1.
Introduction ^
[1]
Near the beginning of this century, the judiciary in Brazil was impacted by technology and mainly computation: we are at the beginning of the transformation of the physical analog in a digital electronic process and which is heavily operated through the Internet. At the forefront of this process in Brazil are higher courts, especially the Supreme Court and the Superior Court, which have converted their paper procedure to the electronic format. Within the judiciary system and the CNJ – National Council of Justice, there has been an implementation of reform policies amongst the state and federal courts, so that they follow the example of the superior courts and promote measures to reform the procedure towards electronic process moving towards E-Justice or electronic justice.
[2]
In parallel to the process of converting the judicial process to an electronic format, another major reform of the judicial system started operating in Brazil, which is the implementation of a system for the collection and statistical analysis of the courts. Through various legal measures, there is a process of implementing a culture of openness to information, so that, through the knowledge of the data of the judiciary and especially its analysis through certain indices, society as a whole, and public entities that deal with strategic management of the judicial system, can identify their problems and plan quick solutions. In addition to e-Justice, there is currently a system of judicial fairness or quantitative justice, also known as Q-Justice.
[3]
The analysis of the performance of courts and tribunals in Brazil has caused some surprise when not creating a feeling of rejection. Many causes can be identified for this negative impact, however, we can point out two main reasons:
- The lack of access to the data regarding the judicial system: in the two decades following the Constitution of 1988, there was hardly a system of data collection and information relating to the judiciary system in Brazil as a whole. There was fragmented and isolated information, but without the preparation of indexes, it was not possible to radiograph the judicial system in statistical terms. From the creation of the National Council of Justice and the implementation of its policies, gradually the data relating to the judiciary began to appear publicly, causing the most varied reactions. The most famous example was the impact provoked by a number of researches called «Top 100 litigants,» in which the federal government and banks emerged as the largest plaintiffs in Brazilian courts, and continue in the same position according to the latest report from the 2011 data.
- The lack of access to the evaluation and control systems of modern Western democracies: in continuation of the previous process, the National Council of Justice also has established itself as a body to implement strategic policies to modernize the judiciary. In this process of modernization, agencies and internal teams to CNJ were created, with the aim of producing knowledge and specific techniques that ensure the means for the implementation of new management policies and control. Consequently, we sought in judicial systems of other countries, like the United States, Mexico and the European Union, for elements and experiences that could also be implemented in Brazil. In a sense, we did not have the culture of data collection, measurement, evaluation and control of the judicial system in Brazil.
[4]
The analysis of judicial systems and their workings are not something new outside of Brazil. The United States and European Union constantly produce statistical reports for analyzing their systems, as well as for evaluation.1
[5]
In this paper, we show some basic data about the Brazilian Judicial System, describe the main systems of data collection for the Brazilian Judiciary, indicate some methodological observations on the availability of public data and statistical research with judicial data, and point out the future prospects of judicial fairness with respect to new computing technologies. The main objective of this paper is to describe, to an international audience, the recent Brazilian experience of judicial reform with the adoption of a strong policy of judicial computerization and creation of a system of judicial statistics and judicial evaluation. The authors do not intend to propose or formulate new theoretical concepts or techniques of legal informatics.
2.
Brazil: some basic judicial data and expenditures of the Brazilian judiciary ^
[6]
In 2012, the annual GDP (Gross Domestic Product) growth was 4.2%. The GDP per capita was US$ 12,670 and the population was 196.5 million2. The GDP in total in 2012 was US$ 2.2 trillion (equivalent to R$ 4.402.537,00). The average income in 2012 per capita per month was 934,5 U.S. dollars (R$ 1.869,05).
[7]
The inputs and endowments of Brazilian judiciary3 power were 20.5 billion U.S. Dollars during 2010, which was equivalent to 1.12% of the national GDP and 2% of the expenses of the Federal Union and the States and US$ 106 dollar (R$ 212,37) per year per inhabitant.
2.1.
