Microsoft Word - IL36.3 Winczorek BSEDIT.docx © Jan Winczorek. Informal Logic, Vol. 36, No. 3 (2016), pp. 371-394. Redundancy of Redundancy in Justifications of Verdicts of Polish The Constitutional Tribunali JAN WINCZOREK Departament of Sociology of Law Faculty of Law and Administration University of Warsaw Krakowskie Przedmieście 26/28 00-927 Warsaw, Poland janwin@janwin.info Abstract: The results of an empirical study of 150 justifications of verdicts of the Polish Constitutional Tribunal (CT) are discussed. CT justifies its decisions mostly on authoritative references to previous decisions and other doxa- type arguments. It thus does not convince the audience of a decision's validity, but rather documents it. Further, the methodology changes depending on features of the case. The results are analysed using a conceptual framework of sociological systems theory. It is shown that CT's justification methodology ignores the redundancy (excess of references and dependencies) of the legal system, finding redundancy redundant. This is a risky strategy of decision- making, enabling political influence. Résumé: On discute des résultats d'une étude empirique de 150 justifications des verdicts du Tribunal constitutionnel polonais (TC). Le TC fonde ses décisions principalement sur des références faisant autorité aux décisions antérieures et sur d'autres arguments de type doxa. Il ne convainc donc pas le public de la validité d'une décision, mais il documente la décision. En outre, la méthodologie change en fonction des caractéristiques de l’affaire. On analyse les résultats en utilisant un cadre conceptuel de la théorie des systèmes sociologiques. On démontre que la méthode de justification du TC ne tient pas compte de la redondance (excès de références et de dépendances) du système juridique, et ne se rend pas compte de la redondance redondante. Cette approche de prise de décision est risquée, car elle permet une influence politique. Keywords: Constitutional Tribunal, content analysis, systems theory, Poland, redundancy 1. Introduction There are many views on how constitutional courts should make decisions. Classical Austrian, German, French and American doctrines all have their different approaches to the matter. Jan Winczorek © Jan Winczorek. Informal Logic, Vol. 36, No. 3 (2016), pp. 371-394. 372 Various authors express various degrees of acceptance for judicial activism and passivism depending on their world-view (directly declared or accepted implicitly). Theories of constitutional adjudication evolve over time in connection with the broader legal and political landscape of a jurisdiction, and the courts' jurisprudence is subject to professional critique of constitutional lawyers. Finally, in the lay discourse decisions on laws' constitutionality are evaluated from a political and a moral perspective. This list is long, but does not yet exhaust all possible perspectives on constitutional jurisprudence. Another possibility is to look at the topic using theoretical perspectives and the methodological lenses of the social sciences. Approaches to constitutionalism in economic analysis of law and political science are well known, but also sociology has a say in this domain. While it is not necessarily the role of a sociologist to evaluate the rightness of courts' jurisprudence from an internalist legal perspective (but see Banakar 2015, chapter 11, Lautmann 2011/1972), s/he is able to grasp its complexities from an external perspective. In particular, s/he may undertake empirical research and develop a conception of legal interpretation aimed at creating an explanatory model of constitutional jurisprudence as a social phenomenon. Such a task can be accomplished even without strictly expert knowledge on current tendencies in jurisprudence and certainly without taking a position in a debate on the rightness of specific judgements or their lines. A natural consequence of this approach is to treat the body of adjudication, together with theoretical conceptions it might reflect, as a Mannheimian ideology. It is thus perceived without any evaluation whatsoever, as a set of beliefs of a particular group of individuals, socially produced and reproduced, and heavily impacting the functioning of the social institutions to which it pertains. In this perspective, court ideology must be considered as an important factor impacting both decision- making and production of justifications for it, such that it cannot be ignored in any attempt to explain the patterns of constitutional jurisprudence. Clearly, even the externalist, disengaged and non- evaluative approach to court adjudication should not prevent the sociologist from paying attention to the complexity and contextual nature of the studied material. This perception must be shared with practising lawyers and legal scholars who apply the internalist perspective. For that reason a sociologist aiming at accomplishing such tasks needs good conceptual tools—a proper theory and an adequate research technique that together will constitute a correct method of collecting and analysing data. Redundancy of Redundancy in Justifications of Verdicts © Jan Winczorek. Informal Logic, Vol. 36, No. 3 (2016), pp. 371-394. 373 This text reports—in a modest scale limited by its volume—on an attempt at using a sociological theory and a research methodology for such a purpose. In the author's view, they both have such a “requisite variety”. 2. Constitutional adjudication as an object of sociological reflection It has been long noted that social sciences tend toward reductionism (see extensively Garfinkel 1981). This pertains also to social-scientific research on law (Teubner 1983:247), including studies of adjudication. Actions of legal actors and decisions of legal institutions are frequently described as dependent on single factors, as in many versions of the “judicial behaviour” approach (see discussion in Dyevre 2010). Authors of such studies are usually at pains to demonstrate legal indeterminacy and the role of extralegal circumstances in decision-making, but do not always produce a consistent picture of the impact such circumstances have. One example could be the continued debate on the role of personality and private views of judges in adjudication. Many empirical studies on the topic have been conducted, sometimes corroborating such an influence, and sometimes falsifying it (See only Weinshall- Margel 2011; Keele et al. 2009; Brace, Langer, and Hall 2000; Sisk, Heise, and Morriss 1998; Segal and Cover 1989). Thus—utility of many individual studies based on such premises notwithstanding—the reductionist approach as a whole is all too often disappointing, also because the theoretical models they utilise are incomplete. One of the few theoretical conceptions in sociology to undertake legal topics such as this issue from a different perspective—in a way taking law seriously—is Niklas Luhmann's social systems theory (see Luhmann 1993; Winczorek 2009). There is no point in discussing it here, particularly as its scope by far extends beyond legal or even socio-legal issues in the broadest sense, and a number of works elaborate on its consequences for constitutions and constitutionalism or develop theories on similar premises (see only Baghai 2015, Teubner 2012, Thornhill 2010, Teubner 2004, Neves 2001). Suffice it to say that the said author accepted universalist assumptions and aimed at creating a comprehensive, deductive social theory, able to describe and explain all manifestations of social life, including legal phenomena. This subsumptive approach to theory building does not yet justify the view that it lacks original conclusions for every specific social field to which it is applied. Jan Winczorek © Jan Winczorek. Informal Logic, Vol. 36, No. 3 (2016), pp. 371-394. 