Normative vs. Positive Approaches

There are two distinct, but not mutually exclusive, debates in the scholarly community regarding separate opinions in collegial courts. The normative debate focuses on the advantages and disadvantages of allowing, or even incentivizing, separate opinions in collegial courts. The positive debate revolves around a related but distinct question: given the possibility of separate opinions, what explains a judge’s decision to break with the majority opinion? Alongside normative theories, one can distinguish three sets of positive theories explaining the incidence of separating opinions: cost–benefit analysis (within rational-choice theories), political economy accounts (in the context of asymmetric information about judicial preferences), and explanations based on legal culture (reflecting historical path-dependence and institutional context).

Normative Theories

Normative arguments supporting the existence of separate opinions surround the benefits of free speech and expression, democratic values, and judicial independence (Alder, 2000; Mendenhall, 2020; Vitale, 2014) and the moral obligation borne by judges when an interpretation differs from the court majority (Bader Ginsburg, 2010; Brennan, 1985). These separate opinions may be justified as a corrective mechanism, thus (a) providing alternative sources of legal arguments, (b) favoring pluralistic interpretations of the law, and (c) improving case law as a consequence of better argued majority opinions (Haire et al., 2013; Mendenhall, 2020). They enhance evolutionary gains within a dialogue between past and future courts (McCormick, 2012; Mendenhall, 2020), hence assisting legal efficiency and balancing transitions across majority and minority views in time. Historically, judicial separate opinions can be construed as a source of unwritten law in the extent that they reflect deposits of legal knowledge (Mendenhall, 2020).

Normative arguments opposing separate opinions point to expected negative effects that dissents and concurrences may have on public confidence in the court and court legitimacy (Salamone, 2013; Stack, 1996; Zink et al., 2009), legal certainty, judicial independence (Laffranque, 2003), the erosion of political respect for the role of international courts and collegiality (Combs, 2021; Raffaeli, 2012), the efficient use of court resources (Raffaeli, 2012; Vitale, 2014), and compliance with court decisions.

The normative value of separate opinions cannot be addressed outside of a given institutional and legal context. In order to understand the normative value of dissent, for example, one must consider the arguments that sustain each jurisdiction’s legal design when accepting or rejecting dissent in the first place. To begin with, one significant factor is the prestige and long establishment of jurisdictions. In terms of legitimacy and public trust, a new-established jurisdiction may benefit from rejecting dissent and projecting an image of a united court. This was the case of the Italian Constitutional Court, which, to this date, still prohibits dissent (Zagrebelsky, 2000). On the contrary, in Portugal and Spain, after their transition to democracy in the 1970s, the countries followed a different path and (since their constitutional courts were established, in the early 1980s) separate opinions are allowed (Kelemen, 2018).

This invites a second important point: When considering appeals to supreme courts, speaking as one voice potentially enhances judges’ guiding role of the lower courts (Vitale, 2014). To counter this argument, though, many legal scholars believe that allowing dissent improves the quality and the depth of legal reasoning (Bader Ginsburg, 2010; Belleau & Johnson, 2008; Blomquist, 2004; Botelho, 2021; Brennan, 1985; D’Amico, 2019; Haire et al., 2013; L’Heureux-Dubé, 2000; Mastor, 2019; Scalia, 1995; Shepard, 2005). Separate opinions are a valuable pedagogical tool; they help clarify judgments, thus making rulings more understandable (Mastor, 2019; Raimondi, 2019). The increased quality of the majority decision experiences a “ripple effect” and is a performative virtue of separate opinions, as the majority judges will improve their argumentation with more convincing arguments than those of the dissenting judges (Mastor, 2019). Notwithstanding the openness to a plurality of interpretations of the law, some other legal scholars have argued, alternatively, that separate opinions fossilize the debate and complicate future changes in interpretation, unless the dissenting judges are replaced by a new generation of judges (Pérez de los Cobos, 2019).