Humans resources ^
[8]
The workforce of the Judiciary system (Federal, State and Labor Courts) consisted, at the end of 2010, of approximately 339,000 employees, of which 16,804 are magistrates (judges) and 321,963 are civil servants. The total number of positions comprises of officially affiliated employees (exception made to employees directly allocated from other public institutions), employees requested from other institutions, outsourced workers, interns and employees commissioned without public service official affiliation.
2.2.
Magistrates and work force per hundred thousand inhabitants ^
[9]
The Judiciary (in the three investigated branches) has, on average of, 9 judges per group of one hundred thousand inhabitants. The highest ratio is in State Courts (6 judges per 100,000 inhabitants), and the lowest one in Federal Courts (1 magistrate per 100,000 inhabitants).
2.3.1.
General case flow data ^
[10]
During 2010, 24.2 million lawsuits were filed in the three Judiciary branches (17.7 million in State Courts, 3.2 million in Federal Courts and 3.3 million in Labor Courts), and at the end of the year there were 59.2 million pending lawsuits. Thus, the case flow of the Brazilian Judiciary in 2010 amounted to 83.4 million lawsuits. 22.2 million decisions were issued, divided as outlined below: 15.8 million in State Courts (representing 71% of the total), 2.9 million in Federal Courts and 3.5 million in Labor Courts.
2.3.2.
Incoming cases per one hundred thousand inhabitants ^
[11]
There were, at the end of 2010, 11.536 new cases for each group of one hundred thousand inhabitants in the three Judiciary branches. The most demanded branch was that of State Courts, with 8.641 new cases for every group of 100,000 inhabitants.
2.3.3.
1st Instance Litigations and Small claims courts ^
[12]
In 1st Instance Courts approximately 20.5 million lawsuits were filed in 2010, 73% of which (in average) corresponded to pre-trial lawsuits, and the other 27% were related to enforcement-stage lawsuits. There were 55.7 million cases pending resolution at the end of 2010, representing an increase of 2% over the previous year.
2.3.4.
Incoming cases per magistrate and civil servants working in the judiciary area per magistrate in 1st Instance Courts and in Small claims courts ^
[13]
Brazilian 1st instance courts (in its three branches) received, on average, approximately 1,290 new cases for every magistrate in 2010.
2.3.5.
Caseload and backlog rates in First Instance and Small Claims Courts ^
[14]
Caseload is the indicator commonly utilized to measure the amount of lawsuits magistrates have to rule on, on average, every year. Each 1st Instance magistrate in the Brazilian Judiciary had, on average, 5,423 lawsuits that could be ruled on in 2010.
3.
Eletronic lawsuit filing ^
[15]
The Justice in Numbers report also brought data on the rates of electronic lawsuit filing, with a view to investigating the level of computer technology adherence of the Brazilian Judiciary and the adoption of new technologies into lawsuit processing methodologies. The indicator is obtained from the division of the number of new electronic lawsuits by the number of new lawsuits in the analyzed instances of Justice (2nd Instance, 1st Instance, Appellate Courts and Small claims courts). It was verified that Federal Courts have continued to invest in the implementation of virtual lawsuits in their courts, with lawsuit virtualization indices ranging from 43% (Federal Court of the 3rd Region) to 82% (Federal Court of the 5th Region). It should be emphasized in particular that the Regional Federal Court of the 1st Region reached a 64% virtualization rate for new lawsuits in the 1st instance courts. On the other hand, also worth mentioning is the low response rate for this indicator in the Labor Courts, which may be a sign of difficulties being found in the adoption of electronic lawsuits in the Labor branch.
[16]
Expenditures on human resources represented 89.6% of the total budget for the three branches of the Judiciary (State, Federal and Labor), a percentage lower than the previous year, which was 90.8%. This decrease is particularly timely, given that since 2006 spending under this heading has always been found at levels above 90%, a fact which undermines important investments for the modernization of the courts and the improvement of its working structure.
