THE TWO DISCOURSES IN COLOMBIAN CONSTITUTIONAL JURISPRUDENCE: A NEW APPROACH TO MODELING...
Publication: The George Washington International Law Review
Date: Saturday, January 1 2005
I. INTRODUCTION
Understanding, explaining, and modeling judicial behavior is important for the legal academic and the political scientist alike. Because legal academics tend to be far more preoccupied with normative questions (Is this a good legal rule? How can it be improved?), the task of
More specifically, I hope to contribute to two separate bodies of literature. First, there is the literature that Western legal scholars created on discourse in Latin American legal systems. This literature is both sparse and rent with considerable ideological preconceptions about U.S. and European legal systems; these preconceptions limit, for the most part, the usefulness of the descriptions.2 Second, there is a large volume of literature created by U.S. political scientists, which is generally focused on U.S. courts and aims to understand judicial behavior. The two most prominent schools of thought in this sub-field, attitudinalism and strategic theory, share the same rational choice underpinning. The challengers to attitudinal and strategic scholars are a loosely defined group of academics identified as legalists; they believe that judicial behavior is best seen not as an attempt to maximize some political policy goal to which the judge is attached, but as a response, at some level, to the judge's notion of what the law is. Unfortunately, the legalist model is extraordinarily underdeveloped, suffering both from a lack of interest from political scientists and from an almost exclusive attention to the U.S. federal courts: Legalist models have thus not gotten very far beyond focusing on mechanical adherence to precedent.
I seek to bridge the gap between the two literatures noted above and to use comparative work to lend some clue as to what a richer conception of the legalist model might look like. I use discourse (what judges say when they decide cases) as a clue to judicial worldviews. Using Colombian constitutional courts in the 1980s and 1990s, I argue that there are presently at least two completely different judicial worldviews about constitutional law in Latin America: (1) a traditionalist Latin American view that minimizes the role of constitutions by focusing on concrete rules, which form a relatively small part of constitutional discourse; and (2) a newer view that focuses on the principles and values behind constitutions, and thus tends to read them broadly. These worldviews are extremely rich, integrating ideas about what law is, interpretative methods, theories of the judicial role in a political system, and substantive values into a fairly coherent whole. Further, each of these worldviews can be identified with a particular type of social actor as its likely carrier: As I will suggest, the traditional worldview is most closely associated with career judges, whereas the alternative view is most likely to be espoused by public or constitutional law scholars. This matching of actors and worldviews, although somewhat crude, helps us to discipline, clarify, and structure the legalist model considerably.
The rest of this work is organized as follows: Part I surveys the descriptive (mostly political-science driven) literature on judicial behavior and shows the common rational choice assumptions of attitudinalism and strategic theory, as well as the theoretical shortcomings of legalism. Part II draws mostly off Weberian social thought to construct a new model of judicial action that focuses on two distinct, rich judicial worldviews that different sets of social actors predictably carry. Part III uses this model to explore the language of judicial opinions in Colombian constitutional courts in the 1980s and 1990s, thus acting both as a preliminary test for the plausibility of the hypotheses and as a chance to flesh out and explore the remarkable richness and internal coherence of the two worldviews that I have identified. Finally, Part IV attempts to link these worldviews-which have been constructed by using the language of judicial opinions-to "real" social phenomena: I suggest that these worldviews affect judicial case outcomes, and at any rate the language of opinions has an independent impact on public opinion and thus on a court's legitimacy.
II. THE DESCRIPTIVE LITERATURE ON JUDICIAL BEHAVIOR: THE FAILURE TO FORMALIZE THE LEGAL MODEL
Scholars have spent almost no effort building and testing models of judicial behavior within a comparative setting.3 Almost all work has thus been done within a U.S. context, virtually always using federal courts, and generally the U.S. Supreme Court. These commentators have grouped theories of judicial behavior into three main types: attitudinal, strategic, and legalist.4 Attitudinalists believe that judges follow their raw, political preferences when making judicial decisions.5 Believers in strategic theory think that judges are led rationally to defect from their raw preferences by the presence of other relevant actors within the systems (like a legislature or higher court judges).6 Finally, adherents to legal theory think that judicial decision-making is best explained as a response to specifically legal variables, like adherence to precedent.7 Although legalism is by far the broadest and richest of these three theories, it should be briefly emphasized that all three combined form only a narrow slice of potential ways in which one could explain judicial behavior.8 This narrowness seems to be largely a function of the social-scientists' emphasis on quantifiability.
Rather than attempting to give an exhaustive summary of the literature in all three camps-a task that others have adequately carried out9- my goal here is to demonstrate two narrower points: (1) in section I (A), that the attitudinalist and the strategic models are really both aspects of one, single theoretical approach-the rational choice approach-which differ not in their theoretical assumptions about human nature but in their factual assumptions about the judicial policy-making environment; and (2) in section I (B), I will attempt to show just how poorly theorized and formalized, if implicitly influential, the legalist model is.
A. The Common Core Assumptions of the Attitudinal and Strategic Models
1. The Attitudinal Model
The attitudinal model seems to be the dominant formalized model of judicial behavior today.10 Put briefly, the model holds that judges come to the bench with certain "ideological attitudes and values" that are essentially political in nature, and they systematically vote in favor of these preexisting political preferences when making judicial decisions.11 "Simply put, Rehnquist votes the way he does because he is extremely conservative; Marshall voted the way he did because he was extremely liberal."12 The "laws" that judges use in their opinions are simple post hoc justifications and not true causes of judicial behavior.13 Of course, many theories of political behavior argue that politicians vote in accord with their revealed ideological preferences;14 the attitudinal theory is thus a transference of a theory from the political to the legal realm. Just as politicians vote in accord with ideological, policy-based considerations, so too do judges.15 The results of empirical testing of the attitudinal model have been somewhat mixed, but most studies have found considerable support for the model.16
2. The Strategic Model
Strategic theorists believe that judges have preferences of one sort or another, but that when they actually give their decisions, the preferences embodied therein are unlikely to match the judges' actual preferences because of the influence of various constraints.17 By constraints, these theorists do not mean senses of rightness or duty that lie within the judges' heads; rather, they mean something external-either other people or institutions. For example, the presence of a legislature that can overturn judicial decisions may influence judicial behavior. To the extent that the judge can foresee this possibility, she may try to hand down a decision that is as close to her personal policy preference as possible without galvanizing a congressional override.18 For lower court judges, one constraint might be the possibility of a higher court judge's reversal of their decisions.19 Finally, internal rules or the presence of other actors within the court itself might influence judges; thus, U.S. Supreme Court justices might refuse to grant certiorari even for a case they thought was wrongly decided because they anticipated that if certiorari was granted, they would lose on the issue once and for all.20 Again, the model here is obviously derived from rational choice models of political behavior-which in turn are derived from economic models of behavior.21 As an example, Ferejohn and Shipan have produced a model of bureaucratic deference in the face of possible congressional action22 that is theoretically identical to Ferejohn and Weingast's theory of judicial deference in the face of possible congressional action.23 Empirical support for the strategic model has been somewhat weaker than that for the attitudinal model.24
3. The Attitudinal and Strategic Models Share the Same Basic Assumptions
Theoretically, attitudinalists could argue that judges rule in accordance with their own ideological preferences honestly, rather than strategically, because for some reason judges simply are not capable of, or prefer not to, act strategically. In practice, however, this is not what they say. Attitudinalists instead say that the factual environment renders strategic action unnecessary, at least for U.S. Supreme Court justices, because, for example, federal judges have life tenure, U.S. Supreme Court justices have no real ambition for higher office, and congressional overrides are rarely a realistic danger.25 "The Supreme Court's rules and structures, along with those of the American political system in general, give life-tenured justices enormous latitude to reach decisions based on their personal policy preferences."26 In other words, both strategic and attitudinal models, in practice, assume that judges are willing and able to act strategically. Where the two theories differ is in their factual assumptions: Strategic models support the belief that judges face various types of constraints that force them to support decisions that differ from their preferred policy points, while attitudinalists believe that the institutional environment leaves at least those judges that they study-generally U.S. Supreme Court justicesfree to make decisions that are exactly in accord with their preferred policies.
