JURISPRUDENCE
https://doi.org/10.1080/20403313.2020.1778289
Routledge
Taylor Si Francis Group
SYMPOSIUM
N Check for updates
Is legal certainty a formal value?
Isabel Lifante-Vidal C
Departamento defilosofia del Derechoy Derecho internacionál privado, University of Alicante, Alicante, Spain
ABSTRACT
Legal certainty is a central requirement for the rule of law. Legal systems should both enable those subject to law to predict human behaviour and institutional reactions and to prevent an arbitrary use of state power against them. The value of legal certainty is usually conceived as a formal value opposed to the values of freedom or equality (material or substantive justice). The purpose of this paper is to discuss this idea and to explore a less formal conception of the value of legal certainty. In this sense, I will argue that we need to depart from a purely formal understanding of the requirements imposed by such value and to consider the need to provide them with a substantive content. So, legal certainty is not completely independent of legal substantive justice and it implies that it is impossible to define ex ante which concrete aspects of justice should not be the law's concern.
KEYWORDS
Legal certainty; predictability; rule of law; justice; formal values; substantive values
1. Legal certainty as a legal value
I would like to begin by pointing out a linguistic discrepancy that exists between Spanish (and, in general, Latin) and English (and, in general, Anglo-Saxon) legal cultures.1 Although the expression 'legal certainty' is not unknown in the Anglo-Saxon legal tradition, in fact it is not so widely used as the expression 'seguridad juridica is in Spanish literature. In the latter, the expression is used to cover almost all the requirements which in the Anglo Saxon context are incorporated under a restricted (sometimes referred to as 'formal' or 'legalistic'2) concept of 'rule of law' (which, in turn, would coincide roughly with the requirements incorporated under the expression 'imperio de la ley3, in the Latin tradition). This restricted conception of the rule of law is the one maintained by, among others, Raz, for whom
CONTACT Isabel Lifante-Vidal @ isabel.lifante@ua.es
^or a detailed analysis of these linguistic discrepancies, see Ricardo Garcia Manrique, 'Acerca del valor moral de la segur-idad juridica' (2003) 26 Doxa. Cuadernos de filosofia del Derecho 477 and César Villegas Delgado, 'La sumisión del poder publico al Derecho en el civil law y en el common law: Estado de Derecho, rule of law y su expansion al ámbito internacionál' (2013) 137 Boletín Mexicano de Derecho Comparado 713.
2For the characterization of this formal (or legalistic) conception of rule of law and its distinction from a substantive (or non-legalistic) one, see Paul Craig, 'Formal and Substantive Conceptions of the Rule of Law: An Analytical Framework' (1997) 3 Public Law 467 and Nicholas Barber, 'Must Legalistic Conceptions of the Rule of Law Have a Social Dimension?' (2004) 17 Ratio luris 474. This designation is criticized by John Gardner, 'The Supposed Formality of the Rule of Law', in John Gardner ed, Law as a Leap of Faith: Essays on Law in General (OUP 2012), who doubts on the suitability of these labels to refer to the features included in the ideal of the rule of law, at least as developed by Fuller or Raz.
3Cf., Rafael Escudero Alday, Positivismo y moral interna del Derecho (Centra de Estudios Políticos y Constitucionales 2000) 444.
© 2020 Informa UK Limited, trading as Taylor & Francis Group
2   @   I. LIFANTE-VIDAL
[t]he rule of law is a political ideal which a legal system may lack or may possess to a greater or lesser degree. That much in common ground. It is also to be insisted that the rule of law is just one of the virtues which a legal system may possess and by which it is to be judged. It is not to be confused with democracy, justice, equality (before the law or otherwise), human rights of any kind or respect for persons or for the dignity of man.4
According to Raz, the literal sense of'the rule of law' has two aspects: (1) people should be ruled by the law and obey it, and (2) the law should be such that people will be able to be guided by it. This second aspect is the most important for Raz's theory:
if the law is to be obeyed it must be capable of guiding the behaviour of its subjects. It must be such that they can find out what it is and act on it [... ] It is evident that this conception of the rule of law is a formal one. It says nothing about how the law is to be made [...] It says nothing about fundamental rights, about equality, or justice.5
Anyway, as Paul Craig has remarked,6 the adscription of Raz to a formal conception of rule of law it is not that clear, since in a posterior work he stated that the protection of basic civil rights is partly presupposed and partly implied by his understanding of the rule of law.7
In Spanish legal literature, the expresion 'seguridad juridica' - that I am going to translate here as 'legal certainty' - refers to one of the fundamental values of our legal system (along with others such as equality and freedom). It is an ethical-political value and one of the requirements of the ideal of the Rechsstaat. Legal certainty as a principle (or value) thus exists on a more abstract plane than other principles such as legality, hierarchy of rules, publicity, non-retroactivity, accessibility, reliability, regulatory stability, res judicata (and many others such as ignorantia juris non excusat), all of which embody regulatory requirements that derive from acknowledging the value of legal certainty. But what is that value? And why is it desirable?