374 Still, in the current context, a number of the theory's features should be mentioned. Most importantly, it is radically anti- reductionist in that it describes the law as a self-contained communication system, organized according to its own internal logic. Even though it responds to external events, the form, scope and impact of such influences eventually depend on the system itself. Consequently, it cannot be steered from the outside. This follows from the fact that all communications referring to it belong to the legal system and are constructed by it. This is the driving factor of system's evolution and in this way—which is denoted by the famous concept of autopoiesis— the nature of law is determined by law itself. A text of every legal provision, every verdict of a court, every doctrinal commentary to both, every motion filed by an attorney, every contract, every discussion among professional lawyers, and every oral communication between lawyers and laypeople and laypeople themselves—are elements of the legal system because they all remain related to all other elements functionally and genetically. Emergence of every new such communication changes the conditions of emergence of further communications. These interdependencies between all elements of the legal system, as well as their emergent nature does not yet mean that systems theory perceives all such elements as equally important. To the contrary—it assumes that they are connected in such a way that the conditions of making further communications, and reproducing them, all depend on other communications. In this view, the legal system is thus structured temporally but also hierarchically and functionally. The consequences of this way of thinking of the conceptualisation of adjudication are far-reaching. While reproduction of communication in the legal system is shaped by the different interdependencies between its elements, it is strongly—and in a number of ways—influenced by the final binding decisions. Crucially, their influence is not exhausted by the fulfilment of a purely validative or thetical function. They also define the meanings of concepts, determine binding values and delimit the possibilities of undertaking effective actions. At the same time, consistently with intuitive lawyer knowledge and many views expressed in legal theory, systems theory stresses the contextual, indexical nature of adjudication itself. In its view an isolated judgement has no relevance for communication in the system, if it may ever exist. It thus may not be uttered in a truly arbitrary way, without an implicit reference to other communications because the conditions of it being communicated are defined by other elements of the Redundancy of Redundancy in Justifications of Verdicts © Jan Winczorek. Informal Logic, Vol. 36, No. 3 (2016), pp. 371-394. 375 system, particularly interpretation theory. Such elements are shaped by political decisions and earlier adjudication, but also theory, doctrine and practice of law. In this light, interpretation methods used by a constitutional tribunal (hereinafter CT) appear as an important factor shaping the functioning of the legal system. Obviously, they are of fundamental importance for legal communication referring to legislation, because they eventually determine whether some norms are binding or not. Yet, they also have more indirect and less clear influence on other communications in the legal system, because they contribute to co-production of other elements in that system. In the Polish context the somewhat ambiguous but graphic term "radiation of jurisprudence" has been coined (Limbach 1999; Łętowska 2006) to describe just that effect. It is supposed to refer to non-determinative, persuasive influences of CT's adjudication on the activity of other adjudicative bodies. Looking at such linkages more analytically, one should say that previous and expected judgements influence the way law is described outside of the CT, because familiarity with these judgements makes it possible to evaluate the chances of success in the future trials. This pertains not only to the CT. Not infrequently, due to functional division of labour in the legal system, CT's verdicts set forth the expected adjudicative standards in cases having a constitutional aspect, which are heard before other courts. The central position of constitutional adjudication in the legal system notwithstanding, the CT itself is connected to other elements of the system, including those that remain outside of its sphere of influence. For instance, the process of appointing justices, even if it is political and not strictly based on merits, determines the minimal contents of judges' professional habitus. The adjudication culture sets forth the dominant ways of referring to the provisions of law and making decisions. It is itself co-determined by legal doctrine, the community of practising lawyers, the history of state's structure and other similar factors. The motions to control constitutionality of a law restrict the substantive scope of adjudication by providing standards of control and defining the subject matter of the case. Furthermore, none of these factors can themselves be seen as the only independent variable, because each of them is co- determined by further factors in the system. These multi-faceted connections between the adjudication methodology of a CT and other elements of the system can aptly be described with the system theoretical concept of redundancy (Luhmann 1993:352f; Luhmann 1985:15f). It refers to the fact Jan Winczorek © Jan Winczorek. Informal Logic, Vol. 36, No. 3 (2016), pp. 371-394. 