Similarly, followers of strategic theory could theoretically believe that judges act strategically to maximize achievement of some set of goals other than their ideological policy preferences. For example, perhaps judges could prefer "legalistic" goals like adherence to precedent, but would have to defect strategically from absolute adherence to those goals given the presence of other institutions with some clout, like the U.S. Congress. In practice, however, this is not what happens. Instead, strategic theorists virtually always model judges as strategically furthering sets of ideological policy goals, which are the exact same goals modeled by the attitudinal theorists.27
What we have, then, are two theories that in practice tend to collapse into one. In both theories, actors are assumed: (1) to have preferences; and (2) to act strategically for the maximization of those preferences.'28 In addition, attitudinalists and strategic theorists both believe in a particular kind of rational choice theory: Specifically, the actors' preferences are assumed to be solely ideological, policy-based goals derived from the political realm. It is important to emphasize that both theories also believe that the proper way to test judicial behavior is to look at what judges actually do, not at what they say: Thus, what matters is the outcome, not the reasoning of the case.29
The most important difference between the two models is how outside pressure is placed on judges: In the attitudinal model, judicial decisions can only be molded by external actors through the appointment process-the initial selection of a liberal or a conservative judge. This model is thus, from one perspective, a cousin of the economists' adverse selection game, where the only relevant fact is the a priori type of the appointed actor (competent vs. incompetent, liberal vs. conservative).30 The strategic model, without denying the impact of the appointment process, suggests that external actors can also influence judicial behavior after this point through the use of incentives such as denials of promotions, reputational losses, impeachment, and reversals. Thus this model resembles the economists' game known as moral hazard, where the emphasis is not on the actors' types but rather on the sorts of incentives they face.31
B. The Unfulfilled Promue of the Legal Model
The very narrowness of the strategic and attitudinal models should give the legal model plenty of breadth in which to work. But this very breadth has been, in many ways, the problem with legalistic theories of judicial behavior. There is nothing like a common answer among believers that "law" drives judicial decisions on two key issues: (1) what are the factors that a legalistic judge pays attention to?; and (2) why do legalistic judges care about law? I will take up these two questions in turn, below. Aside from breadth, another problem with the legalistic model has been that most of its proponents are not political scientists seeking to create relatively formalized models about how judging works; instead, the model's most strident proponent has been mainstream legal academia.32 These legal academics have made assertions about the nature of law in a non-formalized manner and in a way that shows little concern for the intermingling of positive and normative argument. This obvious intermingling with normative argument has made legalistic theory an easy target for the seemingly less biased, more purely descriptive attitudinalist and strategic models,33 while the failure to make any attempts at formalizing legalist theory has left it in a muddled state, easy prey as a foil for the other two theories, and unsupported by much of the formalized empirical evidence that is generally considered acceptable in political science.34
1. Legalism means adherence to what?
What does it mean for a judge to be legalistic? Much of the defining, unfortunately, has occurred at the hands of the theory's opponents. Segal and Spaeth, the most prominent attitudinal scholars, define the legal model as follows: "the belief that, in one form or another, the decisions of the Court are substantially influenced by the facts of the case in light of the plain meaning of statutes and the Constitution, the intent of the framers, and/or precedent."35 Of these elements, it is obvious that the key element to Segal and Spaeth is precedent. Specifically, Segal and Spaeth have attempted to test the legal model by observing whether U.S. Supreme Court justices who dissented from one decision changed their minds and adhered to the now precedentially-supported view in later decisions on the same issue.36
One can make several obvious criticisms of a legalistic model focusing almost exclusively on precedent. First, it seems thin: It would be fairly strange to think that a legalistic judge would simply determine what relevant precedent was on point and then would apply that precedent.37 A better model of legalistic behavior would have to be richer, and comparative work would help to draw out that richness. A second problem with Segal and Spaeth's model is that it envisions legalism as operating as an external constraint on preference-directed behavior, rather than as a state of mind internalized by the judge in question.38 The second conceptualization seems more realistic. Finally, the Segal and Spaeth model strikes one as a bit of a straw man because it reflects a legal model that is far too mechanistic in nature: As has been noted by critics, formalism has not been in vogue for nearly a century and today's mainstream idealist jurisprudence is far from mechanistic or formalistic in nature.39 The post-realist critics of Segal and Spaeth, however, have failed to identify many clear, testable competing conceptions: Moving from the lofty domain of idealist jurisprudence to the more grounded realm of the rigorous, empirically-testable model has proven exceedingly difficult. Without such a model, post-realists leave themselves unprotected against Segal and Spaeth's critique that the legal model seems consistent with any judicial behavior, and that such a model is not falsifiable.40
Given the evident problems with Segal and Spaeth's precedentcentered model, it is not too surprising that some other scholars have tried to produce models that they perceived as more realistic.41 The most interesting and important of these combined theoretical and empirical efforts is by Richards and Kritzer, who try to identify and test the influence of what they call "jurisprudential regimes."42 By jurisprudential regime, Richards and Kritzer refer to "a key precedent, or a set of related precedents, that structures the way in which the Supreme Court justices evaluate key elements of cases in arriving at decisions in a particular legal area . . . jurisprudential regimes function as intervening variables between factors influencing justices' decisions and the decisions themselves."43 Thus, although Richards and Kritzer, like Segal and Spaeth, do look at precedent, they consider it in very different ways. In the Richards and Kritzer model, precedent is not some constraint to mechanically apply to similar future cases; rather, sets of precedents and the ideas embodied in them form part of the internal consciousness of the justices, filtering between the raw data of the cases and the decision that is eventually reached.44 Thus, in their model, the division of a free-speech incident into content neutral vs. content-based tracks served as a jurisprudential regime,45 and they found that the presence of this regime had an empirically verifiable effect on the Court's decision-making: cases decided after the formulation of this regime were decided differently to a statistically significant degree than those decided before it was determined.46 The Richards and Kritzer approach to understanding legalism is promising because it gets at the idea that the judge sees legalism as a sort of richly-defined worldview within which she acts. Richards and Kritzers's "jurisprudential regimes" are fairly narrow; they self-consciously avoid constructing a macroworldview of judicial behavior in favor of smaller-scale, issue-specific notions.47 Without faulting this approach, this Article's aim is to articulate just this sort of macro-worldview-or, rather, two competing macro-worldviews-in Latin America. One advantage of comparative work is that it tends to allow the most basic, constitutive notions of an institutional order to become clear.
2. Why Does the Legalist Judge Act in a Legalist Way? Preference vs. Role
Along with the issue of what factors the legalist judge pays attention to, the even more basic, if much less studied, question of how the legalist judge's human nature should be conceptualized stands as equally muddled. Treating legalism as a preference of the individual judge would achieve the easiest reconciliation with attitudinal and strategic theory.48 Thus, instead of desiring decisions in accord with their personal ideological views, judges would desire decisions that they saw as being legally sound. Such an approach to legalism might occasionally be useful to allow ease of testing an integrated attitudinal-strategic-legalist approach, but it has the drawback of seeming rather unrealistic. It does not seem quite right to say that judges prefer legalist over nonlegalist outcomes in the same way they prefer apples over oranges. Instead, a more complex phenomenon is occurring. The judge is trained and socialized to act a certain way, so she may not perceive herself as having much choice and may feel duty-bound to determine cases in accordance with legal principles.
Some legalists have tried to adopt more realistic, if less tractable, notions of human behavior, based on some idea of "role" or dutythe judge's conception of what it means to be a judge or what she ought to be doing as a judge.49 One theory, for example, suggests that the institution of a court itself has a sort of normative structure, and the actors within the institution somehow adopt this structure as their own normative framework.50 Thus, U.S. Supreme Court justices gain their role conceptions from the norms swirling around the Court.51 This institutionalist theory doubtlessly has some force, and it may explain a bit of the narrative I will tell about Colombian constitutional jurisprudence. One could also argue, however, as does this Article, that the source of much of one's "role" conception lies not in the court on which one is currently sitting, but somewhat further back in time. Thus, one might consider the source of one's professional training, or the course of one's career prior to appointment on a top-ranking court.52
C. The (Sparse) Descriptive Literature on Judicial Behavior in Latin America
The existing descriptive literature on Latin American judicial behavior falls into either the attitudinal or the strategic camps, with legalist work being essentially absent.53 For example, Larkins has argued that Argentina's recent president Carlos Menem appointed members to the country's Supreme Court that were dedicated Peronists and close associates of the president and thus extremely compliant towards his program.54 The Argentine Supreme Court, therefore, served Menem's shift towards an executive-led, delegative democracy.55 Obviously, this is essentially an attitudinal argument. Recent work by Helmke fits more comfortably into the strategic paradigm. She argues (and empirically demonstrates) that Argentine justices, although generally possessing ideological preferences that are very close to those of the nation's incumbent presidents (who generally appointed them), sometimes strategically defect from the president who appointed them.56 The reason why they defect, Helmke posits, is basically because they want to keep their jobs, and various means, both formal and (mostly) informal, exist for new presidential administrations to purge the old Supreme Court and put in place their own appointees. Thus, for the sake of extending their judicial tenure, judges have strong incentives to defect from the incumbent president's policy at the end of his term in an attempt to please the future president. An external actor's-in this case, a future president's-exertion of pressure, therefore, alters the preferred policy position of judges.