To some authors, allowing dissent would project an image of a conflicted jurisdiction, which would hinder the court’s authority (res judicata) and generate legal uncertainty. Under this understanding, the secrecy of judicial deliberations is of paramount importance and allows the quietness and peacefulness that is required to conscientiously deliberate (Pérez de los Cobos, 2019). On the other side of the spectrum, it has been argued that revealing dissent increases a court’s legitimacy by transparently showcasing how the deliberation was the result of a serious weighting of different arguments (D’Amico, 2019; Saint-Hilaire & Sirota, 2020; Stephens, 1952). The weakness, however, is that identifying majorities and minorities—and which judge is part of which coalition—reinforces the image of politization of the courts, endangering its social legitimacy (Pérez de los Cobos, 2019).

From a normative perspective, unanimity and justice are not synonyms. Under this argument, legal authority is not determined by the (in)existence of dissenting opinions. Therefore, the normative discussion reflects the moral authority or the persuasive effect of court decisions (Raimondi, 2019). At a more fundamental level, concealing a disagreement within the court to convey to the public an appearance of monolithic truth is interpreted by some as a form of legal paternalism (Botelho, 2021). It follows that the authority of a deliberation does not flow from the number of votes in favor, but from the law itself. The fact that a specific deliberation was unanimous or approved by five or nine judges does not change its legal force (Botelho, 2021; Rousseau, 2000). And, perhaps most importantly, to circumvent the destabilizing effect of separate opinions, the majority ruling should be as well reasoned as possible (Botelho, 2021).

Conceptually, freedom of expression is also an argument in favor of separate opinions (Kelemen, 2013; Lynch, 2004). In this sense, allowing dissent is beneficial for mimicking the democratic arena and its plurality of opinions, either majority or minority. In other words, the sibylline and prospective capacity of dissent should be in harmony with the idea of the legal framework as an evolving and mutable reality (Botelho, 2021). Alongside it, one has the notion of “demosprudence,” that is, the jurisprudence improving democracy and its plurality of opinions (Flanders, 1999; Guinier & Torres, 2014; L’Heureux-Dubé, 2000; Raimondi, 2019). As today’s dissent might be tomorrow’s majority, separate opinions reveal “a kind of constituent metalanguage” that encourages the evolution of the law (Mastor, 2019, p. 127). Separate and majority opinions influence one another, even though such an impact is more diachronic than synchronic (Botelho, 2021). As checks and balances are the foundation of each states’ constitutional design, so dissent plays the same role balancing the majority’s power within the judicial system (Musmanno, 1958).

Moreover, the independence of the judiciary is often considered when allowing dissent. Judicial independence is both external (before the other powers) and internal (before other judges). For precisely that reason, some scholarship has argued that allowing dissent promotes judicial independence (L’Heureux-Dubé, 2000; Saint-Hilaire & Sirota, 2020).

On the other hand, a relevant argument for not allowing dissent is the culture of collegiality (Alder, 2000). Collegiality rests upon two premises: unanimity and anonymity (Vitale, 2014). To counter this, some stress the false dichotomy in equating unanimity with authority and clarity. Furthermore, when forced to reach a unanimous decision, unanimity is not real, but a “pseudo-unanimity” (L’Heureux-Dubé, 2000). Quite simply, there is no such thing as “unanimity by decree” (Botelho, 2021). If anything, the separate opinion can be deemed as a mere moment of “poetic protest” (Shepard, 2005).

Oftentimes, arguments against separate opinions condemn its trivialization and instrumentalization by judges wanting to self-promote themselves. As a former president of the Spanish Constitutional Court once wrote, the most difficult role of the president is to “manage [judicial] egos” (Pérez de los Cobos, 2019). One particular concern is the “professional dissenters,” who show a pathological and narcissistic attraction to dissent to distinguish themselves from their colleagues (Botelho, 2021; Jori, 2019; Pérez de los Cobos, 2019; Rugge & Volpe, 2016). A distinction clearly needs to be made between a great dissent and a great dissenter (Tushnet, 2006). Still, it is not surprising that some dissenters who acted as prophets with honor were crucial for the evolution of the law (Barth, 1975). As a result, the most widespread and recognized dissents were ones related to subjects such as social justice, the right to bear arms, and fundamental rights (including but not limited to racial justice and women’s rights). Even if some might think dissents sound hollow at that juncture, they are a consolation for the losing party and make the ruling more acceptable overall (Raimondi, 2019). For example, dissenting opinions could offer cues of procedural justice (Baird & Jacobi, 2009; Bentsen, 2019).