[17]
The computerization of the judiciary in Brazil is being done through the collaborative construction of a complete electronic system of judicial proceedings, the Electronic Judicial Process (PJe in Portuguese). This system was developed at the federal level and is being adopted by Federal and State Courts through the signing agreements. In the next chapter we describe the main characteristics of the PJe.
4.
The PJe (Electronic Judicial Process) and the computerization of courts in Brazil ^
[18]
The main and most extensive project of computerization of the judicial courts in Brazil is called PJe, Electronic Judicial Process. It is a system developed by CNJ – National Council of Justice, a body created in 2005 and it has, among other functions, to establish strategic, management and planning policies for the Brazilian judiciary as a whole.
[19]
The formulation of the PJe began in 2008 under Federal Courts, and has been taken over permanently by CNJ in 2010. The purpose of this system is to provide the Brazilian Judiciary System a full system of judicial proceedings, since the entry of a case into the judicial system to its archive, ensuring its integration with the old system of proceedings. PJe is quite comprehensive and involves civil and criminal cases from all specialties of the Brazilian judiciary, the Labor Courts, Federal, State or Military Courts. The system is also multi-user, intended to be accessible to all actors involved in judicial proceedings, i.e., the magistrates, the employees of the judiciary, lawyers, prosecutors and parties.
[20]
In 2010, 54 courts and tribunals were already participating in the formulation and development of the PJe. Those same courts have adopted the PJe, through the signature of agreements between the involved courts and CNJ.
[21]
Until the year 2013 three versions of the system have been developed, 1.0, 1.2 and 2.0.
[22]
The PJe has several interesting features that deserve to be mentioned:
[23]
With respect to direct assistance to the judge/lawyer: The PJe allows documents, such as petitions, to be produced internally within the system and not externally to the system. It has a built-in text editor that enables this feature. Until now there is not one electronic database of jurisprudence incorporated into the system. Currently, each court in the Brazilian system offers access to its jurisprudential bases on its website. The PJe allows uploading of files and electronic documents and communication between parts of the process through the internet.
[24]
With respect to the administration and management: the PJe electronically manages the procedure of cases and unifies procedural tables, so that in a previous situation, even one procedural act might have different names in various courts, in the new system there is no divergence of denomination for the same acts. In this new system there is also an automatic replication of information process management: the reports sent to the CNJ from different courts that make up the annual report «Justice in Numbers», will not need to be filled manually. With the integration of systems, the data collection becomes automatic. Reports can also be automatically generated within the court, for better internal management of the judiciary body. Another feature of the PJe is its flexibility: it allows administrators to adapt the system and the permissions for users, so the judicial proceedings may be more rigid or more flexible, as the desired configuration by the system administrator. Thus the IT team of Courts need not be triggered in every system change.
[25]
Administration between the court and the parties: the PJe has a feature that, in addition to allowing the procedural actors to write a petition within the system, allows models to be archived in pre-formatted documents, so that, once the variables are inserted into the fields, the document is automatically generated.
[26]
In a broad sense, the PJe intends to integrate all actors in the process: judges, judicial employees, parties, lawyers, prosecutors, administrative and supervisory body.
[27]
Together with the computerization of the courts, the CNJ made in recent years another major reform in the judiciary, through the implementation of a system of judicial statistics and an evaluation of various aspects of performance of the judicial system. In the next chapter we describe the Brazilian system of judicial statistics and point out some methods that should be adopted.
5.
The actual situation of Q-Justice in Brazil ^
[28]
Actually, within the scope of the CNJ and the Supreme Court, the main data collection systems of the Judiciary in operation are:4
- Justice in Numbers
- The Supreme Court in Numbers.
5.1.
CNJ ^
[29]
The CNJ system provided statistical information on the judiciary system for the first time in 2003 and 2004, initially developed from the previous system, the National Judiciary Data Bank (BNDPJ). In 2005, Statistical Judiciary System was created (SIESPJ), through Resolution no. 4, on August 16, 2005, from the office of the president of the CNJ. Resolution no. 15 of April 20, 2006 regulated and set contents and formulas of the system. Resolution no. 15 was repealed and the normative basis of the system is currently established by Resolution no. 76 of May 12, 2009.