Both the attitudinal and strategic models, particularly the former, have received fairly good empirical support within the United States; these models would probably perform at least as well in Latin America. In particular, given the lack of effective institutional protections for judicial tenure in Latin America, strategic theory would seem to be a fruitful avenue for future research. This Article is therefore not meant to imply that the types of political pressures identified in the attitudinal and strategic models never occur in Latin America. Instead, this Article will demonstrate that the lack of legalist work in Latin America is a major lacuna, and it would be a significant mistake for scholars seeking to understand Latin American judicial behavior to fail to take account of judicial role conceptions.
D. A Short Summary of the Literature Review
We have seen, first, that the strategic model and the attitudinal model are really the same theory, with both importing rational choice assumptions derived from political science and ultimately from economics. The only real difference between the models is in the point at which control can be exerted on the judicial process by outside actors: In the attitudinal model, like in the adverse selection game, outside actors can only exert control during the appointment process through selection of the desired "type"; while in the strategic model, as with moral hazard, outside actors can exert control after appointment through manipulation of incentives. These two models dominate formalized descriptions of judicial behavior, and are the only game in town in Latin American research. Legalism stands against these two models as a rather jumbled mess. Yet the most promising way of conceiving the legalist model would seem to focus on judicial role conceptions-rather than preferences-and would attempt to construct a relatively rich judicial worldview.
III. A NEW THEORY OF JUDICIAL BEHAVIOR IN LATIN AMERICA
At the outset, this Article's goal for the present, unlike those of the attitudinal and strategic scholars, is not to explain case outcomes. Instead, this Article seeks to explain the judicial self-conception as expressed in the language of judicial opinions.57 This parallels a move made by some other legalist scholars.58 Part IV will attempt to provide some link between the judicial self-perceptions that this Article has constructed and case outcomes, and will also attempt to show that the language used in opinions is independently worth studying for its impact on public opinion.
A. Weber and Legal vs. Economic/Political Rationality
The theory that this Article will construct here has its roots loosely planted in Weberian social thought. Weber gives us, first, a suggestion that the strategic and attitudinal models would seem insufficient to explain legal behavior.59 The rational choice model embodied in these two models is a worldview. It is a worldview that matches pretty well with our society's understanding of modern economic rationality. The rational economic actor conceives of himself as being called upon to act strategically-or purpose rationally, to use Weber's term-for the end of maximizing profit.60 Modern economic rationality, in Weber's view, does not seem too different from modern political rationality. The heart of the professional politician's role is also to act in a very strategic manner, for the purpose of maximizing power.61 Behind this game for power, hopefully, lies some cause, some ideological policy, that the politician wants to favor.62 Because Weberian economic rationality is roughly a variant of rational choice theory, and because Weberian political rationality is so similar to Weberian economic rationality, rational choice theories do a good job of predicting many aspects of political behavior. In both the political and economic spheres, however, rational behavior is a construction, not something deeply imbedded in human nature: Entrepreneurs and politicians act the way they do because of their understanding of what it means to have the role they hold within society.63
Things change significantly when one comes to judges, because rationality for a judge is constructed quite differently than either political or economic rationality. Weber seems to consider modern judicial rationality to be somewhat similar to modern bureaucratic rationality.64 The key difference between political rationality and judicial-bureaucratic rationality is that an actor subject to the latter feels himself ultimately not to be free to choose his own end or cause; instead, a web of norms chooses his end or cause and constrains his behavior.65 The attitudinal claim that judges bring their personal policy preferences to bear on cases clashes sharply with the self-perception of the judicial actor. Furthermore, because bureaucratic rationality par excellence is defined as the obeying of commands, there would seem to be considerably less scope for strategic conduct.66 Weber is not nave; he recognizes that bureaucrats and judges can and often do act strategically in their personal selfinterest.67 This is not their core role conception, however; and the content of this core role conception, one would presume, would significantly affect their behavior.
B. A New Model - The Two Worldviews in Latin American Constitutional Law
If, according to Weber, we should hesitate to apply theories developed to explain political behavior to the legal realm, then what sort of theory might we use as a replacement? Again Weber gets us started. This Article will not use Weber's specific categorization of legal systems,68 but the broader point Weber-along with many others-makes is very important: He sees the way that law, in the modern world, is a sort of autonomous system that hangs together and makes sense in the eyes of its actors (lawyers and judges).69 Legal actors possess a distinctively legal orientation to the world, and this orientation seems unified and systematized to them. To act rationally from a legal perspective is to act meaningfully within that worldview. Thus, the first step-although surely not the last-to understanding the behavior of judges is to reconstruct their worldviews.
Of what is such a legal worldview composed? Weber suggests, and one might safely assume, that it consists primarily of two components: (1) a notion of what law is (a sort of jurisprudence); and (2) a notion of what the judge is supposed to do with that law (a sort of political theory of the judge's role). Note that this definition of a worldview is somewhat different than Richards and Kritzer's "jurisprudential regime."70 They define jurisprudential regimes as encompassing only understandings of what to do within a single area of law. The worldviews expressed here are intended to be far more comprehensive, encompassing the constitutional judge's entire professional self-conception.71
There are at least two quite different worldviews that Latin American constitutional judges possess at the present time. The first, which I will call-using a closely related concept from LopezMedina72- traditionalism-positivism represents the classic way of seeing law in Latin America. When translated into the constitutional realm, it means that courts will interpret constitutions just like ordinary statutes; thus constitutions will not be seen as having much impact on the legal order. The second worldview, which I will call new constitutionalism, defines itself quite differently. It sees constitutional interpretation as being a different enterprise from ordinary statutory interpretation, and thus views constitutions as documents that should be read broadly and with the document's hierarchy of ideals in mind. This Article will flesh out the conceptions much more broadly in the case study below.
For now, though, it is more important to theorize why a given constitutional judge might adhere to one of the above worldviews versus the other. Most of the answer appears to lie in intellectual cleavages within the legal profession. Traditionalism-positivism is still the dominant outlook within Latin American societies; carriers of new constitutionalism perceive it as being something quite new and different within Latin America, part of a transnational network of high-level academic/judicial discourse moving towards expansive constitutionalism. The focus here will be on the judges' legal experiences before becoming constitutional judges. Pre-constitutional-court professional experience is likely to make a big difference: Career judges spend their lives immersed in the ordinary legal system,where traditionalism-positivism is still unrivalled; furthermore, these judges, due to their long exposure with the ordinary system, would seem unlikely to be sympathetic to claims that constitutional interpretation has special rules. At the opposite end of the spectrum, constitutional or public law professors would expect to have the most exposure to new constitutionalist ideas during their careers, and would likely see themselves as harbingers or entry points for a discourse that they see as being new, transnational, and very scholarly.73
With these ideas in mind, and at significant risk of seriously understating the amount of complexity involved, we can proceed to make a very crude table matching worldviews with their likely carriers:74
IMAGE TABLE 1TABLE 1: THE TWO WORLDVIEWS AND THEIR CARRIERS
These classifications, as crude as they are, can help us predict which worldview a court's decisions will exhibit. A career judgedominated court should demonstrate a faithfulness to traditionalism-positivism, while a constitutional law scholar-dominated court should adhere more closely to new constitutionalism. This Article will examine this hypothesis below, using Colombia as an example. In fact, the model constructed here, like attitudinal theory, can be usefully conceptualized through using the adverse selection game. There are two different types of actors-constitutional scholars and career judges-each carrying a very different attitude towards constitutional law. The existence of these two types opens up possibilities that could be exploited by the relevant political authorities during the appointment process (ie. politicians could select carriers of worldviews that gain these politicians political advantages).