When first approached, legal certainty could be understood as the capacity granted to us by the law to foresee human actions and their consequences (the institutional response to our actions) to a certain extent.8 Accordingly, accepting legal certainty as a value is tantamount to considering its existence as positive and, in this regard, accepting that we have good reasons to strive to attain or, better still, to maximise it.
This positive evaluation is often based on the consideration that the predictability offered by legal certainty is instrumentally necessary (though not sufficient) for the development of personal autonomy. The existence of standards which provide us with a certain
4Joseph Raz, The Authority of Law (Clarendon Press 1979) 211. 5ibid 214. 6Craig (n 2) 485.
7See Joseph Raz, Ethics in the Public Domain, Essays on the Morality of Law and Politics (Clarendon Press 1994) 376.
definition taken from Manuel Atienza, Introduccion al Derecho (Barcanova 1985) 116. Atienza distinguishes three levels of certainty: order [orden], security [certeza] and certainty stricto sensu [seguridad juridica en sentido estricto]. On the first of these levels, predictability would be given either by the law or by any other system of social control (religion, social mores, etc.). The second level, security, is much more demanding, and is linked to a series of principles recognised under developed legal systems (legality, publicity, res judicata, ignorantia juris non excusat, etc.) as well as to their system-atisation. Lastly, the level which Atienza refers to as legal certainty stricto sensu invokes the capacity of a certain body of laws to make the values of liberty and equality predictable or, in other words, certain. From this point of view, this third level of legal certainty, which encompasses the first two, is conceived by Atienza as secondary to the other two components of justice (freedom and equality).
JURISPRUDENCE   @ 3
degree of predictability in our social relations is a necessary (though not sufficient) condition for the development of personal autonomy.
Joseph Raz has also claimed, in a similar sense, that the 'observance of the rule of law is necessary if the law is to respect human dignity'.10 Respecting human dignity entails treating humans as persons capable of planning and plotting their future. Thus, respecting people's dignity entails respecting their autonomy, their right to control their future:
The law can violate people's dignity in many ways. Observing the rule of law by no means guarantees that such violations do not occur. But it is clear that deliberate disregard for the rule of law violates human dignity. It is the business of law to guide human action by affecting people's options [...] The violation of the rule of law can take two forms. It may lead to uncertainly or it may lead to frustrated and disappointed expectations. It leads to the first when the law does not enable people to foresee future developments or to form definite expectations (as in cases of vagueness and most cases of wide discretion). It leads to frustrated expectations when the appearance of stability and certainly which encourages people to rely and plan on the basis of the existing law is shattered by retroactivity lawmaking or by preventing proper law enforcement, etc.11
Any analysis of legal certainty must then be based on its consideration as a value, insofar as it is an instrument towards other goals which we also consider to be valuable: in individual terms, it allows us to develop personal autonomy, and in social terms, it forms part of the institutional framework (the rule of law) that makes it possible to develop human rights, that is, justice.
However, one characteristic usually attributed to legal certainty is its formality, although it is unclear what this refers to, or where the distinction between formal and substantive values lies. I believe that there are at least two possible explanations for this dichotomy.
A first interpretation attributes the term formal to those values whose development depends only on the formal aspects of the law, regardless of their substance. I am aware that the term formal does not clarify a great deal. I use it here in relation to the four categories of formality distinguished by Atiyah and Summers:12 (1) authoritative formality, according to which both validity and hierarchy are dependent on the source of origin; (2) content formality, which refers to the degree to which the content of a legal norm is determined by fiat rather than by direct reference to any substantive reasoning; (3) interpretive formality, which refers to the priority given to literal interpretation over interpretative methods which allude to underlying goals or values; and (4) mandatory formality, which refers to the extent to which otherwise substantive considerations may be overridden.
9See Francisco J Laporta, El imperio de la ley. Una vision actual (Trotta 2007). 10Raz (n 4) 221. "ibid 221-22.