376 that elements of the system are connected in such a way that the removal of one of them does not result in a radical change of the entire system, even if that element is connected to others and co- defines them. In other words, the identity of every element of the system is defined by a network of relations with other elements, which are also further related to one another. In such a network causal relationships cannot be easily determined. Still, having information about some of the network's elements allows us to determine the identity of other such elements—hence the network is redundant. In system-theoretical terms (not just those of Luhmann, but in the entire broad field), redundancy so perceived has two related features. On one hand, the term refers to oversaturation and excessive interconnectedness of references, implicit and explicit in the system. This is useful, for instance, in situations where continuous or error-free operation of particular technical systems must be safeguarded and the uncertainty of their failure be reduced (Streeter 1991). Yet on the other hand, it suggests that redundant communication is spurious. In terms of information theory, redundancy is thus unnecessary duplication of information to be transmitted, such that it increases the size of a message, decreases the speed of its delivery and negatively influences its overall efficiency even if it reduces entropy (Skyttner 2007: 249-250). This is well visible in the case of adjudication. On one hand, a particular way of judging and justifying decisions, once established in the legal system, is sustainable because it is produced by many references within that system. For instance, a change of rules on how the justices of a CT are appointed does not suffice, ceteris paribus, for a radical transformation of its jurisprudence, even if it might to some extent result in changes in the habitus of persons holding such position. Similarly, alterations in how motions to CT are written might allow it to change the way of adjudicating, but such an occasion might not be used if the dominant concept of law is not changed. Such interdependencies can be counted by dozens. On the other hand, the multitude of references which contribute to decision-making is very difficult to reflect and communicate within the limits of a decision's justification, requiring that CTs act both selectively and comprehensively. It is not a coincidence that many courts of final instance, the Polish CT included, produce very long and complicated justifications of their decisions. This, hypothetically, might be an outcome of the attempt to provide as detailed arguments for the decision as they can be, such that they provide the maximum chance that the verdict will be accepted by possibly wide Redundancy of Redundancy in Justifications of Verdicts © Jan Winczorek. Informal Logic, Vol. 36, No. 3 (2016), pp. 371-394. 377 audiences. Seen in this way, the necessity to reduce underlying complexity of the constitutional adjudication is an important limitation of CTs' legitimacy. 3. Content analysis as a method of studying adjudication Examination of utility of this conceptual framework for the study of constitutional adjudication must be postponed until empirical data is presented. This in turn requires that the method of producing it is made clear. The study under description utilised content analysis, a research technique allowing insight into cognitive categories used by lawyers or any other group of creators of symbolic communication (See generally Krippendorff 2004). Its main operating principle is to ascribe "codes" to individual segments of studied material, such that it can synthetically report on their meaning, and then to analyse such codes quantitatively or qualitatively. The organised nature of this method stems from the fact that it aims at minimising researcher's subjectivity in how the communication is perceived. This is achieved by utilising systematic code-list building strategies, segmenting the material according to certain assumptions, using triangulation and selecting content for analysis systematically. All these aspects distinguish content analysis of such legal material as court decisions and their justifications, from a standard juridical approach, particularly legal analyses of adjudication. It is comprehensive, usually refers to the entire corpus of texts or their systematically selected sample and normally pays little attention to the importance of individual verdicts. Furthermore, normally content analysis does not undertake any attempt to verify the correctness or incorrectness of court decisions and their justifications. Instead it describes the inter-subjectively measurable features of the material studied. While American authors of a state of art remark that content analysis is used often enough to be considered as a standard method in the legal scholarship (Hall and Wright 2008), in Europe such studies are much less frequent or at least infrequently published in international outlets. Still yet, even in the American context such studies are conducted rarely enough not to be canonical in the legal scholarship. For that reason content analysis is at times rediscovered by new authors who apply it for their own purposes (Hall and Wright 2008). Its uses can yet be catalogued. It is possible to distinguish analyses of contents of decisions, inquiries into their justifications and the factual references they make, as well as Jan Winczorek © Jan Winczorek. Informal Logic, Vol. 36, No. 3 (2016), pp. 371-394. 378 studies exploring consequences of specific verdicts or legislation for further legislation and jurisprudence. They are conducted as both qualitative and quantitative (Hall and Wright 2008). One important aspect of the different takes on content analysis, stressed by authors of said article, is that it allows one to notice common features of decisions which otherwise were not seen as linked or are particularly complex. This allows one to notice an adjudication pattern or an argumentation pattern invisible from the viewpoint of standard juridical methodology (Hall and Wright 2008:91—92). These features of content analysis as a research methodology correspond with the aforementioned assumptions of systems theory. The method allows one to notice the general and systemic features of CT's verdicts, as well as their linkages with doctrines and adjudication in other segments of the legal system. At the same time, by taking an external perspective and being programmatically blind to the disputes within legal theory or jurisprudence, it calls for inter-subjective description of judgements under study. All those advantages notwithstanding, studies of adjudication utilising content analysis have their weaknesses. They are by nature limited to the published material and do not take into account the deliberation processes, possible negotiations between judges in deciding panels nor other group dynamics. They also disregard the perspectives of individual judges, put aside psychosocial aspects of adjudication, and completely ignore the organisational dimension of adjudication. In that sense they are selective, treating courts as a black boxes. Still, by being sensitive to complexity of studied judgements they stress the aspect of constitutional adjudication that is also relevant for an observer located within the legal system, such that has a fundamental influence on the functioning of the system as a whole. 4. Presentation of results of the study In the remaining part of this paper the sketched theoretical perspective is juxtaposed with the results of an empirical study, accomplished by means of the research technique under description. In the course of doing so, the outcomes of a larger research project on constitutional interpretation completed in the years 2010-2012 by a University of Warsaw team led by T. Stawecki (Stawecki and Winczorek 2014) are utilised. The purpose of the project was to determine the patterns of constitutional interpretation in the adjudication of Polish CT. Redundancy of Redundancy in Justifications of Verdicts © Jan Winczorek. Informal Logic, Vol. 36, No. 3 (2016), pp. 371-394. 