Finally, a caveat: This categorization of courts into carriers and worldviews is a contingent one, holding true only in Latin America and only at the present time. It is not meant, of course, to be a universal typology of judicial behavior.
IV. THE Two WORLDVIEWS AND THEIR CARRIERS: THE COLOMBIAN CASE
Having formulated a model, it is time for at least a preliminary investigation of its plausibility. This Article will examine the model using Colombian high courts hearing constitutional claims-the generalized Supreme Judicial Court in the 1980s, and the specialized Constitutional Court from 1991 to the present. The data on the composition and prior experience of the judiciary comes entirely from newspaper articles, and is collected in detailed form in the tables in the Appendix, while the source for judicial discourse comes mostly from formal written judicial decisions, although for the post-1991 period, this Article relies on some of the constitutional judges' writings in academic journals.75 The methodology is admittedly somewhat eclectic: For the pre-1991 period, I read-or at least skimmed-hundreds of cases, including virtually all of the constitutional jurisprudence from 1985 to 1990. In the 1990s, where far more constitutional opinions have been written and where the length of the average opinion has increased considerably, I had no choice but to be more selective: I focused on reading cases that dealt either with jurisprudential problems or with major social issues.
What I aim to show is that career judges dominated the 1980s Supreme Judicial Court and therefore demonstrated the Traditionalist-Positivist worldview, while constitutional law scholars dominated the post-1991 Constitutional Court and this court has thus gravitated sharply toward new constitutionalism. In the process, this Article will flesh out these two worldviews and give the reader some sense of just how rich they are: jurisprudence, political theory, interpretative craft, and substantive values are all combined in an essentially seamless whole.
A. Background
Prior to 1991, the Supreme Court of Justice (CSJ), a body of twenty-four members, held the power to hear constitutional claims.76 The CSJ functioned as a court of cassation as well as a constitutional court, and thus heard a wide variety of ordinary statutory claims.77 The body was arranged into four smaller chambers: the criminal, civil, labor law, and constitutional chambers.78 The full body of twenty-four heard all constitutional claims following a study by the constitutional chamber.79 In 1991, Colombia drafted a new constitution, which, among other things, changed the machinery for hearing constitutional claims. A new, specialized Constitutional Court, hearing constitutional issues exclusively and composed of nine members (initially seven, changed to nine after 1993), became the highest constitutional authority.80 This new institutional structure has persisted up to the present day.
B. The CSJ in the 1980s: Career Judges and Traditionalism-Positivism
1. Composition of the Court
It is difficult to find evidence of the composition of the CSJ in the 1980s, but the evidence that I have found strongly indicates that this was a career judge-dominated court. On November 6, 1985, terrorists affiliated with guerillas fighting the Colombian government invaded the CSJ offices.81 They killed about one-third of the court.82 Thus, in this abnormal situation, the government replaced much of the court at the same time, making it relatively easy to track the appointments in the press.
Of the twelve initial nominees to fill these vacancies, six appear to have been career judges.83 The newspapers described only four, including one of the career judges, as having substantial legal academic experience.84 These raw numbers would seem to substantially overstate the representation of legal academics and understate the representation of career judges: As the attack wiped out the court's entire four-person constitutional chamber, the members of this particular chamber seemed to have been very heavily oriented towards legal academia. Three of the four initial replacements in this chamber were academics, and the other was a member of the top national administrative court. When all four of these initial nominees declined because of fear or other reasons, the four new nominees were all legal academics.85
The academia-heavy, career judge-light composition of the Constitutional Chamber makes sense if one keeps in mind that pre-1991 Colombia used a centralized system of judicial review; almost all constitutional claims under the old constitution could only be heard by the CSJ.86 If other courts could not hear constitutional claims, then it would be tough to recruit judges from other courts to serve in the constitutional chamber, as that chamber's chief function was to study and recommend results for constitutional cases. Members of the other chambers dealt mostly with ordinary, non-constitutional cases in their respective areas, but recall that the full body composed of all twenty-four members actually decided constitutional claims.
If one roughly extrapolates the data for the eight replacements for the other three chambers (civil, labor, and criminal) and treats the constitutional chamber as a wholly separate entity with no predictive power for the composition of the rest of the court, then one might expect a full CSJ of twenty-four members to be composed of maybe thirteen or fourteen career judges and only five or six magistrates with substantial academic experience.87
Note further that this was a court where justices served for life and were replaced by vote of the remaining members of the court itself,88 which would seem to be peak conditions for the maintenance of a court dominated by a technocratic, civil-service-like career judiciary.
2. The Worldview of the Court
i. Interpretive Method and Judicial Role
Judicial worldviews are easily analogized to spheres: they have no obvious starting point for purposes of analysis. Nonetheless, in this case one can most easily begin to understand the judicial worldview by considering the general issue of constitutional interpretation. The CSJ addressed this issue in sentence number 51 of 1988, where it confronted the issue of whether a statute could properly penalize the mere act of running away after being ordered to jail, given the general interest in liberty enshrined in the preamble of the national constitution.89 The precise issue at hand-which would not give any court much difficulty-and the broader decision not to make the preamble directly enforceable is not incredibly important. What is important is some of the court's more general language about constitutions.
The court began by noting that norms can be either should-be's (current arrangements) or want-to-be's (aspirations), and preambles fall primarily in the aspirational category. Legal rules ("normas juridicas"), on the other hand, fall into the should-be category. The court stated that it knows this because of the simple grammatical logic "brilliantly expressed" by Hans Kelsen in his Pure Theory of Lam. Enforceable legal rules, unlike constitutional preambles, "impute a determined consequence to a certain antecedent."90 There is a division, then, between legal rules-things that state in their own text when they are triggered and describe the sanction for triggering them-and other norms like general principles and values. The status of these necessarily vaguer norms within the system is reduced to mere aspiration. Furthermore, the court noted that constitutionalism inevitably involved the comparison of lesser legal rules with constitutional legal rules, which are of a higher order. It acknowledged, however, that only
poorly can we try to determine the constitutionality of a legal precept compared with a principle or value of diverse kinds .... [F] or determining if something conforms with another thing or not, we should compare homogenous entities, not heterogeneous ones, as would be, in the present case, if we compared a rule with a principle or value.91
Thus, in addition to being aspirational, constitutional values and principles seem ill-suited on practical grounds to application in real constitutional cases-they are too different in nature from ordinary statutory law. When put together, these two critiques render problematic the idea of a broad constitution with considerable penetration into the rest of the legal order; most of the language in all constitutions would not meet the Kelsenian definition of a legal rule, but instead looks like a vaguer principle or value.92
Finally, the court sought to link its jurisprudential points to some conception of the proper judicial role within a separation of powers system. In a traditional civil law system, the notion is especially strong that judges must only apply law; they must not make it.93 The court applied this traditional theory as follows:
The confrontation of many rules articulated in the Constitution and the law with [principles and values], would give rise to grave contradictions, according to the focus of the political doctrine with which you looked at them. If this occurred . . . the constitutional judge would become a legislator, and, what's more, a constituent, starting down the road towards legal uncertainty and . . . arbitrariness.94
In summary, the majority envisions a jurisprudence where only legal rules have legal efficacy, while principles and norms are purely aspirational in character, and where principles and norms are too different from legal rules to really fit into the legal structure. Further, the majority backs this up with a narrow theory of judicial role.95
The seven dissenters imagined a very different, and far more expansive, task for constitutional values and principles. There was little historical or other evidence that values and principles (including those in the preamble) were intended to be merely aspirational.96 Furthermore, the Kelsenian definition of the efficacious legal rule is far too simple: Oftentimes legal rules did not come with any kind of consequence or sanction attached in and of themselves; the sanction was found elsewhere in the legal order. Thus, the dissenters saw the legal order "systematically and comprehensively," as a united whole and with its guiding principles in mind.97 The dissenters also made the important point that constitutions are inevitably political and must be applied with their given broad political philosophies in mind if they are to be enforced sufficiently.98 This, however, emphatically did not mean that courts could not apply them objectively.99 What the dissenters wanted, in sum, was a constitutional order that went beyond mere logical interpretation and entered the realm of teleology.100 Principles and values existed to guide the application of constitutional rules and to broaden their scope beyond what would be allowed by classical positivism.101
It is interesting that the court saw statutory interpretation in the same way as constitutional interpretation: It is a logical enterprise, not a ideological one.102 One key point, then, is the way that these pre-199 !judges minimized the differences between statutory interpretation and constitutional interpretation: They basically saw the two as the same enterprise. Again, the vision is that of law, whether statute or constitution being about low-to-the-ground rules. The ordering is vertical in the sense that a higher ranking rule in the constitution outweighs a lower ranking one in a statute, but horizontal or flat within each rank-a constitution, like a statute, is composed of a series of rules that are autonomous and not controlled by higher-order norms. As I have already said, the inevitable result is to downplay much of the significance of the constitution, because the constitution is a very different type of document from a statute-it is far vaguer in form. Thus many types of claims simply were not computable under the constitution.103 What survived within the constitutional order? Primarily, it would seem, those provisions that were framed as rather specific, particular rules, like the prohibition against ex post facto laws and the right to a defense.104
The basic story so far-of flat constitutions-needs to be qualified in just one respect. If most of the provisions in the constitution were enforced individually, without regard to some unifying, guiding principle that might stand behind them, the structural provisions were treated differently. Here, separation of powers acted as a spirit: It was, in the court's words, a "principle" that formed "one of the pillars of Constitutionalism as a national philosophy of division and management of powers in service of liberty."105 The court thus derived several specific doctrines from this principle, including, for example, the idea that "all actions of a branch of government in activity that properly belongs to another branch require an express mandate in the constitution, since the contrary would . . . contravene the philosophy of the Constitution."106 Thus, if Congress wanted to adjudicate or perform administrative functions, it would need to find explicit authorization for the particular act of adjudication or administration from the constitution.107 Related to this principle, the court developed a family of important doctrines limiting the president's extraordinary powers under states of emergency and states of siege.108 This last set of cases in particular had a significant impact on many of the executive's important policies,109 leaving the general impression that most of the court's constitutional cases in the 1980s, and particularly most of its important cases, were argued and decided on these sorts of structural grounds.