12Cf., PS Atiyah and Robert Summers, Form and Substance in Anglo-American Law (Clarendon Press 1987) 11 etseq.; Robert Summers, La naturaleza formal del Derecho (P Larranaga tr, Fontamara 2001) 113-20. It should be pointed out that, as far as these two authors are concerned, formality (in any of its meanings) is a question of degree, and that reasons may have varying levels of formality and/or substantiality. A legal reasoning is formalist when it degenerates towards one or more of the formal aspects of the law, to the detriment of its substantive aspects. As it happens, both authors reject excessive insistence on form and on substance equally (cf., Atiyah and Summers, ibid., 28 etseq.) In a later work, Summers distanced himself somewhat from the form/substance dichotomy, outlining a much broader definition of the significance of form in the law, though maintaining his criticism of the excessive formalism denounced in his 1987 work with Atiyah. Cf., Robert Summers, Form and Function in a Legal System. A General Study (Cambridge University Press 2006) 61, note 32.
4   @   I. LIFANTE-VIDAL
In a second, and not totally unrelated interpretation, the contraposition between formal and substantive values aims to underline the idea that certainty, as a value, exists independently from substantive values, i.e., those which embody the ideal of justice (freedom, equality, etc.), insofar as certainty may be a feature of any fundamentally fair or unfair system.
In my view, these two interpretations of the implications of the formal nature of legal certainty run the risk of being exaggerated, giving rise to an unacceptable formalism.14 And in this regard, the aim of this work is precisely to defend a concept of legal certainty in which its formality is relativised, as well as the sharp distinction between different legal values.15 Some authors criticise this more substantialised concept, as it would imply including any evaluative requirement under the umbrella of legal certainty. This would thus lose its own particular references and become synonymous with justice, hindering the work of clearly identifying the different values and, in the case of conflict, balancing them.16 However, I believe that this conclusion does not follow; it is possible to show how the substantive content of legal norms - and not just their formal aspect - affects predictability and how the optimum development of the value of legal certainty is not independent of other legal values, without confusing it with those values.17
This idea should lead us to relativise the distinction between aspects of justice that should or should not be the law's concern: the determination of what expectations have to be taken into account and protected by the rule of law cannot be determined ex ante with exclusively formal parameters. The law cannot be seen as a closed system, which does not mean, obviously, that each and every aspect of material justice must be a matter of law, but rather that it could become so.
2. The dimensions - and requirements - of predictability18
We now analyse the essence of predictability, that situation which legal certainty demands be generated or, better still, maximised. Predictability, in general, refers to the capacity to anticipate events. In the case of legal certainty, the events to be foreseen are human actions
13This idea occasionally is presented as distinguishing between formal and substantive justice. The first would refer only to the correct application of pre-existing legal norms to specific cases, while the second would refer to the justice of the norms. According to this interpretation, legal certainty would thus be connected only to the former.
14According to Atiyah and Summers (n 12) 28 et seq., a judicial reasoning is formalist when one or more of the formal aspects present in the law degenerate to the detriment of its substantial aspects. Both authors reject excessive insistence on form and on substance equally.
15The same position is held in Spain by Atienza, Introduction al Derecho (n 8) and Antonio-Enrique Pérez Luňo, La seguridad juridica (Ariel 1991). Atienza's proposal consists of defending three different levels of legal certainty, each involving different requirements. The maximum connection with other values occurs at the final level (in which the different values are reconciled). Similarly, Escudero Alday (n 3) 503 et seq. defines two levels of legal certainty; the first (which he calls 'certeza') would be unrelated to matters pertaining to the justification (of source and content) of the law, while the second (for which he reserves the expression 'seguridad juridica') would include respect for the basic rights. My thesis, however, is that a certain materialisation of the requirements of this value also occurs on the level which these authors refer to as 'certeza'.
16ln this regard see also Raz (n 4); Robert Summers, 'A Formal Theory of the Rule of Law' (1993) 6 Ratio luris 127; Federico Arcos Ramirez, La seguridad juridica: una teória formal (Dykinson 2000) 395-403; Garcia Manrique (n 1) 484.
17Here I should point out that from an integrating or holistic viewpoint, such as the one to which I would subscribe, cognizance of certain values will depend necessarily on others, and all are supported by values which we could consider to be ultimate. This vision would be very close to that maintained by Dworkin, for example, and is developed in greater detail in one of his later works. Cf., Ronald Dworkin, Justice for Hedgehogs (Belknap Press of Harvard University Press 2011).
18The ideas presented here have been developed by me in greater detail in some previous work. See Isabel Lifante Vidal, 'Seguridad juridica y previsibilidaď (2013) 36 Doxa. Cuadernos de filosofia del Derecho 85; 'Ignorancia de la ley y seguridad juridica' (2015) 18 Teória & Derecho. Revista de pensamiento juridico 16.
JURISPRUDENCE   @ 5
and their consequences (the situation resulting from those actions). This may refer to natural actions (such as driving on the right or on the left) or institutional actions (legal acts), performed either by individuals (e.g., a dismissal) or by legal bodies (imposing a fine, awarding compensation, granting or denying a permit, etc.).19
Predictability may exist to a greater or lesser extent. This in no way implies that there is a metric that can be used to quantify degrees of predictability,20 among other things because predictability does not exactly imply that actions or their consequences have actually been predicted, but merely that they could have been, i.e., that the conditions are in place for the consequences to be anticipated.