Positive Theories

By Nature of Cases

The interaction between the information conveyed by a separate opinion and the legitimacy of the court plays an important role (Amaral-Garcia & Garoupa, 2017). More controversial laws may be associated with more dissent due to the intensity of the political debate that they imply. Alternatively, more controversial laws might require a more decisive intervention by a collegial court, which hinders judicial legitimacy and, in turn, deters potential separate opinions. The interaction between the nature of the cases and separate opinions is, therefore, complex (Muro et al., 2020).

In their study of the Canadian Supreme Court, Songer et al. (2011) showed that dissent is strongly related to legal ambiguity in the law, but also to political conflict, institutional structure, and variations in the leadership style of the chief justice. Hanretty (2015), in his research about the Estonian Supreme Court, found that the main determinant of judicial disagreement was the appropriate scope of constitutional redress, not necessarily the political salience of the case.

Furthermore, Bentsen (2018) examined the Supreme Court of Norway and concluded that rising rates of dissent in the 1990s reflected both changes to judicial discretionary power and the salience of human right cases. In a second article by the same author, Bentsen (2019) further documented that these dissents found significant support among opponents of a court decision. Thus, according to him, separate opinions could be perceived as a mechanism of procedural justice in more controversial cases.

Bricker (2021) investigated the production of court decisions and separate opinions in six different courts in Europe. He concluded that consensus depends on the complexity of the case: more complex cases induce more separate opinions.

Finally, in other experiences, dissent tends to be more likely when cases are more controversial (especially regarding separation of powers) and when the largest opposition party is politically more relevant (Bagashka & Tiede, 2018).

The nature of cases subject to adjudication by a collegial court is inevitably determined by docket control. Skiple et al. (2021) compared tax decisions in the Norwegian and Danish supreme courts. The authors found that discretionary dockets are associated with higher dissent rates than mandatory dockets.

By Nature of Separate Opinions: Oral Versus Formal

An individual judge may disagree, partially or fully, with the majority opinion and yet decide not to file a formal separate opinion (Laffranque, 2003). Disagreement without written separate opinion is less costly (for example, no formal document must be written and no major elaboration on a legal explanation need be advanced), but it is also less likely to generate a strong benefit (different audiences are unlikely to pay attention to a mere registration of a negative vote with no additional written explanation). Hence, mere oral disagreement could be seen as a compromise between voting with the majority (thus never revealing a discord) and filing a formal separate opinion (expressing an individual and/or distinct view, partially or fully, in written terms).

The distinction between oral disagreement and filing a separate opinion is a matter of degree. Once there is public disagreement, the cost of disrupting a norm of consensus is borne. In a sequential decision, when considering whether to file a separate opinion, the disruption to a consensus norm is a sunk cost to the individual judge. The additional decision is mainly about revealing to the world the individual reasons for such collegial disagreement (and possibly the specific identity of who has disagreed).

Reflecting the existing theories of dissent, the balance of costs and benefits in favor of mere oral disagreement (with no separate opinion) could depend on the nature of the case or the characteristics of the individual judge. For example, judges could consider a certain case sufficiently important to justify exposing public disagreement but not important enough to go a step further and file a separate opinion. The importance of the case can be understood to be legal (varying as a function of the area of the law) or political (the judge’s appointer, party, or certain audiences, even public opinion matter for this purpose). Still, it could also be that certain individuals are inclined to explain their disagreement more often because of professional background. A former legal academic or politician could be more or less inclined to write fluently about the law (or public policy) than a career judge, a former lawyer, or a standard bureaucrat as a function of post-court career opportunities (Raimondi, 2019). Age and tenure in the court may be associated with a learning curve on appropriate behavior in collegial courts and could influence the ultimate decision of filing a separate opinion. In a more mundane view of collegial courts, a lazy judge could simply prefer to register disagreement without recording the exact reasons about the disagreement or even disclose his/her identity to the general public.