Why was the court so willing to imbue the constitution with a spirit when it came to structure, but nowhere else? Some of this might be rooted in the division, particularly strongly seen with civil law countries, between public and private law.110 Codes regulate private law, constitutions public law, and organization of the state seems like the most appropriate subject for public law. Another piece of the puzzle would seem to deal with the concept of judicial role touched on earlier.111 The traditionalist-positivist theory of judicial role is marked by a great fear of acting politically, as the legislator; separation of powers is a political theory, but it seems to be less political than substantive values like dignity or equality. Making sure that the proper branches are performing political acts might seem very different than striking down one branch's substantive political act with an opposing substantive political conception.
ii. Substantive values
The jurisprudential theory of a "flat" constitution (and its related conception of judicial role) was linked, in the Colombian context, to certain substantive values. In the absence of a strong overlay of constitutional values to fill the legal order, "code values" played this role instead-those essentially politically conservative values reflected in the old nineteenth century codes, particularly the Civil Code.112
A wonderful example is found in a 1981 case where the court explored whether a law giving illegitimate children less of a father's estate than legitimate children was unconstitutional.113 The plaintiff, citing a number of constitutional provisions, made an argument based on principles of equality and dignity.114 The court complained about the "lack of rigor in the formulation of the charges, and concretely in the enunciation of precepts supposedly infringed," and stated that the plaintiff was "opposing her own, personal idea of justice to the conception of justice formulated by the legislator in her norm . . . [thus supposing] that her idea of justice is that of the constitution."115 The court also noted, dismissively, that the plaintiffs view reflected a "pure idea of immanent justice."116
The court placed the Civil Code against the notion of the broad, value-laden constitution. After rehearsing at some length the history of the (unequal) treatment of illegitimate children under the evolving Code, the court noted that the provisions served "valuable goals": "the preservation of sexual stability, the defense of the family, the identification of the father ... ; [these] are values of social order and collective security that it would be senseless to try and ignore."117 The court added that "the unequal treatment ... of children born outside wedlock compared to those born inside is a predominant legislative phenomenon in history" and that this was "easily" shown as being the result of "the constant and profound interest of the community ... in maintaining the regularity, solidity, and consistency of that original and basic focus of society that is the family."118 Furthermore, such a deeply-rooted legislative enactment was entitled to far more respect than the plaintiffs pure constitutionally-grounded theory of justice, because the enacted Code's values represented "a harmonization of the interests in conflict .... [that] supposes not only an axiological criterion, but [also] a painstaking sociological study of the interests in the era in which the legislation is transmitted."119 "Law," then, is the "positivist-historical product of what the community thinks and feels," and not something "abstract, ideal, and transcendent."120
Thus, the values of the statutory legal order, expressed most powerfully in the Codes, become the source of values across the entire system. Tested by the history of society, they are entitled to far more respect than some speculative values derived by judges from the constitution. Obviously, the conception of separation of powers and judicial role looms large again. A dissenter accused the majority of being "merely positivistic" and of lacking "ideology"the court, the dissenter argued, should have tested the legal rule against some normative conception of justice.121 This critique misperceives that the majority's notion of positivism is itself ideological-it is grounded in the historical values of enacted legislation. It is, thus, an ideology, and a rather conservative one.
This reverence for conservative code values extended beyond traditional regulation of family law and into the economic realm. A 1988 case considered the clash between an 1887 Civil Code provision providing that "ownership ... is the real right to a corporeal thing for enjoying and disposing of it arbitrarily, so long as it is not contrary to the law or the right of another" and constitutional provisions noting that property is a "social function entailing obligations."122 The majority held that there was no conflict: "The adjective 'arbitrarily' [in the Civil Code] is tempered in the same rule by prohibiting uses that violate the law or the rights of another, all of which implies that it is not an absolute faculty . . . ."123 Despite dating from a different era, the court fully adapted the Civil Code provision to the social enmeshment of property contained in the New Deal era124 constitutional clauses.
A close read of the decision, however, reveals that it was not simply the Civil Code provision that the court was reinterpreting; if the court was understanding the Code provision in a strangely communitarian light, it was also moving the constitutional provisions on property further towards individualism. The court, citing a variety of old cases,125 argued that the definition ... of ownership ... of the Civil Code, with its content as a subjective and individual right, has been considered as an essential element of its social function, thus that its constitutional protection obeys fundamentally the particular interest in ownership and also those of the community for the satisfaction and advancement of its social ends . . . .126
The individualistic conception of property enshrined in the Civil Code seems to have heavily influenced this scope of constitutional protection. As a dissenter noted, the court adopted this "individualistic and egoistic" conception of property, which the constitutional provisions of the 1930s tried to neutralize, into the new constitutional order: "[I]n Colombia they reform the texts but not the attitudes."127 This complaint expresses an interesting issue: As wedded as the traditionalists-positivists were to a positivist vision of the law, they had, through time, developed a very strong, intensely normative vision of the Code. The court could use this sort of spirit to strike down a claim to a distinctively normative spirit within the constitution, as it did in the family law case above, or instead the court could infuse the constitution itself with this spirit, rather than opposing the constitution and the Code, as here with private property. The Code, then, becomes the light behind the constitution, illuminating one of its aspects (private property) and giving it some real normative life. The court sometimes read the constitution through the Code.
One should not get the impression that the property area has been an extraordinarily active one, or that the court has developed a nineteenth-century conception of property in its jurisprudence. The court has recognized that property is a social institution and that legislators have fairly broad discretion to regulate it and related economic rights.128 Still, the court has been willing to imbue this section of the constitution with a spirit and to enforce it with some activeness. Most of its work here, unsurprisingly, has been in the expropriation area, where the court has noted that although property rights entail social obligations, they still exist in strong form and the state cannot take them without either full compensation or through the extinction of property rights through prescription.129 Via the compensation route, the court has insisted rigorously that the compensation be equal to the market value of the property taken.130 The prescription route is much more interesting, because here the court has stated that this constitutional method "of developing the social function of property"131 is an "ancient"132 concept regulated by principles found in the Civil Code.133 Thus, the court gives the constitutional mandate of the social function of property much of its life through a general principle it derives from the Code.