379 "The patterns" were understood as empirical regularities in how the constitution had been interpreted, observable in a corpus of justification of verdicts. In the part discussed here, the study was conducted as a content analysis of a quota sample of 150 justifications of CT's judgements, selected from all judgements concluding the case on the merits. Sample selection scheme is described in Table 1 (For more detail see Winczorek 2014b). The selection process was organised in such a way that verdicts marked with particular types of signatures and issued before and after the Polish constitution of 1997 came into force were proportionally represented1. Population Sample Until 1997 After 1997 Sum Until 1997 After 1997 Sum Signe d N % N % N % N % N % N % K 217 41,9 7 434 31,2 0 651 34,1 4 17 41,9 7 34 31,2 0 51 34,4 0 Sk 0 0,00 546 39,2 5 546 28,6 2 0 0,00 43 39,2 5 43 28,6 2 P 40 7,74 297 21,3 5 337 17,6 7 3 7,74 23 21,3 5 26 17,7 1 U 137 26,5 0 83 5,97 220 11,5 4 11 26,5 0 7 5,97 18 11,7 1 W 102 19,7 3 0 0,00 102 5,36 8 19,7 3 0 0,00 8 5,48 S 21 4,06 31 2,23 52 2,73 2 4,06 2 2,23 4 2,75 Sum 517 100 1391 100 190 8 100 41 100 109 100 15 0 100 % 27,1 0 72,9 0 27,1 0 72,9 0 Table 1: Sample selection scheme The research team, consisting of six members, analysed the contents of corpus using a software package “RQDA” aiding content analysis (See Winczorek 2014a). The analysis only covered a part of every justification—it was limited to the section presenting CT's own reasoning, not the parties' positions. 1 In this (non-probabilistic) sense the sample is representative. The main criterion for credibility of result and possibility of projecting them onto the population are the p values calculated for Chi2 test. It is useful to note that seen in this way the sample was in fact layered. Jan Winczorek © Jan Winczorek. Informal Logic, Vol. 36, No. 3 (2016), pp. 371-394. 380 This included marking a segment of text—which in the view of a researcher constituted a relatively complete utterance—with a synthetic code describing the contents of the segment. For the purpose of the study such chunks of text were called "arguments"2. Because perceptions of text are subjective, the study employed the principle of personal triangulation—every justification was coded independently by two different researchers. Since every researcher in the six-person team coded 50 justifications, in total 150 individual justifications were analysed. As a consequence, 300 instances of analysis of justifications were included in the dataset. Altogether, 15080 segments of text were marked with at least one code, and the total volume of coded text came to 149 million characters. This means that on average every justification under study comprised approximately 50 arguments, and that total volume of text in segments was approximately twice as large as overall volume of justifications. The latter outcome follows from the fact, that every segment of text could be coded with an arbitrary number of codes and the codes could be crossed. One advantage of the approach was that all persons engaged with the study (hereinafter: “coders”) were educated as lawyers. Most of them could be perceived as experts in legal interpretation. This made it possible to use relatively complicated code categories that at the same time were in principle not unfamiliar to legal theory. The list of codes was developed in an inductive procedure. Before the main study was executed, a pilot study was conducted in which every coder independently coded 5 justifications of verdicts of CT. At that stage coding was free—not limited by pre-established set of codes and without superimposed coding unit. Instead, codes were developed in vivo. Because judgements were selected in such a way as to be as different as possible, and because the coders did not communicate during the pilot study, a large list of categories was produced, coming to 312 unique codes. Once the results produced by individual coders had been compared and their meaning discussed, their number was greatly reduced. Eventually 37 first-order codes were used in the main study, describing utterances of CT along with 12 second-order codes, which made it possible to identify the type and features of a normative act to which such an utterance pertained. A few 2 The usage of the term in this paper is imprecise and has no grounding in any legal (or otherwise) theory of argumentation. Redundancy of Redundancy in Justifications of Verdicts © Jan Winczorek. Informal Logic, Vol. 36, No. 3 (2016), pp. 371-394. 381 examples of usage of codes are provided in Table 2, for full list of codes see Table 4 below. Code Example of coded segment Ref. arguments of adjudication— argument of CT's own adjudication— further justified The constitution stipulates (…) that common courts in principle bear the duty of providing justice in the name of the Republic and their competence is in fact presumed (…) In this way—as pointed out by the CT in its verdict (…) the constitutional standard is that „regular legislation may only determine the type of court (…)” P 09/02 arguments of adjudication— argument of CT's own adjudication— no further justification The meaning of the CT's verdict of 8 November 2000 is to point at the necessity of balancing the universities' decisions on the amount of tuition. Taking this as the correct way of solving the problem, CT points out at the fact that in the case under review it is reflected by the said (…) scope of judicial review. Sk 39/05 argument of limits of CT's authority In the case under review CT is not obliged to evaluate the constitutionality of (…) the statute in the light of article 17 of the Constitution, even if it finds it to be the standard of control included in the petition. To derogate a provision it is sufficient to establish that it is inconsistent even with one provision of the Constitution K 30/06 reference to the normative text Introduction of measures aiming at coordinating duties of public authorities with guarantees of art. 21 par. 1 and 64 par. 1 of the Constitution belongs to the lawgiver. As an example, the regulations can be mentioned (…) which are included in (…) statute on spatial planning (…) statute on environmental protection (…) statute on water regulation (…) K 20/07 argument by a reference to the principle under interpretation Among the freedoms and personal rights of citizens the constitution regulates basic principles of criminal responsibility. This includes principles of nullum crimen sine lege and the presumption of innocence (praesumptio boni viri), expressed in article 42 par. 3 (,,,) The principle of innocence expresses the basic safeguard of freedom and human rights. As such it fulfils a protective function. The imperative which follows from it orders to consider everyone innocent until the court states otherwise. P 01/99 semantic arguments - argument of legal language (…) The regulation upkept the entitlements to exemptions from import duties for persons who are resettling, yet currently they are described as „natural persons, who under law on foreign currencies are foreign persons staying permanently in the country for at least 12 months” K 05/96 Table 2: Examples of coding. Coded segments are abbreviated for more clarity. All translations by the author. The analysis of the latter category of codes shows that most utterances of CT in justifications of verdicts pertain to statutes and other normative acts, not to the constitution. This was measured by calculating the frequency of individual codes Jan Winczorek © Jan Winczorek. Informal Logic, Vol. 36, No. 3 (2016), pp. 371-394. 