Predictability may be imagined as a continuum, with absolute certainty at one end and absolute uncertainty at the other. With regard to human behaviour, we always find ourselves somewhere in the middle, as the social practices that operate as stabilising factors generate a certain degree of predictability. Moreover, in the case of the law, we find norms that, to a certain extent, must be accessible, clear, stable, enforceable, non-retroactive, non-contradictory and applied consistently and regularly;22 and these are all characteristics that help to increase predictability. In this regard, the law could be seen as a social practice that is particularly useful for the generation of predictability.
However, when we say that a legal system (or a certain sector of it) can provide a greater or lesser degree of predictability, what are we really comparing? To answer this question, we should be aware of a second characteristic of predictability, one that is not so unanimously invoked, i.e., that as well as existing in degrees, it is also complex, insofar as it projects in several dimensions, each of which, in turn, also exists in degrees. In particular, I believe that predictability exists in three dimensions:23 (1) the objective dimension: what is it that can be predicted? (2) the subjective dimension: who can make the prediction? and (3) the temporal dimension: when can it be predicted to occur? As we will see below, each of these dimensions justifies different regulatory requirements.
19Many analyses of legal certainty have focussed essentially on the predictability of the decisions of the judicial bodies. However, we should not forget that we are also interested, perhaps fundamentally so, in directly predicting the actions of individuals (independently of whether these are later aired in the courts): giving assistance in case of need, driving on the right, complying with a clause in a contract, etc. The configuration of the law as a coactive order contributes to generating this type of predictability, which we could classify as primary; but many legal requirements are also justified, at least in part, by this interest in being able to predict citizens' behaviour. Thus, for example, the general requirement of good faith, the doctrine according to which, in the business world, nobody can go against their own acts, or the principle of ignorantia legis non excusat.
20ln my opinion this would be an enterprise doomed to fail, at least from the perspective of its regulatory aspects (a sociological study on the perception of predictability, for example, would be another story). However, Gometz maintains otherwise, expressly, as it seems, setting out to measure the degree of predictability attained by different legal systems. Although I do not agree with Gometz on this point, many of the arguments outlined here regarding the dimensions of predictability are directly inspired by his captivating work, although the reader will discover that the dimensions discussed may not coincide exactly with those proposed by the author. See Gianmarco Gometz, La certeza juridica como previsibilidad (D Moreno Cruz and D dei Vecchi tr, Marcial Pons 2012).
21 Some refer to this idea by pointing out that this property is dispositional, like the 'solubility' of sugar (in the regard, see the detailed analysis made by Gometz (n 20) 157 et seq.
22These characteristics would go to make up what Lon L Fuller, The Morality of Law (Yale University Press 1969), called the internal morality of the law.
23We could add a fourth dimension: reliability, or the probability of predictions being correct. In this fourth dimension, a legal system is more predictable the more reliable the predictions it allows us to make. Although this dimension is very important, it depends fundamentally on the degree of efficacy of the law, here taken for granted. In this regard, Perez Luho (n 15) outlines the need, alongside what he considers to be the structural requirements of legal certainty, for also complying with its functional requirements, which are, precisely, those concerned with the efficacy of the law (which is, ultimately, what will determine the reliability of the predictions that the law allows us to make).
6   @   I. LIFANTE-VIDAL
2.7. What can be predicted? The precision requirement
First of all, a legal system (or, better still, a certain sector of a legal system) will be more predictable the more things it allows us to predict and the more accurately it does so. Thus, determining the degree of predictability in this first dimension will depend on whether it is possible to know precisely beforehand which behaviours are prohibited, mandatory or permitted by the law and the legal consequences established for certain behaviours (or for certain situations), as well as the conditions under which those consequences, which may be procedural, temporal, economic, or of any other nature, may be generated.
In this regard, we could, for example, say that a legal system which predicts that the consequence of a certain action will be prison is less predictable than one which predicts a sanction of one to six years in prison. And this, in turn, is less predictable than another which predicts that the sentence will be four to six years, and so on. However, we should not forget that this objective dimension of predictability is, in turn, complex and encompasses many aspects which may also be of relevance to predict. Leaving criminal law aside, we may imagine that, in a certain situation (for example, damage caused to us by a neighbour), we may be interested in foreseeing, more or less accurately, the course of a hypothetical judicial decision (whether or not it is favourable to our interests), its content (the sum of the compensation), the time it will take to be adopted (or become definitive), the time during which the consequences of the situation may be appealed against (is it time-barred? and on what does the answer to that question depend?), the financial cost of the trial (are there fees/expenses to be paid?), etc.