3. Summary of the 1980s CSJ
We have seen that the CSJ in the 1980s was a court dominated by career judges, and we have also seen that it displayed a traditionalist-positivist worldview. It read constitutions as being flat-composed of a series of fairly specific, lesser-order rules rather than as rules subordinated to and understood in light of higher constitutional principles and values. The result is that the court has generally downplayed the significance of the constitution. We have seen that this jurisprudential theory is linked to the traditional Latin American civil law conception of separation of powers and judicial role. The one exception to this general interpretative theory is in the structural area of defining which branches of government should perform which roles; here a more value-oriented jurisprudence has developed. Finally, the court's jurisprudential theory has led to a situation where traditional values from the codes have filled the gap in overarching legal ideals left vacant through this method of constitutional interpretation. Thus, this jurisprudence had a predictably, if loosely, conservative bent.
C. The Constitutional Court in the 1990s and Today: Law Professors and New Constitutionalism
1. Composition of the Court
The composition of the Constitutional Court in the 1990s was radically different from the composition of the CSJ in the 1980s. The CSJ, as we have seen, was a court controlled by career judges. In contrast, the first Constitutional Court selected in 1991, composed of seven members, had only three magistrates with substantial prior judicial experience, yet five of the seven were legal academics.134 Colombia selected a new, nine-member court in 1993; this court was composed of six academics and only two career judges.135 Thus, by 1993 the percentage of academics on the Court compared to the old CSJ had risen from roughly twenty-three percent to sixty-six percent, while the percentage of career judges had fallen from about fifty-five percent to twenty-two percent.136 As further evidence of legal academics' domination of the new Constitutional Court, one notes an explosion of public-law scholarship from the nine members of this new court, particularly in the early 1990s.137
Though important, explaining the reasons for this shift is difficult and well beyond the scope of the present Article. Partly, prominent politicians, including then-president Cesar Gaviria Trujillo, understood that legal academics, with their tendency to question certain aspects of traditional Latin American jurisprudence, would be more in accord with a progressive, new-constitutional political agenda. The shift was also partly a result of the pragmatic need to find people who understood constitutional law: since, as has been noted, the pre-1991 system utilized a centralized system of judicial review, the career judiciary-below the CSJ itself, which monopolized judicial review under the old system-was a bad place to find judges that were competent in constitutional law.
2. Jurisprudence of the Court
i. Interpretive Method and Judicial Role
A crucial starting point for the new court has been that constitutional interpretation is somewhat different from ordinary statutory interpretation in degree if not in kind.138 Constitutional doctrine starts from the notion that mere legal rules are insufficient to cover all individual cases. Rules would leave gaps in the legal order without some additional, creative judicial work:
The text of the law is not . . . susceptible to being applied mechanically in all cases, and that justifies the necessity of having the judge interpret and apply it, integrating it and giving it coherence, thus that equality can be realized in the most complete constitutional sense.139
This is far truer of the constitution than of ordinary statutes, given the vague character of constitutions. Some method is needed to bring clarity to the constitution and to make it effective: This method essentially starts from fundamental principles and values, According to Justice Rivadeneira, the constitution is largely comprised of these principles, which, even though they have an "open character," must not be "subordinated to the other, more concrete constitutional precepts .... Far from an opposition between the abstract and indefinite and the concrete and specific . . . interpretation should try to integrate [them] in a harmonious whole."140 Similarly, according to the court in an important 1995 case, general principles are an important way of making sense of the otherwise gap-filled constitution without giving way to judicial arbitrariness.141
If broad principles are thus suddenly relevant to constitutional interpretation, how exactly should the court use them? Based on a reading of Dworkin, the court seems to want to break down these general concepts into two groups: values and principles.142 Values identify the ends of the state and are chiefly aimed at the futurethey are things like "coexistence, work, justice, equality, knowledge, liberty, and peace captured in the preamble to the constitution. Also they include [things enumerated in one of the initial articles like] service to the community, general prosperity, . . . participation, etc."143 These ends are so broad that they exist "to resolve a problem of interpretation in which the sense of the law is in play, not to be applied directly . . . ."144 Principles are quite different. These are narrower concepts that "express norms for the present" and "consecrate general legal prescriptions that imagine a recognized political or axiological delimitation, and, in consequence, restrict the space of interpretation."145 These include things like the social state of law, participatory and pluralist democracy, prevalence of the general interest, and respect for human dignity.146 Because of their greater specificity, principles, in addition to being an "inescapable guide to interpretation," can also sometimes be applied directly to solve constitutional questions.147 In sum, rules, principles, and values all lie on a continuum in which, "as they gain generality they increase their space of influence but lose concreteness and capacity to apply directly to the concrete case."148 The court must interpret rules in light of principles and values to achieve coherence and harmony throughout the constitution as a whole, and must place these principles and values in a hierarchy to avoid high-level conflicts between general concepts.149
The post-1991 Constitutional Court thus abandoned the notion of the flat constitution where only specific legal rules had enforceable content for a complex, hierarchical constitutional structure that general principles and values dominated. They believed that they were doing something that was both quite new150 and fairly international151 in outlook. They even selected a new hero: In place of the old worship of Kelsen, the new court cited Dworkin's theories.152 This new approach should not be confused with a judicial emphasis on policy; policy arguments have never had much play in Latin American law.153 The argumentation from broad principles that Colombian constitutional jurisprudence has emphasized since 1991 is very different from the kind of pragmatic policy arguments that are familiar to us in modern, post-realist American law.
Lopez-Medina, in his study of Colombian "pop" jurisprudence, has emphasized the new court's role in changing the notion of precedent.154 This change in precedent, however, was only a part of the new court's general change in approach to the constitution, despite its obviously important consequences (if the Constitutional Court could succeed in forcing other courts to follow its holdings). Also, these precedential changes were, in fact, subsidiary to the new interpretive methodology explained above. When the legal order is seen as being composed of specific legal rules interpreted through the use of logic, precedent would seem to the actors involved as unnecessary-it should be simple enough, the traditional view goes, for judges to come to the right decision independently. Once the constitutional order becomes seen as gap-filled and is consequently complexified through the use of teleological methods of interpretation and vague general concepts like values and principles, then the need for precedent appears pressing to avoid blatant inconsistency in application of constitutional concepts. Indeed, the court used the dangers of inequality to justify a ratcheting up of the value of its own precedents in the 1990s.155 The court's steps towards altering the types of decisions it reached156-allowing: (1) holdings where a statute would only be considered constitutional if applied a certain way (conditional decisions);157 and (2) sentences adding something to a statute and thus correcting its omission (integrated decisions)158-should also be seen as subsidiary to its more general change in attitude towards constitutional law. These maneuvers were not exactly new,159 but they were applied far more confidently and far more often in the post-1991 period than before. They spring from the court's increasing sense that the constitution-and not something else, like the Codes-constituted the main well-spring of normative ideals within the legal order, and thus that the rest of the legal order must be infused with and, when necessary, reshaped by constitutional ideals.