382 referring to the constitutional interpretation among all codes, and their share in the volume of the entire corpus of justifications (in the remaining part of this paper the variable is referred to as "relative volume"). It turned out that only about 25% of utterances of CT pertains to the constitution, measured both by relative volume and frequency of codes. This data also shows that the distribution of both volume and frequency was limited, which implies that CT's judgements are similar in this respect. Descriptives show that in the third quartile of judgements both frequency and volume come to approximately 0.3. This means that only in 25% of judgements the references to the constitution exceeded 30% of CT's entire communication. The same data still suggests that there exist both justifications in which CT makes no reference at all to the constitution and justifications where most of the text is devoted to it. To explain the circumstances in which the CT is more eager to refer to the constitution, a logistic analysis was conducted. The dependent variable was the volume of references to the constitution, and over a dozen factors describing the judgements were used as independent variables. After a number of attempts a regression was obtained whose coefficients, after being converted to odds rations, show the increase of likelihood that the CT refers to the constitution if certain conditions are met. Estimate Odds-ratio Pr(>|z|) (Intercept) -1.845 0.158 0.00 Verdict issued after (0) or before (1) 1997 0.609 1.838 0.00 Verdict signed K (1) 0.254 1.289 0.23 Verdict signed Sk (1) 0.428 1.535 0.06 Verdict signed P (1) 0.249 1.283 0.27 Verdict signed U (1) 0.173 1.189 0.48 Number of justices in the panel -0.019 0.981 0.23 Number of dissenting opinions 0.036 1.036 0.67 Entering of verdict into force postponed (1) -0.026 0.974 0.01 Proceedings discontinued (1) 0.027 1.028 0.74 Decision in favour of petitioner (1) 0.359 1.431 0.00 Table 3: Logistic analysis: dependent variable—volume of references to constitution Redundancy of Redundancy in Justifications of Verdicts © Jan Winczorek. Informal Logic, Vol. 36, No. 3 (2016), pp. 371-394. 383 The outcomes of this analysis support the conclusion that the volume of references to the constitution is influenced by the year of passing the verdict (the chances are 1.8 times greater after the year 1997 than before that date) and by the fact that it was issued as a result of a constitutional complaint (1.53 times greater). The analysis also reveals that the verdicts in favour of the petitioner are characterised by a greater volume of references to the constitution (1.43 times compared to unfavourable verdicts). This outcome suggests that in the constitutional interpretation an important role is played by the normative context, shaped by, among other things, the entering into force of the new constitution. This event provided new standards of constitutional adjudication which contributed to the increase in the volume of references to the constitution. Still a significant effect is likely owed to the contents of the motion filed with the CT (in particular it is plausible that constitutional complaints are drafted in a different way than other types of motions and on the average refer to a larger number standards of control). An important finding is also that the CT justifies its verdicts in a different way depending on the conclusion. This shows that the CT adjusts its strategy of justifying a verdict to the contents of the decision it had made. Still the most significant result of the study—one that demonstrates the potential of contents analysis in the studies of adjudication—is the distribution of frequency and volume of individual codes referring to constitutional interpretation. These measures show the relative popularity and role of individual types of argument in justifications of verdicts. The outcomes are summarized in Table 4. Type of argument Frequency Relative volume arguments of adjudication—argument of CT's own adjudication—further justified 0.13 0.19 arguments of adjudication—argument of CT's own adjudication—no further justification 0.12 0.10 doctrinal arguments 0.08 0.07 argument by a reference to the principle under interpretation 0.08 0.09 argument of limits of CT's authority 0.05 0.05 reference to the normative text 0.05 0.03 argument of nature of legal institution 0.04 0.05 argument by a reference to the principle concluding interpretation 0.04 0.05 argument of proportionality and weighting (principles, 0.04 0.05 Jan Winczorek © Jan Winczorek. Informal Logic, Vol. 36, No. 3 (2016), pp. 371-394. 384 Type of argument Frequency Relative volume values) systematic arguments - argument of legal system's structure 0.04 0.04 systematic arguments - other types of systematic argument 0.03 0.03 semantic arguments - argument of legal language 0.03 0.03 argument of ratio legis 0.03 0.02 argument of discretionary powers of an authority 0.02 0.02 argument of social consequences 0.02 0.02 reference to values and other non-legal norms 0.02 0.02 semantic arguments - other types of semantic argument 0.02 0.02 comparative arguments - internal comparative argument 0.02 0.02 systematic arguments - argument of methods of regulation and branches of law 0.02 0.01 systematic arguments - argument of construction of a legal act 0.01 0.01 classical legal reasoning (maxims) 0.01 0.01 argument of formal binding force of law 0.01 0.01 arguments of adjudication - argument of adjudication of Polish courts 0.01 0.01 semantic arguments - argument of autonomous meaning of legal terms 0.01 0.01 systematic arguments - argument of type of legal provision 0.01 0.01 semantic arguments - argument of provision being unclear 0.01 0.01 argument of notorious facts 0.01 0.01 arguments referring to sustainability of law or legal change 0.01 0.01 systematic arguments - argument of idealisations of legal system 0.01 0.00 reference to legislative works 0.01 0.00 argument of social change 0.01 0.01 comparative arguments - external comparative argument 0.00 0.00 reference to group interests and social groups 0.00 0.00 semantic arguments - argument of natural language 0.00 0.00 arguments of adjudication - argument of adjudication of ECHR 0.00 0.00 arguments of adjudication - argument of adjudication of ECJ - - Sum 1.00 1.00 Mean 0.03 0.03 Table 4: Frequency and relative volume of codes - codes before aggregation Redundancy of Redundancy in Justifications of Verdicts © Jan Winczorek. Informal Logic, Vol. 36, No. 3 (2016), pp. 371-394. 