In their study of the Spanish Council of State, Garoupa et al. (2022) found that the individual decision to file separate opinions, when there is already disagreement in the collegial decisions, is more related to professional background and demographics in the collegial court broadly defined (more specifically, tenure in the institution), and less to political variables or case-specific variables (except for decisions involving Spanish autonomous communities).

By Nature of Separate Opinions: Reasoned Versus Boilerplate

According to rational choice theory, judges balance costs and benefits when contemplating a separate opinion. Judges should write a formal separate opinion when the possible benefits (such as the impact on the law or external recognition) outweigh costs. However, if a separate opinion is to be filed in a formal way, judges must decide which kind of opinion to cast—a long, detailed, reasoned dissent or a simple boilerplate dissent. To illustrate, some dissents are longer than the ruling itself, or than famous books, such as some of Shakespeare’s plays (Penrose, 2018). In some cases, separate opinions are written similarly to a paper submitted to a law review (Fried, 2000; Raimondi, 2019). In fact, it has been said that judges should keep in mind that “the influence of their opinions is often in inverse proportion to their length” (Kolb, 2013, p. 1017).

By backwards induction, the decision on whether or not to dissent should take into account the subsequent decision concerning the type of dissent. A rational judge should go for a reasoned dissent when the matter justifies a long, legal pondering of arguments. The same rational judge should opt for boilerplate dissents when the case does not answer a very important legal question. The immediate consequence of these observations is that dissenting in important matters is costlier (because it involves long and complex reasoned dissents) while dissenting in less important cases is less costly (since the judge will file something like a mere template).

At the same time, one can envisage that individual benefits from dissenting are also more acute in important cases (at least, in terms of external visibility) than in less important cases (which have little impact on the law or on legal and political debates).

Therefore, rational choice theory cannot unambiguously predict the exact outcome on the balance of costs and benefits. In fact, it could be that the net benefit is positive for important cases (because legal impact is more significant than drafting a reasoned dissent), for less important cases (because filing a boilerplate dissent is almost costless), or for both.

In their study of the Supreme Court of Argentina, Muro et al. (2020) found that more important cases have a lower likelihood of carrying a dissenting opinion. Yet, when the authors broke down dissents by type (that is, reasoned comprehensive dissents versus boilerplate dissents), they found that majority decisions carrying dissents tended to be longer, but only for reasoned dissents. Furthermore, they showed that reasoned dissents were more likely to occur in important cases.

By Nature of Separate Opinions: Individual Versus Collegial

Theories of dissent focus on explaining why and under which conditions individual judges decide to write a separate opinion when they disagree from the collegial decision. Yet separate opinions can be strictly individual or collegial too. In the latter case, individual judges who disagree with the majority write a different opinion as a group. Broadly speaking, there is a continuum of possibilities. One polar case is a court decision supported by a majority and a separate opinion endorsed by a minority in a collegial institution. The opposite polar case is a court decision supported by a plurality and a multitude of separate opinions by individual judges.

From a conceptual viewpoint, the individual decision to join a collegial separate opinion or file a solo separate opinion can be understood as mimicking the first-order decision of agreeing or disagreeing with the court decision. For example, the costs of writing an individual separate opinion are more significant than joining a collegial separate opinion. Similarly, the prestige and influence acquired by authoring a separate opinion are presumably higher than when merely joining a separate opinion written by another judge.