Most analyses of legal certainty tend to link the degree of predictability exclusively to the presence of the four formal features of the law already mentioned: formality of source, formality of content, interpretive formality and mandatory formality. From this perspective, the law will be more predictable if its norms, which must be identified by their authoritative source or origin, can be applied independently of the substantive reasons behind them (for which they must adopt the form of rules of action and have complete semantic autonomy24) and their interpretation is limited to simply confirming their literal meaning. The ideal situation would then seem to be one in which only merely deductive reasoning would be required to apply the law. Any evaluative term, any attribution of a discretionary power, or even any interpretive activity, would undermine this conception of predictability.
However, the requirement that all decisions be predictable through merely subsumptive procedures, as well as being impossible, is not even desirable.25 Of course, this is not to be stated from predictability tout court. Now, the advisable kind of predictability to aim for in certain cases might be distinct from that formal sort. This non-formal predictability will
See Ruiz Manero's criticism of Laporta's vision regarding the requirement that legal rules enjoy complete semantic autonomy in order to attain the 'ideal' that applicative reasoning be deductive. Cf., Juan Ruiz Manero, 'Las virtudes de las reglas y sus Ifmites. Una discusion con Francisco Laporta' in M Atienza and J Ruiz Manero (ed), Para una teoria postpositivista del Derecho (Palestra-Temis 2009) 188.
'Along with all the problems of unintentional indefiniteness that abound in law (ambiguity, vagueness, contradictions, gaps, etc.), we also find - as Hart pointed out - that absolute definiteness would not be desirable either. Given that it is impossible to know the properties of all future cases - and their combination - in advance, it is also impossible to determine what the relevant aims will be. So it appears 'the need to leave open, for later settlement by an informed, official choice, issues which can only be properly appreciated and settled when they arise in a concrete case'. Herbert LA Hart, The Concept of Law (2nd edn, Clarendon Press 1994) 128.
JURISPRUDENCE   @ 7
imply a degree of flexibility in legal systems (and for these purposes a low degree of formality concerning their content). As an example, we could take those cases in which it is more important to predict that a certain interest (e.g., the wellbeing of a child) will be protected, than the details of how that will be done. In these cases, the law may regulate behaviour by means of delegating discretionary powers, so that those who apply the law take the most suitable measures, according to the circumstances of each case.26 In order to be applied, these powers require reasoning, to adapt the end to the means and weigh the different reasons in play, going beyond mere subsumption. However, and this is the interesting part, they are not exempt from the possibility of rational control and, in this regard, retain a certain degree of predictability. The effective generation of predictability will depend here - though not here alone - on having developed a social practice favourable to the requirements of this rational control over the exercise of power.
On the other hand, there are times when the requirement that the reasoned application of the law demands only deductive reasoning based on pre-existing normative premises also fails to ensure predictability. This would be the case, for example, when aid is granted according to a very formalised scale, but the total amount to be distributed is fixed, so that the aid that each recipient is going to receive will depend on the total number of applicants (a factor which may be totally unpredictable for each applicant).
2.2. Who can predict? The accessibility requirement
The second dimension, the subjective one, refers to what we consider the extension of predictability: those in a position to make the predictions. In this regard, a legal system will generate a greater degree of predictability the easier it can be used by citizens to make predictions. The first requirement linked to this dimension is obviously that of making legal norms public and accessible. A secret rule, or one to which access is restricted, will completely destroy predictability in this dimension.
In this case, of course, predictability is also variable, being dependent upon a series of diverse factors which may make access easier or more difficult, such as when norms are published only in subscription channels or on noticeboards located inside offices with restricted opening hours, etc. However, the fact remains that, alongside this formal accessibility, there also exists a more substantive dimension, one which refers to citizens being able to effectively know and understand the applicable law. When referring to this real possibility of knowing the norms, some authors invoke transparency, referring to something more than the mere formal possibility of access, i.e., that the norm not only be known, but also understood by those to whom it is addressed. Many factors can play a role here, such as clarity of wording, level of technical complexity, legal references, the dispersion of regulatory competence (having to know beforehand which public authority is competent in a certain matter), the proliferation of so-called omnibus statutes (those which regulate very disparate matters), etc. These factors can all make it very difficult, or practically impossible, for the lay citizen (and even non-specialised jurists) to be
In a previous work I characterised the requirements for the exercise of discretionary powers to be justified. Cf., Isabel Lifante Vidal, 'Poderes discrecionales' in A Garcia Figueroa (ed), Racionalidad y Derecho (Centra de Estudios Polfticos y Consitucionales 2006).