385 It shows that the most frequently used method of justifying its decisions by CT and the most significant in terms of volume, is making references to CT's own earlier decisions. This interpretive strategy is visible in approximately 25% of all units of text identified by researchers which refer to the constitution and about 30% of the volume of the corpus so defined. This means that the standard form of justifying verdicts in their constitutional aspect utilises the following argumentative pattern: "CT believes X because it had believed X in the past". The second most important reference made by CT is legal literature, and only the third most important type of argument is based on principles or values (arguably this type of interpretation is frequently just a modified semantic argument, based on a crude reference to the legal text) along with common textbook techniques of legal interpretation. The former includes mostly linguistic and systematic methods. The references to adjudication of other courts are very infrequent both when it comes to Polish and international courts. These patterns of interpretation can hardly be seen as multi-contextual. The distribution table reproduced here just does not support the view that the CT is eager to justify its constitutional decisions by using a number of different parallel arguments, nor do correlation analyses of different types of arguments used by CT, omitted for brevity. Since comprehensive presentation of the outcomes of a statistical analysis of frequency and volume of 37 individual codes is difficult, the codes were aggregated. Similar types of argument were merged into 12 broad groups. This simplification of complex data makes it possible to reproduce two-way tables that show the differences in frequency and relative volume of codes depending on select independent variables. Table 5 demonstrates that CT uses different methods of interpretation when it refers to the constitution and other legal acts. This contradicts the theory that legal interpretation is methodologically unified across the legal system. In the case of the constitution, self-citations are utilised much more frequently than in the case of regular legislation. As far as statutes are concerned, the most important interpretive techniques are textual—linguistic and systematic. In that case only a minor role is played by arguments which M. Matczak (2007) calls external. Jan Winczorek © Jan Winczorek. Informal Logic, Vol. 36, No. 3 (2016), pp. 371-394. 386 Arguments Constitution Statutes Other legal acts Frequency Relative volume Frequency Relative volume Frequency Relative volume n 2032 4635 836 of own authority 0.25 0.29 0.10 0.11 0.10 0.11 of values and prinicples of law 0.18 0.21 0.07 0.08 0.05 0.06 systemic 0.14 0.12 0.24 0.24 0.30 0.30 linguistic 0.12 0.09 0.21 0.20 0.25 0.27 doctrinal 0.08 0.07 0.07 0.06 0.02 0.01 functional 0.08 0.08 0.11 0.11 0.09 0.08 of distribution of power 0.07 0.07 0.05 0.04 0.05 0.04 teleological and refering to intentions 0.03 0.03 0.06 0.06 0.04 0.04 comparative 0.02 0.02 0.04 0.04 0.06 0.06 of authority of other courts 0.01 0.01 0.04 0.04 0.02 0.02 systematic 0.01 0.01 0.01 0.02 0.01 0.01 of interests 0.00 0.00 0.01 0.01 0.01 0.01 Table 5: Frequency and relative value and type of legal act - aggregated codes As in the case of frequency and volume of references to the constitution, also the characteristics of arguments can to some extent be explained by independent variables. The outcomes of such an inquiry are presented in Table 6. It shows—and so do other similar analyses not reproduced here—that the type of argumentation is influenced by the same factors as in the case of references to the constitution. These factors are: the date on which the judgement was passed, type of procedure (particularly, the constitutional complaint) and whether or not the verdict was in favour of the petitioner. This confirms that different types of justification of verdicts are used whenever CT refers to the constitution and regular legislation. At the same time this result suggests that the decision variable in justifying verdicts is the extent to which the CT refers to the constitution. If it decides to make such a reference, the methodology of justification changes. Redundancy of Redundancy in Justifications of Verdicts © Jan Winczorek. Informal Logic, Vol. 36, No. 3 (2016), pp. 371-394. 387 Arguments Relative volume Until 1997 After 1997 Verdict unfavourable Verdict favourable n 415 1617 1291 661 of own authority 0.25 0.31 0.25 0.39 of values and priniciples of law 0.21 0.2 0.2 0.22 systemic 0.16 0.11 0.13 0.09 linguistic 0.11 0.09 0.09 0.08 of distribution of power 0.1 0.07 0.1 0.03 doctrinal 0.06 0.07 0.08 0.05 functional 0.05 0.08 0.09 0.04 teleological and referring to intentions 0.03 0.03 0.02 0.03 comparative 0.02 0.02 0.02 0.03 systematic 0.01 0.01 0.01 0.02 of interests 0 0 0 0 of authority of other courts 0 0.01 0.01 0.01 Table 6: Constitutional interpretation until 1997 and after 1997, constitutional interpretation in judgements favourable and unfavourable for the petitioner. Differences in argumentation related to favourability of the verdict are particularly interesting. As demonstrated in Table 7, in case of verdicts that could be described as coherent with petitioner's intent, the references to CT's own authority (i.e. to the fact that it has already uttered a particular view), come to almost 40% of volume of its statements about constitution. This also strongly supports the view that the justifications of CT's verdicts are relatively coherent methodologically. Jan Winczorek © Jan Winczorek. Informal Logic, Vol. 36, No. 3 (2016), pp. 371-394. 388 Arguments linked to outcome unfavourable for the petitioner "Neutral" arguments Arguments linked to outcome favourable for the petitioner doctrinal (0.078— 0.053) of limits of CT's authority (0.065— 0.024) of nature of legal institution (0.057— 0.027) of notorious facts (0.0053—0.0028) of legal system's structure (0.036—0.024) of discretionary powers of an authority (0.032— 0.009) of social consequences (0.024—0.012) other types of semantic a. (0.023—0.007) classical legal maxims (0.014—0.005) of formal binding force of law (0.013—0.009) of type of legal provision (0.012 - 0.002) sustainability of law or legal change (0.009— 0.005) of social change (0.005—0.0006) group interests and social groups (0.002— 0.0003) reference to the principle under interpretation (0.089— 0.095) reference to the principle concluding interpretation (0.046— 0.047) reference to the normative text (0.03— 0.03) other types of semantic a. (0.023—0.007) argument of legal language (0.026— 0.023) reference to values and other non-legal norms (0.02—0.02) of adjudication of Polish courts (0.01— 0.01) of methods of regulation and branches of law (0.013—0.01) of idealisations of the legal system (0.004— 0.005) external comparative a. (0.004—0.003) of adjudication of ECHR (0.0—0.003) reference to legislative works (0.002—0.002) of legal language (0.0— 0.002) of adjudication of Polish courts (0.0001— 0.0) of CT's own adjudication—further justified (0.15—0.28) of CT's own adjudication—no further justification (0.10— 0.12) of proportionality and weighting (0.046— 0.053) of ratio legis (0.013— 0.03) internal comparative a. (0.016—0.025) of autonomous meaning of legal terms (0.006— 0.015) of construction of a legal act (0.01—0.02) Table 7: Arguments favourable and unfavourable for the petitioner For that reason, the differences in the type of argumentation in verdicts favourable and unfavourable for the petitioner were further analysed using disaggregated codes. The outcomes obtained, as presented in Table 7, show that the volume of Redundancy of Redundancy in Justifications of Verdicts © Jan Winczorek. Informal Logic, Vol. 36, No. 3 (2016), pp. 371-394. 