 Bobbio 1994: 51-53.

11 “As a normative fact, necessity is, all at the same time, a non-voluntary fact and an unqualified fact” (Bobbio 1994: 52).

12 From the standpoint of the international law principle of effectiveness, as considered, e.g., by Hans Kelsen (Kelsen 1945: 118 f., 350, 367, 370; Bobbio 1994: 54-55), the successful revolution leaders’ (usurpers’, or Golpistas’) self-authorisation to produce new norms (which is usually embodied in normative acts in the form of an “Organisation Statute” or a “Process Decree”) is turned into a legitimate act: it does, of course, remain illegitimate from the standpoint of the pre-existing legal order that has been destroyed by a revolution or a coup d’état.

13 “In other words, a normative fact is a fact which brings into existence (or gives life to), and hence has as a consequence or result, a legal norm” (Bobbio 1994: 47).

14 Celano 2016: 13.

15 Celano 2016: 15.

16 Celano 2016: 13.

17 Celano 2016: 19.

18 Celano 2016: 22.

Top of page


Bibliographical reference

Pierluigi Chiassoni, Do jurists need pre-conventions?Revus, 33 | 2017, 37-43.

Electronic reference

Pierluigi Chiassoni, Do jurists need pre-conventions?Revus [Online], 33 | 2017, Online since , connection on 30 July 2021. URL:; DOI:

Top of page

About the author

Pierluigi Chiassoni

Professor of Law at the University of Genoa, Italy

Pierluigi Chiassoni
Tarello Institute for Legal Philosophy
Via Balbi 30/18
16126 Genoa


By this author

Top of page


All rights reserved

Top of page
  • OpenEdition Journals
Search OpenEdition Search

You will be redirected to OpenEdition Search