8   @   I. LIFANTE-VIDAL
acquainted with the applicable rule or understand it in detail. In this regard, the words of Joaquin Costa y Martinez in early twentieth-century Spain come to mind: first of all, nobody knows the law in its entirety, and only an insignificant minority know a part -and not a large part - of the applicable laws at any given time. Moreover, it would be impossible for the majority, or even for this minority, to know them all. Accordingly, Costa y Martinez considered that labouring under the 'presumed knowledge' of the law would be 'an absolute travesty and the greatest tyranny ever exercised in history'.28 In his opinion, the presumption should be the exact opposite.
The same reasons that make it impossible to know the law completely also confer particular importance on a factor that generates legal certainty: the regulatory coherence offered by legal principles.29 If, like Llewellyn,30 we make a distinction between the certainty of laymen and the certainty of lawyers, we have to admit that, in the case of the layman, legal predictability does not only depend on the total predetermination of the content of legal decisions by means of applying pre-existing rules. What also and above all guides the general population and what, therefore, contributes to generating the necessary predictability for their plans to be developed satisfactorily, is, in particular, the recognition of the legal principles that underlie those rules, and ultimately the congruence between those legal norms and the ways of life - i.e., the proximity between legal norms and social norms.31
Waldron uses very similar terms with regard to another of the requirements for legal certainty: the principle of prospectivity. In his opinion, this principle, which precludes retroactive or retrospective laws,32 is based on requirements having to do with the systematic nature or integrity of the law, rather than on the idea that only prospective law can really guide the conduct of individuals. In this regard he maintains:
I have been trying to emphasize the point of systematicity. Our law works as a system; and it works to the extent that the integrity of the system can be held together. The principle of prospectivity - that we should make law in a forward-looking way, not retroactively or retrospectively - along with other Rule-of-Law values is key to that systematic integrity.33
Cárcova has referred to this phenomenon, which encompasses all of the difficulties of knowing and properly understanding the contents of the law, as the opacity of the law. Cárcova's analysis also incorporates cultural, ethnographic and anthropologic factors, which would affect the proper understanding of the law in multicultural societies. Cf., Carlos M Cárcova, La opacidad de Derecho (Trotta 1998).
28Joaq uín Costa y Martinez, El problema de la ignorancia del Derecho y sus relaciones con el estatus individual, el referendum y la costumbre (Civitas 2000) 31.
29The role of principles as generating legal certainty has been outlined, among many others, by Habermas, who picks up from Dworkin on this point. Cf., Jürgen Habermas, Facticidad y validez. Sobre el derecho y el Estado democrático de derecho en términosde teória deldiscurso (M Jimenez Redondo tr, Trotta 1998) 272 et seq. In Spain, this role has been emphasized particularly by Manuel Atienza and Juan Ruiz Manero, 'Sobre principios y reglas' (1991) 10 Doxa. Cuadernos de filosofía del Derecho 101.
30Karl N Llewellyn, Präjudizienrechtund Rechtsprechung in Amerika. Eine Spruchauswahlmit Besprechung (Weicher 1933) 58.
31Cf., Lon L Fuller, 'American Legal Realism' (1934) 82(5) University of Pennsylvania Law Review 429.
32Jeremy Waldron establishes a distinction between what the law considers retroactive and retrospective: 'Retrospective legislation is legislation that attaches some legal consequence now and for the future to an event or transaction that took place in the past [...] Retroactive legislation is more radical. A retroactive law is one that operates on past events as though it were in force when the past event took place [...] it would be a mistake to suppose that what I am calling retroactivity is necessarily worse than retrospectivity [...] Increasing the penalty is retrospective, because what it does is attach a new consequence for now to an action that society already had the right to punish. Imposing a penalty on an act previously regarded as innocent is retroactive, because it changes the status of the action ex post facto from innocent to criminal. The distinction is a delicate one and one can imagine a number of intermediate cases where it would not be so clear'. Jeremy Waldron, 'Retroactive Law: How Dodgy was Duynhoven?' (2004) Otago Law Review <www.nzlii.org/nz/journals/OtaLawRw/2004/8.html> accessed 19 September 2018.
33ibid.
JURISPRUDENCE   @ 9
The usual justification for the condemnation of retroactivity deals with the proposition that it is impossible to use rules to guide people's conduct if the rules are not prospective. However, as Waldron points out, a retrospective norm often does not affect the possibility of guiding behaviour. Firstly, because what really guides people's behaviour are their moral beliefs; and, secondly, because there are many norms which aim not to guide the conduct of citizens, but to correlate certain consequences to certain states of affairs.