389 arguments of a particular type depend on characteristics of the verdict. The first number in brackets signifies the relative volume of an argument in the unfavourable verdicts and the second number in the favourable verdicts. Clearly, the verdicts in favour of the petitioner are much more broadly justified by references to CT's own adjudication. This is a circumstance once dependent from CT's own decision, but impossible to change or negotiate at the time of passing the verdict. The unfavourable verdicts more often use references to the legal doctrine, the formally limited competences of CT, and the nature of the legal institutions. All these circumstances by definition lay outside of the scope of power of CT. 5. Discussion The outcomes of the study are in principle coherent with earlier findings of some of the researchers involved (Stawecki, Staśkiewicz, and Winczorek 2008). They show that Polish CT prefers doxa-type arguments. The arguments revolve around statements which cannot be questioned because of quasi- objective factors. Challenging them might be outside of the scope of discretion of the adjudicative body and the petitioner questioning it, stand against the dominant views of constitutional lawyers, or they might simply be logically infallible. Such quasi-objective arguments include references to earlier (and hence impossible to change) verdicts, linguistic and systematic (quasi-objective) interpretive schemes, and views of the legal scholars (a circumstance independent from CT itself). This finding appears as coherent with a view that discretionary powers of adjudicative bodies should be limited, and the perception that political views and direct (not mediated by principles of law) references to values are unacceptable in the courts' adjudication. In this way some sort of objectivisation of CT's own decision is achieved, which seems coherent with the ideology of bound legal decision-making (Wróblewski 1992:265ff), dominant in Poland. Still, the study shows that the readiness of CT to make references to the constitution depends on the contents of decisions which need to be justified. The same holds for methodological types of arguments used to that end. This said, independent of the situation, the predominant methodology of constitutional interpretation used by CT cannot be seen as multi- contextual, combining different types of argument, epistemologically and methodologically divergent but resonating with different audiences. In fact, the aims of this Jan Winczorek © Jan Winczorek. Informal Logic, Vol. 36, No. 3 (2016), pp. 371-394. 390 methodology are not persuasive, but demonstrative. It serves the purpose of making it apparent that a particular decision is necessary and that it just cannot be different rather than presenting persuasive reasoning which can be evaluated and opposed with a counter-argument. If the outcomes of the study reported here are not fundamentally flawed, it follows that they justify a conclusion that the predominant methodology of justifying verdicts by CT limits the scope of references it makes in its opinions. Out of many possible types of argument which could be used and which refer to different elements of the legal system— jurisprudence, theory, values and institutions—it selects just those which refer either to CT itself or to legal text. One possible theoretical interpretation of this finding makes use of the concept of redundancy. While in excess of references, which according to systems theory is a defining feature of any social system, including the legal system, it is not reflected by the CT in its verdicts. Redundancy of the legal system is thus implicitly considered redundant—irrelevant for the proper functioning and efficacy of CT's adjudication. This shows that Polish CT, despite producing long and detailed justifications in fact fails to deliver wide-ranging arguments for its decisions. In some sense, such an interpretive strategy is conservative but safe. As it were, systems theory suggests that social systems, including the legal system are necessarily and naturally redundant. They are either characterised by excess of relationships between their elements or they cease to exist. Whether a CT is interested in this or not, the system has exactly such a characteristics. It is also a feature of constitutional adjudication, which is possible, understandable and effective only when it has an established position in the network of elements constituting the legal system. This in turn means that there exists a large probability that the justifications which today appear as intellectually unquestionable will be so in the predictable future. This stems from the fact that they are produced to that effect by the network of relations in which they are nested. Even when some of the elements of the system are amended—for instance the mechanism of appointing justices is changed, the readiness of regular courts to cite CT's decision is changed, and the parliament's willingness to execute them is altered—the entire network will accommodate the change. Yet on the other hand, treating redundancy as redundant is risky. The legal system undergoes evolutionary processes, as a consequence of which its seemingly constant elements change. This happens exactly because they are observed as constant, and translates into a chance that today's adequate and effective Redundancy of Redundancy in Justifications of Verdicts © Jan Winczorek. Informal Logic, Vol. 36, No. 3 (2016), pp. 371-394. 391 description of the legal system will not be such tomorrow. In the context relevant here the quest for independence of adjudication from political conditions, as expressed by the frequent references to internal arguments, might in reality be perverse. Such references facilitate deploying instrumental strategies against constitutional adjudication because they make it easier to estimate reactions of a constitutional court to the passing of particular legislation and other political decisions. Whenever a number of elements of the system is changed— which might be a contingent phenomenon or an outcome of coordinated action, for instance politically motivated—a redundant perspective on the system's redundancy might thus have consequences surprising the constitutional court itself. In other words, by using the strategy of redundancy of redundancy the CT might become a tool of political power even if prima facie, and in its own intent, it is completely independent from it3. A question in its own right is why does CT treat redundancy as redundant. A preliminary answer has been suggested when the general tendency of Polish justice system to rely on the ideology of bound decision-making was mentioned. It is a relatively constant feature of the Polish legal culture and offers good opportunities for explanation. Yet giving a full answer to that question exceeds beyond the limits of this text and perhaps requires a separate broader empirical study, such that it would look at the history of adjudicative strategies of Polish courts in a longue-durée perspective. 