Norms of consensus shape separate opinions in two moments—when deciding to disagree and when deciding the number of separate opinions to be formally written. In their seminal article, Caldeira and Zorn (1998) documented that these consensual norms impact concurrences and dissents substantially at the same time. Thus, consensual norms are likely to be a substantial influence in the time-pattern and number of separate opinions.

The Fundamental Divergence Between Private and Social Motivations

In the section “Theories of Separate Opinions in Collegial Courts: An Overview,” there was an important distinction between normative and positive theories of dissent in collegial courts. From a rational choice theory perspective, one can expect the normative theories to suggest the optimal rate of separate opinions while the positive theories should explain the observed rate of separate opinions. Broadly speaking, they should differ unless judges understand the available normative theories and act by them.

It is conceivable that judges in collegial courts care about normative theories (as embodied by collective reputation mechanisms in the way suggested by Garoupa & Ginsburg, 2015). However, judges are also likely to care about their individual reputation. It seems only natural that there is a divergence between individual motivations to dissent or concur and collective or efficiency motivations to dissent or concur. Nevertheless, the direction of the externality effect, that is, this divergence between individual and collective goals could easily go both ways. One can imagine having too many separate opinions because individual decisions do not fully consider the normative losses generated by separate opinions. Still, one can also imagine not having enough separate opinions when individual decisions fail to internalize the normative gains generated by separate opinions.

It does not seem easy to answer the question on whether there are too many or too few separate opinions without some recognition where the optimal number lies. Similarly, the rules established by collegial courts to regulate the production of separate opinions should be designed to facilitate such optimal or desirable number.

The Venice Commission (2018), on its report on dissent behavior, advised that separate opinions should only be used at ultima ratio or, in other words, with parsimony and only when all other attempts to find unanimity have failed. The rationale is to avoid abusing separate opinions. In fact, some warn that the generalization of separate opinions will lower their utility. Hence, separate opinions must be written selectively (Pérez de los Cobos, 2019, p. 118).

The Venice Commission (2018, p. 13) further recommended first, that separate opinions “should focus on explaining that the matter could be dealt with differently, perhaps, in a better way, but not that the solution chosen by the majority was of poor quality”; and second, that a “judges’ code of conduct or ethics” should be adopted, to define red lines that should not be crossed. Against the first recommendation, some scholars have argued that it “would be regrettable if judges hesitated to discharge their adjudicative responsibility, which individual (as well as collective) independence protects, for fear of making their colleagues look bad” (Saint-Hilaire & Sirota, 2020). As far as the second recommendation is concerned, some arguments are against the development of formal codes of conduct. To put it bluntly, formal codes of conduct could be used to “harass judges, and so to undermine the publication of dissenting or concurring views is meant to protect” (Saint-Hilaire & Sirota, 2020). If anything, judicial ethics and good sense should suffice when deciding whether or not to write a separate opinion (Botelho, 2021; Saint-Hilaire & Sirota, 2020).

Is dissent a symptom of a healthy jurisdiction or, on the contrary, does it reveal its dysfunctional dissonance? The answer must consider the frequency of separate opinions (a balanced number of dissents as suggested by Botelho, 2021), but also the substance of these opinions. Some legal commentators also highlight style, suggesting that the dissent should be written cordially, refraining from ad hominem attacks or from feeding the narcissistic self-promotion of judges. In other words, according to these legal scholars, the “me against the world” motto or the “stellar judge” longing to attract public attention is not consistent with what a functional dissent should be (Botelho, 2021). To sum up, concerning style, each judge should master the refined skills of moderation, the talent of “disagree without being disagreeable” (Little, 1999, p. 686), and the ability to communicate the disagreement in an elegant, respectful, and empathetic way (Botelho, 2021; Hambro, 1956; Kulawik, 2011; Pérez de los Cobos, 2019; Wedeking & Zilis, 2017).

Yet, at the same time, additional studies have revealed that, when writing separate opinions, judges tend to use more rhetoric and persuasion techniques. It has been reported that such a writing style had a better impact on future majorities of the court (Belleau & Johnson, 2008; Hinkle & Nelson, 2017).