If we are conscious of the importance of principles with regard to legal certainty, we will also understand how substantive, and not just formal, reasons play a fundamental role in generating predictability. In this regard, it is these same reasons (which, in any case, are also based on the certainty requirement) that lead us to accept teleological and evaluative arguments in legal interpretation35 - in any legal system, independently of whether or not they are expressly foreseen in an authoritative source - justifying the displacement of one of the types of formality (interpretive formality); or to accept what we could call atypical illicit acts3,6 (such as the abuse of rights), and which would mean displacing another type of formality, namely mandatory formality.
Until now we have maintained that a certain legal system is more predictable the more it allows those who are subject to it to make predictions, but in reality what is important is not the number of subjects capable of making an abstract prediction but how many of the relevant subjects (or, in other words, subjects potentially affected by the regulation) are capable of doing so. In this regard, we should bear in mind that some regulations typically affect certain classes of subjects, with conflicting interests and different possibilities of real access to information (and not only in the sense of formal possibilities): citizens/the administration; consumers/suppliers; employers/employees, etc. At times the same legal norm may generate different degrees of predictability in each kind of affected subjects. Take for example consumer protection law, which appears to be designed precisely on the basis of this conflict between kinds of subjects with different levels of access to regulations and to the information needed to determine their consequences. It should also be borne in mind that predictability may be predicated on very different types of events: primary conduct (e.g., a dismissal or breach of contract), judicial decisions (determining whether a dismissal is fair or unfair, establishing compensation, etc.), the moment in time when the decisions are made or take effect, etc. It is important to take into account that each kind of individuals affected by a certain regulation may be fundamentally interested in the predictability of certain aspects and that, sometimes, the regulations may increase the relevant predictability for one of the parties at the cost of decreasing it for the other.
We may also consider that the well-known principle that 'ignorantia juris non excusaf allows us to distinguish between two types or kinds of subjects whose predictability may be affected differently. On the one hand we have the unknowing subject, unacquainted with the content of the norms that are to be applied to him and regarding whom, obviously, this principle may imply less predictability. However, on the other hand, there may be third
34MacCormick also points out that even though the stability and certainty demanded by the rule of law are often set against the rhetoric or argumentative nature of the law, the important role played by coherence should not be overlooked. Cf., Neil MacCormick, 'Retorica y Estado de Derecho' (1999) 33 Isegorfa 5.
35This would of course be the thesis held by Dworkin, for whom it is precisely the nature of the interpretive activity which implies recurring to this type of argument. Cf., Ronald Dworkin, Law's Empire (Belknap Press 1986) 52 etseq.
36This terminology was proposed by Manuel Atienza and Juan Ruiz Manero, llicitos atipicos: sobre el abuso de derecho, el fraude de ley y la desviacion de poder (Trotta 2006).
10   @   I. LIFANTE-VIDAL
parties affected by the application of the norm (for example a norm that regulates when a right expires) and for whom applying the principle will increase predictability, as the occurrence of certain consequences is not dependent on a hard-to-access detail: the third party's understanding of the applicable norms. Accordingly, and even though it may sound paradoxical, the principle of ignorantia juris non excusat can also be said to contribute to legal certainty.
2.3. Until when can a prediction be made? The stability requirement
The third dimension of predictability is stability in time. From this perspective the degree of predictability of a legal system depends on how far into the future the predictions can be made. The prediction is always made at a specific time and refers to future events. The reach of the prediction can vary in length; thus the longer the predicted term is, the more predictable a legal regulation will be. It is precisely this aspect or dimension of predictability that justifies certain legal requirements such as that of a specific type of non-retroac-tivity or that of regulatory stability.
However we can distinguish between two meanings of'stability',37 both connected with predictability, and relatively independent of each other. The first, more formal, meaning understands stability as the absence of changes: legal norms must have a minimum duration in time to allow subjects to plan their medium- and long-term behaviour. From this perspective, frequent legal changes lead to a lack of predictability. In the second, less formal, meaning, stability is understood as continuity (coherence), rather than a simple absence of changes. With regard to this second meaning, we have to assess the content of any changes made, to determine whether or not they imply instability and, therefore, whether or not they affect legal certainty. A certain regulation may be more or less stable under the first and/or second meanings. At times, certain normative changes (i.e., a certain lack of stability as under the first meaning) may not affect predictability, but actually increase it - imagine that a rule is worded ambiguously and that changing it would eliminate this ambiguity, or imagine a reform which consists in eliminating a rule that was incoherent with other rules or principles.