3 The author of this text has the rare priviledge of reviewing his own ideas in the light of political developments which took place long after these ideas had been coined and the empirical study based on them had been performed. The political developments of 2015 and 2016 in Poland—which according to many commentators mark a deviation from the principles of rule of law, and whose principal element is a sweeping reform of the CT effectively disabling the institution, yield a question on the sustainability of the entire institution and the conditions of possibility of reverting from rule of law. It appears, that the model developed in this paper explains the development quite well, suggesting that the doxa-oriented interpretaion agenda, based on ideology of bound-decision making can be turned against the CT—which has actually happened. For reference on developments see CDL-Opinion on amendments to the Act of 25 June 2015 on the Constitutional Tribunal of Poland, adopted by the Venice Commission at its 106th Plenary Session (Venice, 11-12 March 2016), AD(2016)001-e. Jan Winczorek © Jan Winczorek. Informal Logic, Vol. 36, No. 3 (2016), pp. 371-394. 392 References Baghai, Katayoun. 2015. Social Systems Theory and Judicial Review: Taking Jurisprudence Seriously. Aldershot. Ashgate. Banakar, Reza. 2015. Normativity in Legal Sociology. Methodological Reflections on Law and Regulation in Late Modernity. New York. Springer. Brace, Paul, Laura Langer, and Melinda Gann Hall. 2000. Measuring the Preferences of State Supreme Court Judges. Journal of Politics 62(2): 387—413. Dyevre, Arthur. 2010. Unifying the field of comparative judicial politics: towards a general theory of judicial behaviour. European Political Science Review, 2(02): 297-327. Garfinkel, Alan 1981, Forms of explanation. New Haven. Yale University Press. Hall, Mark A., and Ronald F. Wright. 2008. Systematic Content Analysis of Judicial Opinions. California Law Review 96 (1): 63—122. Keele, Denise M., Robert W. Malmsheimer, Donald W. Floyd, and Lianjun Zhang. 2009. An Analysis of Ideological Effects in Published Versus Unpublished Judicial Opinions. Journal of Empirical Legal Studies 6 (1): 213—39. Krippendorff, Klaus. 2004. Content Analysis An Introduction to Its Methodology. Thousand Oaks: Sage. Łętowska, Ewa. 2006. Promieniowanie orzecznictwa Trybunału Konstytucyjnego na poszczególne gałęzie prawa [The Radiation of Constitution Tribunal's Jurisprudence onto Other Branches of Law]. In Księga XX-lecia orzecznictwa Trybunału Konstytucyjnego, ed. Marek Zubik. : 353-377. Warszawa: Trybunał Konstytucyjny. Lautmann, Rudiger. 2011/1972. Justiz-die stille Gewalt: teilnehmende Beobachtung und entscheidungssoziologische Analyse. Frankfurt: Springer-Verlag Limbach, Jutta. 1999. „Promieniowanie” konstytucji na prawo prywatne. [The Radiation of Constitutions on Private Law]. Kwartalnik Prawa Prywatnego 3: 405-414. Luhmann, Niklas. 1985. Einige Probleme mit ‘reflexivem Recht.’ Zeitschrift Fuer Rechtssoziologie 6 (1): 1—18. ———. 1993. Das Recht der Gesellschaft. Frankfurt am Main: Suhrkamp. Matczak, Marcin. 2007. Summa iniuria: o błędzie formalizmu w stosowaniu prawa. [Summa iniuria. On the Formalist Error in Applying the Law] Warszawa: Scholar. Neves, Marcelo (2001). From the Autopoiesis to the Allopoiesis of Law. Journal of Law and Society, 28(2): 242-264. Redundancy of Redundancy in Justifications of Verdicts © Jan Winczorek. Informal Logic, Vol. 36, No. 3 (2016), pp. 371-394. 393 Segal, Jeffrey A., and Albert D. Cover. 1989. Ideological Values and the Votes of U.S. Supreme Court Justices. American Political Science Review 83(2): 557—65 Skyttner, Lars. 2006. General Systems Theory: Problems, Perspectives, Practice. Singapore: World Scientific. Sisk, Gregory C., Michael Heise, and Andrew P. Morriss. 1998. Charting the influences on the judicial mind: An empirical study of judicial reasoning. New York University Law Review, 73(5): 1377-1500. Stawecki, Tomasz, Wiesław Staśkiewicz, and Jan Winczorek. 2008. Between Polycentrism and Fragmentation The Impact of Constitutional Tribunal Rulings on the Polish Legal Order. Warszawa: Ernst&Young. Stawecki, Tomasz, and Jan Winczorek. 2014. Wykładnia konstytucji: Inspiracje, teorie, argumenty. [Constitutional Interpretation: Inspirations, Theories, Arguments]. Warszawa: WoltersKluwer. Streeter, Calvin, 1991. Redundancy of Social Systems: Implications for Warning and Evacuation Planning, International Journal of Mass Emergencies and Disasters. 9(2):167-182. Teubner Gunther, 1983. Substantive and Reflexive Elements in Modern Law, Law & Society Review, 17(2): 239-286. Teubner, Gunther. 2004. Societal Constitutionalism: Alternatives to State-Centered Constitutional Theory? Storrs Lecture at Yale Law School, Oct. 7, 2003, accessed 10 May 2016, http://www.jura.uni- frankfurt.de/ifawzl/teubner/dokumente/societal_constitutiona lism.pdf Teubner, Gunther. 2012. Constitutional fragments: societal constitutionalism and globalization. Oxford: Oxford University Press Thornhill, Chris. 2010. Niklas Luhmann and the sociology of the constitution. Journal of Classical Sociology, 10(4): 315- 337. Weinshall-Margel, Keren. 2011. Attitudinal and Neo- Institutional Models of Supreme Court Decision Making: An Empirical and Comparative Perspective from Israel. Journal of Empirical Legal Studies 8 (3): 556—86. Winczorek, Jan. 2009. Zniknięcie dwunastego wielbłąda. O socjologicznej teorii prawa Nikasa Luhmanna. [The evanescence of the twelfth camel. On sociological theory of law of Niklas Luhmann] Warszawa: Liber. Jan Winczorek © Jan Winczorek. Informal Logic, Vol. 36, No. 3 (2016), pp. 371-394. 394 ———. 2014a. Wykorzystanie oprogramowania R i RQDA w jakościowo-ilościowej analizie treści orzeczeń Trybunału Konstytucyjnego. [Using R and RQDA in a qualitative- quantitative content analysis of judgments of Constitutional Tribunal]. Przegląd Socjologii Jakościowej 2: 138—61. ———. 2014b. Wzorce wykładni konstytucji w świetle analizy treści uzasadnień orzeczeń Trybunału Konstytucyjnego. [Patterns of constitutional interpretation in the light of a content analysis of judgments of Constitutional Tribunal In Wykładnia konstytucji: Inspiracje, teorie, argumenty, ed. Tomasz Stawecki and Jan Winczorek. 387—460. Warszawa: WoltersKluwer. Wróblewski, Jerzy. 1992. The judicial application of law. Dordrecht:Kluwer. i The author would like to acknowledge the participation of the following persons in the creation of the empirical material presented in the study: Tomasz Stawecki, Wiesław Staśkiewicz, Marcin Matczak, Krzysztof Kaleta, and Marcin Romanowicz. They all participated as research team members in a project led by T. Stawecki in the years 2010-2012 at University of Warsaw, which included an empirical study designed by the present author. The author would also like to thank anonymous reviewers for their useful comments on the earlier version of this paper. All errors and omissions are the author's own, and the interpretation of data presented in this paper might be or might not be consistent with other research team members' interpretations.