A changing theory of judicial role has accompanied this new theory of a complex, value-laden constitution. The old notion of separation of powers emphasized the legislator as law-maker and the judge as law-applier, and inspired an extraordinary fear of judges making law. The new theory instead marks out a "creative"160 role for the judge in the "creation of law."161 Jurisprudential arguments about the changing nature of law back this new theory. Society and law have both become more complex.162 On the one hand, as we have already seen, vague values and principles instead of specific rules have dominated the legal, or at least constitutional, order, and it requires considerable judicial work to integrate legal rules with higher principles of the legal order.163 On the other hand, the post-1991 judges have tended to note a gap between judicial rules on the books, which are relatively abstract, and complex, fact-specific, modern social reality; only creative judicial action can bridge this gap.164 Thus, in addition to upward-looking work aimed at coherent integration of legal principles and rules, the judge is also engaged in a downward-looking task, required to mesh complex social facts with legal rules. As the court noted on one occasion, "the work of the judge cannot be reduced to a simple, mechanical application of [legal rules] to concrete cases, because that ignores the complexity and singularity of social reality . . . ."165 Judges on the post-1991 court have a somewhat increased penchant for delving into the material, rather than merely formal, situation faced by those groups hurt by a given law, and for using empirical, social scientific information to get a picture of social reality.166
Beyond these jurisprudential arguments for judicial role changes, there has also been some effort on an institutional level to alter the traditional theory of separation of powers. The political reality in modern, interventionist Latin American states has, according to some judges on the court, been one of "overwhelming growth" in the power of the executive branch and a "loss of political leadership" from the legislative branch.167 The legislature, traditionally the most legitimate part of a democratic state, has lost legitimacy as it has declined in leadership and fallen prey to special interest groups.168 This void in legitimacy must be filled by the court, which must take an active role to defend "the institutional order" and to reestablish a "true equilibrium and collaboration between the three powers."169 Otherwise, the "president will predominate."170 The judiciary cannot fill this void, however, simply by applying traditional-positivist conceptions of law. These traditional conceptions of law, like traditional conceptions of the legislature, are no longer seen as legitimate.171 Modern social actors, rather than looking for law applied hierarchically top-down, instead want consensus-based solutions that are particularized to specific social situations.172 The creative judge, who works at the level of individual dispute resolution, is best placed to integrate broad legal values, specific legal rules, and the specific factual situations of the actors to arrive at suitable solutions.173 In the modern order, the judge has become "the carrier of the institutional vision of the general interest."174
ii. Substantive values
Just as the flat constitutional theory of the pre-1991 CSJ was connected, albeit loosely, with certain substantive norms, the complex, hierarchical constitutional theory of the post-1991 Constitutional Court also seems to have some connection with a broad set of substantive values. The values imbedded in the old codes are no longer the chief source of normative spirit for the legal order; now the broad principles and values found in the constitution itself can play this role. These values tend to be relatively progressive in character. The court demonstrated this point quite early on; in a 1992 decision, Justice Giro Angarita Baron noted that the judiciary could no longer ignore the constitution's "generous" list of social, cultural, and economic rights as judges has done under the old system: The court had to give them judicial effect.175 In a decision later that same year, Justice Baron held that human dignity was the "supreme" principle of the 1991 constitution.176 This, of course, put the Colombian charter within an international family of modern, liberal, dignity-centered constitutions.177 Other members of the court have not seriously challenged the dignity-focused nature of the constitution.
A summary of some important cases since 1991 demonstrates the generally liberal bent of recent jurisprudence, as well as its connection to new constitutionalist interpretative methods. In 1994, the court held that possession and consumption of drugs for personal use could no longer be criminalized in Colombia.178 The court focused on broad constitutional principles of liberty, autonomy, and particularly "free development of personality," deducing the holding from these ideals.179 As the court noted:
When the state resolves to recognize the autonomy of the person .... this is defined as everything that corresponds to the zone of ethics: it leaves it up to the individual to decide ... on the good and the bad, on the sense of her existence. If the person resolves, for example, to dedicate her life to hedonistic gratification, it does not interfere with that decision while that form of life, in the concrete, and not in the abstract, does not harm anyone else .... If the right to free development of personality has any force inside our system . . . the rules that make consumption of drugs a crime are clearly unconstitutional.180
Hence, the court was interpreting the constitution as having an underlying philosophy that was liberal, and almost libertarian, in its content.181
In recent years, the post-1991 court has turned away from its early emphasis on liberty and formal aspects of equality182 towards an emphasis on social, cultural, and economic rights and a conception of material equality. For example, the court has used the constitutional right to dignified housing to strike down several laws dealing with home loans. In 1998, the court struck down a law forbidding prepayment of loans where interest was charged.183 In 1999, the court invalidated a law tying movement in a formula used to index the interest rates on home loans for inflationary changes to general changes in the national interest rate.184 The court's general approach in these cases was to start with the general, exceptionally broad concept of the "social state of law," beneath which existed other broad ideas like human dignity, "service to the community," and "the creation of a just order."185 The court then linked social, cultural, and economic rights, and particularly the right to dignified housing, as integral parts of these broader ideals.186 The right to dignified housing and its subparts (such as the state's obligation to provide adequate long-term housing), even though they were intended to be progressive measures and not to be fully applied immediately,187 seemed to have enough content once filtered through the broader concepts to be applied directly to the factual situation. For example, the state had an obligation, under the new and non-individualist order, to "establish [ ] specific plans for the less wealthy classes of the population."188 The general technique, again, is quite comfortable starting from the broad values at the top of the constitution and working its way down to narrower, but still quite un-Kelsen-like, notions of law. The narrower rights are applied in light of the broader ideals, and once again, the general philosophy of the constitution seems liberal in content.
The court's recent decision to invalidate the application of a value-added tax (VAT) to basic goods and services used almost an identical methodology.189 The court began by citing the social state of law as the basic principle of interpretation for the other norms involved.190 It then held that the VAT violated a right to a minimum of support for living,191 a right that seemed to come from a combination of constitutional clauses.192 The VAT case also strongly defended the use of socio-economic context to resolve cases193 and conducted such an exploration, including detailed empirical social-scientific data.194
We should be wary of thinking that the political-substantive values favored by adherents of the post-1991 interpretative method have been particularly uniform. A 1994 decision, for example, required the criminalization of abortion while using classic new constitutionalist interpretative methods: The right to life was seen as trumping any countervailing rights of the mother, given that it was the "ontological substratum for the existence of the other rights" and thus must be higher than other rights in the hierarchy of principles and values.195 Still, there is certainly a loose, understandable relationship between interpretative method, conception of judicial role, and substantive values. Some decisions of the new court, like the decision to decriminalize personal drug use and the decision to strike down the VAT, would have been simply unthinkable under the old style of jurisprudence, whereas some decisions of the old court, like the decision to deny equality in inheritance to illegitimate children, are equally unthinkable under the new style.
3. Summary of the post-1991 Constitutional Court
As we have seen, this was a court composed largely of constitutional law scholars, and it strongly espoused new constitutionalist jurisprudence. This court no longer viewed the constitution as flat and enforceable only through relatively specific rules; rather, the court saw the constitution as a complex amalgam of rules, principles, and values, ordered in a vertical hierarchy for purposes of interpretation. The court also recognized social reality as being quite complex. It linked this new jurisprudential theory both with a new conception of the judicial role-the creative judge, restoring an imbalance in the balance of powers-and loosely with a set of progressive values that seem imbedded in the constitutional text.
V. WHY DOES THIS MATTER? LINKING JUDICIAL WORLDVIEWS TO "REAL" SOCIAL PHENOMENA
A. Worldviews and Outcomes
The attitudinalists believe that there is no causal link between what a judge says in his opinion and the outcome that that judge reaches-the outcome is caused by the judge's substantive, political policy views, whereas the opinion is simply ex post legitimation. As Segal and Spaeth have noted,
We . . . considered the legal model, which holds in one form or another that justices make decisions influenced by the facts of the case in light of plain meaning, the intent of the framers, and precedent. While the Court uses these factors to justify its decisions, they do not explain their outcome.196
The case study of Colombian judges demonstrates that it is implausible to assert that there is no causal link between judicial wordviews, as manifested in opinions, and judicial outcomes, for reasons rooted in the interdependency of ideas.197 That is, the very richness and complexity of the relationships between substantive policy values and two other types of ideas explored above-theories of judicial interpretation and ideas about what facts are relevant in a given type of case-make the notion of unilateral causality seem hard to sustain. Political policy preferences are part of judicial behavior in Colombia, but they are also intertwined with other ideological constructs in a dense web of reciprocal causation.
First, discussed in Part III above, substantive values in Colombia have some relationship with interpretative methodologies. I have generally drawn the causal arrows as running from interpretative theory to policy values, but I recognize that they probably run in the reverse direction as well. This relationship should not be seen as a particularly tight one: As shown above, a wide range of values can fit within each worldview. For example, the post-1991 court has been much more disruptive of the prevailing governmental political economy in recent years, when it has turned towards material equality and social, cultural, and economic rights, than it was in its early years when it focused on formal equality and liberty. But there is some loose relationship nonetheless. An attitudinalist thus might still be able to use clusters of policy values-roughly, liberal vs. conservative-to predict judicial behavior, but he would be missing the underlying causes of those values themselves-conceptions about the nature of legal interpretation and judicial role.