To comply with this temporal dimension of predictability, it is not enough to simply incorporate a diachronic perspective, under which the law would be conceived as a dynamic system, i.e., a series of elements changing over time; it is also essential to incorporate its central aspect as a social practice developed over time and aimed at achieving certain objectives considered valuable from the point of view of those participating in the practice. From this perspective, the practice must have a certain continuity, at least in its fundamental principles. It is precisely those values which the practice aims to develop, that allow us to identify it as a coherent, meaningful whole (in other words, one aimed at attaining certain ends). Citizens may be interested in making long-term predictions, even incorporating the probability that normative changes may be made in this timeframe. To take the example of the social security laws that regulate retirement pensions, what citizens want to know is not necessarily the exact amount of the pension they will receive in the future, but rather the general principles of the normative framework that will be applicable at the, as yet unknown, time of their retirement: will
See Arcos Ramfrez (n 16) 265.
JURISPRUDENCE   @ 11
pensions still be adequate? Will the later years of paying into the system still weigh more when determining the amount they will receive? Etc.
3. Summing up
As we have seen, a certain regulation may be very precise (have a high degree of predictability in the first dimension) but, nevertheless, have a low degree of accessibility to its subjects, or be too variable to allow long-term planning. On the contrary, an imprecise regulation may be highly accessible and very stable, etc. There are many possible combinations and, at times, what increases predictability in one dimension may also be beneficial - in terms of increasing predictability - in others. Nevertheless, on other occasions, the relation may be inverse - greater precision may lead to less accessibility or less stability. However, these dimensions are, in turn, complex, and what, for example, generates predictability for one kind of subject (e.g., businesses) may destroy it for other kinds of subject (workers, consumers), or a very precise wording (high content formality) may generate incoherence with other norms, making it necessary to interpret it in a less literal way (reducing interpretive formality). Which of these dimensions of predictability is more relevant is not a matter that can be established using merely formal parameters, which do not take substantive reasons into account. These evaluations cannot be made in the abstract, but rather special attention must be paid to the different assets and interests at stake in each case. From that point on, we are in a position to understand why it is not easy to determine in the abstract what is required by legal certainty and when we are in the presence of violations of this principle. The question is not so much that of determining whether or not predictability is affected, but rather that of determining which kind of predictability we consider of legal value in each case, that is which kind of predictability obliges us to maximise the value of legal certainty.
It should be borne in mind that the modulation to be made to the requirements of legal certainty will differ depending on the assets or interests at stake in each specific legal sector (criminal law, commercial law, inheritance law, labour law, etc.) and depending on the procedural instance in question (the same predictability is not required from the behaviour of citizens, from legislators, judges, officials, etc.). This will give rise to different balances between the different requirements deriving from certainty, aimed at maximising predictability in its different dimensions: precision, accessibility, stability. This is a balancing which we could consider internal to legal certainty itself. In some cases, it is more important to increase the precision of the predictability, while others demand a generalised predictability or predictability for certain groups (as in the case of consumer protection), and yet in others it is fundamental to maintain a certain stability. However, alongside this internal balancing, it is also necessary to weigh other legal values with which it may conflict; a balancing which we may consider external to the requirement of certainty.
In short, and in response to the question raised at the beginning of this section, I believe that we could say that the predictability which we consider valuable and which, accordingly, the principle of legal certainty obliges us to maximise, is that which affects reasonably grounded expectations (i.e., expectations which have to be considered legitimate in the light of the principles and values recognised by the law itself). If this is the case, in a constitutional state of law committed to a series of substantive values, legal certainty
12   @   I. LIFANTE-VIDAL
may not continue to be conceived as a value dependant solely on the formal aspects of the law, and, moreover, cannot be developed in equal measure independently of the justice or injustice of the law on which it is predicated. And, in this sense, it is difficult to separate ex ante which aspects of justice should be the law's concern and which should not.
ORCID
Isabel Lifante-Vidal © http://orcid.org/0000-0003-3348-2299
Acknowledgements
A previous version of this paper was presented and discussed at the II International Meeting on Law and Justice. What Aspects of Justice Should Not be the Law's Concern? (30 June, 1-2 July, 2016, Carlos III University, Madrid). I am very grateful to all the participants in this Seminar and most particularly to Raquel Barradas de Freitas and Timothy Endicott for their interesting and detailed observations. I am also very grateful to Victoria Roca, who carefully read different versions of this article.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Funding
This work was supported by the Spanish Ministerio de Economía y Competitividad under Research Project DER2017-86643-P ("Una teória postpositivista del Derecho").
Atienza refers to this interconnection between legal certainty and other components of justice, underlining its nature as a dependent value. Cf., Atienza, Introduction al Derecho (n 8) 118-19 and, more recently, Manuel Atienza, 'Seguridad jur-fdica y formacion judicial' in C Cruz Moratones, C Fernandez Blanco and J Ferrer Beltran (ed), Seguridad juridica y demo-cracia en Iberoamerica (Marcial Pons 2015).