Furthermore, a judge's ability to maximize political policy preferences depends on the types of facts he sees as relevant in a given case, and these facts themselves are a product of judicial worldview.198 As we have seen, traditionalists-positivists tend to want to decide cases on structural grounds-the facts that they would seem to focus on are thus facts dealing with the procedural legitimacy of the rule-making process.199 New constitutionalists tend to see nitty-gritty social scientific facts-they try to see the material, rather than formal, situations of those affected by laws.200 Obviously, the facts seen in a case will affect the way that a court decides that case. Attitudinalists could claim that a judge's preexisting political policy preferences predetermine the judge's vision of which facts are relevant in a given case, but this does not seem likely.
If roles predictably affect outcomes, then one interesting insight of my theory is that, like attitudinal theory and unlike strategic theory, it focuses attention back on the selection processes for judges. The key to determining judicial behavior is to see what type of social actor-career judge vs. public law scholar-was appointed, rather than to look at incentives judges face once they are already on the bench-like threats of reversal, censure, and removal. Still, if certain judicial worldviews, which are predictably carried by certain social actors, increase the probability of a given set of linked substantive case outcomes, and if politicians know this, might politicians try to select social actors that are likely to carry worldviews which are favorable to those politicians' preferred substantive policy outcomes? Thus, the old conventional wisdom of the liberal politician appointing a liberal judge and the conservative politician appointing the conservative judge would be complicated a bit; politicians would instead appoint actors with legal worldviews that, after a working through of interpretative methodologies and the like, were likely to produce favorable substantive outcomes for those politicians. In future work, I hope to be able to explore whether these sorts of appointment practices have occurred in Latin America.
B. Worldviews and the Legitimacy of the Court
Segal and Spaeth seem to argue that the language of an opinion itself is not an important social phenomenon: What really matters is what drives judicial outcomes, not the language judges use to get there.201 This ignores the impact that judicial opinions, independent of outcomes, can have on public opinion. Jonathan Miller has recently taken a stab at this issue in the Latin American context, arguing roughly that the decline in the legitimacy of the Argentine Supreme Court can be explained as a shift away from the traditionalist-positivist jurisprudence of the old court towards a new, socially responsive jurisprudence in which the court must rely on-using Weber's categories of legitimacy-charisma rather than law as the ultimate source of its authority.202 When forced to battle for charisma with other, more naturally charismatic bodies like the president and the legislature, the court has not fared well.203
The biggest problem with this theory is its use of charisma as a catchall black box for all modern legal action outside of the traditional-positivist sphere.204 Post-1991 Colombian constitutional judges see their decisions as grounded in factors that are just as objective and legal as the underpinnings of traditionalism-positivism; they do not see themselves as doing something that is arbitrary or grounded only in their personal or institutional magnetism. However, Miller's explanation for the stylized fact that Latin American courts have lost legitimacy in recent years is onto something. What he may be trying to get at is a social difference as to what counts as "legal action." The attitudes of the Colombian career judges identified in this Article regarding what law is may be closer to the attitudes of most ordinary people in Latin American society. The attitudes of the constitutional scholars, which are self-consciously elitist and transnational in orientation, may be further from these normal social attitudes. If this is so, it suggests that shifts from traditional-positive views to new constitutionalist views may threaten the power of courts as institutions. This is ironic, given that proponents of these views envision giving courts a more central role within the separation of powers system than was true traditionally,205 that traditional Latin American law has long been perceived as being out of touch with social reality,206 and that one goal of the new constitutionalists has been to help bring law closer to that social reality.207 A related theory might be that any change from one type of legal worldview to another, at least if insufficiently gradual, destabilizes judicial legitimacy.
VI. CONCLUSION
Colombian constitutional judges have worked under two quite different, yet equally rich, legal worldviews. Each of these worldviews combined interpretative theories, ideas about judicial role, and substantive policy goals. Equally important, each was predictably linked to a certain type of legal actor who functioned as its carrier-career judges in one case, legal scholars in the other. Rational choice is not the only perspective that is capable of producing structured, testable theories of judicial behavior. Legal scholars potentially have much to contribute to the literature on judicial behavior.
A complete theory of judicial behavior requires more than a general theory of human nature; it requires a specific understanding of how and what lawyers, judges, and others think about law. Those of us who are comparative scholars should say even more: We must study not just the understandings of the legal community in general-although this of course will always have relevance to the extent that lawyers in one country see themselves as sharing worldviews with lawyers elsewhere-but those understandings in the exact area that we are studying. Ultimately, this is because law is largely an autonomous sphere of society that courts and legislatures have constructed quite differently from other spheres of our social life. Law and economics, rational choice, and related theories of economic rationality are problematic partly because they miss this point: They posit universal accounts of motives and values that simply do not exist, and they tend to ignore the things that people (judges, for example) actually say.208
The ending point need not be the discourses that legally-oriented individuals use in their own interactions: judges can lie to others and they can delude themselves. Still, these discourses must at least be our starting point.
IMAGE FORMULA 2IMAGE FORMULA 3IMAGE FORMULA 4IMAGE FORMULA 5IMAGE FORMULA 6IMAGE FORMULA 7IMAGE FORMULA 8IMAGE FORMULA 9IMAGE FORMULA 10IMAGE FORMULA 11IMAGE FORMULA 12IMAGE FORMULA 13IMAGE FORMULA 14IMAGE FORMULA 15IMAGE FORMULA 16IMAGE FORMULA 17IMAGE FORMULA 18IMAGE FORMULA 19IMAGE FORMULA 20IMAGE FORMULA 21IMAGE FORMULA 22IMAGE FORMULA 23IMAGE FORMULA 24IMAGE FORMULA 25IMAGE FORMULA 26IMAGE FORMULA 27IMAGE FORMULA 28IMAGE FORMULA 29IMAGE FORMULA 30IMAGE FORMULA 31IMAGE FORMULA 32IMAGE FORMULA 33IMAGE FORMULA 34IMAGE FORMULA 35IMAGE FORMULA 36IMAGE FORMULA 37IMAGE FORMULA 38IMAGE FORMULA 39IMAGE FORMULA 40IMAGE FORMULA 41IMAGE FORMULA 42IMAGE FORMULA 43IMAGE FORMULA 44IMAGE FORMULA 45IMAGE FORMULA 46IMAGE FORMULA 47IMAGE FORMULA 48IMAGE FORMULA 49IMAGE FORMULA 50IMAGE FORMULA 51IMAGE FORMULA 52IMAGE FORMULA 53IMAGE FORMULA 54IMAGE FORMULA 55AUTHOR_AFFILIATIONDAVID LANDAU*
AUTHOR_AFFILIATION* Law clerk for the Honorable Sandra L. Lynch, U.S. Court of Appeals for the First Circuit. A.B. 2001, Harvard College; J.D. 2004, Harvard Law School. My sincerest thanks to Christine Desan, Jonathan Miller, Duncan Kennedy, and Ramon Eduardo Madrinan Rivera for their assistance with this project.
IMAGE CHART 56APPENDIX: DETAILED EMPIRICAL DATA ON COLOMBIAN JUDGES, LISTING THEIR PRIOR CAREER EXPERIENCE
IMAGE CHART 57APPENDIX: DETAILED EMPIRICAL DATA ON COLOMBIAN JUDGES, LISTING THEIR PRIOR CAREER EXPERIENCE
IMAGE CHART 58APPENDIX: DETAILED EMPIRICAL DATA ON COLOMBIAN JUDGES, LISTING THEIR PRIOR CAREER EXPERIENCE
In addition, make sure to read these articles:
- Institutionalizing global wars: State transformations in Colombia, 1978-2002
- COLOMBIA'S URIBE GOES FOR SECOND ROUND
- Judiciary.
- The jurisprudence of uncertainty
- LATIN AMERICAN OIL OFFICIALS STRESS THE NEED FOR STABLE CRUDE PRICES
- The politics of research access to federal court data
- Judicial Review in State Supreme Courts: A Comparative Study
- Destigmatizing the Reassignment Power
- Gender Effects in the Courts of Appeals Revisited: The Data Since 1994
- Whither the race? A comment on the effects of the Delawarization of corporate...
- Premier Announces New Rational Choice Rx(R) Program; Percutaneous Coronary Intervention...
- CONSTITUTIONALISM AND NEW DEMOCRACIES
- A defense of empirical legal scholarship
- Emergence from the Dilemmas of Justice Reform