[30]
Basically, SIESPJ has the basic statistical indicators divided into the following categories:
I – Inputs, allocations and utilization levels:
a) Revenue and expenses;
b) Structure.
II – Litigation:
a) Workload;
b) Rate of congestion;
c) The ability to appeal and retirement decisions.
III – Access to Justice;
IV – Profile of Demands.
5.2.
The Supreme Court in Numbers project ^
[31]
The Getulio Vargas Foundation, through the Law School of Rio de Janeiro and with the support of the School of Applied Mathematics, recently launched the Supreme Court in Numbers project. The goal of the project is to perform quantitative analysis on the role of the Supreme Court, based on quantitative and statistical discussions about the nature, function and impact of the actions of the Supreme Court in Brazilian society.
[32]
The project aims to work from the database of the Supreme Court, composed of more than 1.2 million cases – 1,132,850 which have already been judged and 89,252 still active, almost 14 million in motion, 240,000 lawyers, and 1 million plaintiffs and defendants and more than 370,000 decisions from 1988 to the present day.
[33]
The first project report has been produced and has already been made available to the public.5
[34]
The methodology of the project is composed of a more widespread qualitative analysis, conduct a quantitative analysis and perform statistics on the nature, role and impact of the actions of the Supreme Court in Brazilian democracy. Since the Brazilian High Courts judge hundreds of thousands of cases per year, an amount much higher than the courts of any other major Western democracy, the object of the study of the project will be the vast databases of the Brazilian courts. As these databases are not available or are often incomplete and have much redundant, incomplete or inconsistent information, the project also aims to develop and employ new computational techniques that allow for the analysis.
6.
Some methodological bases for Q-Justice in Brazil ^
[35]
Through a brief examination of the state of the Q-Justice in Brazil, it is possible to conclude that the statistical system of the Brazilian judiciary is in its initial steps, almost completing a decade of existence. For a Q-Justice to be effectively transparent, it is necessary that the databases of the judiciary are widely publicized in order to allow researchers from various fields, such as lawyers, sociologists and political scientists, to submit the mass of data to comparative analysis and where conclusions can be freely established.
[36]
Both the site of the CNJ and STF – Supreme Federal Court currently provide data in the format of reports where the information has been analyzed and purified by the internal team of the organ. For a Q-Justice effectively open and democratic, some recommendations on methodological issues can be highlighted:
a. There needs to be a culture of transparency and free availability of data.
[37]
The recent Brazilian Law on Access to Information (Law No. 12,527 of November 18, 2011), in force since May 2012, already represents a significant breakthrough in this regard. The judicial system cannot be a closed system, it should be a system opened to society and especially to public consultation and scientific research. Indirectly, the major impediment to the realization of transparency is the technical interface. The major question, in current terms, may be well placed: how civil society can have access to the judicial system using modern information technologies? On the other hand, how the Courts may make use of information technologies to provide civil society with their relevant information? In this sense, the next item is of paramount importance.
b. The data provided by the judicial system should be standardized in the usual format database.
[38]
Courts should disclose their information in open formats, accessible and readable by computer systems. Unfortunately, the vast majority of reports are published in PDF format, which is usually a closed document and of difficult issue, as it needs to be converted to another format. It would be recommended that the formats of the reports were available in the usual extensions database as XML, JSON or CSV. A lot of the time of the social scientist, who focuses on the data of the Courts, is spent on manual collection of research information in spreadsheets, where it could be operated automatically by a computer program.
c. It would be advisable to standardize the format of the data supply for the entire judiciary.
[39]
For the information to be analyzed by a computer program, any character is significantly different. In the format of a number in a report, for example, the comma use is quite different from the point use. In certain cases, a report uses the same number to different formats, such as 3,99 and 3.99. Without human intervention, a computer has no way of interpreting the two formats as equal. The same observation holds for counting spaces and the use of uppercase and lowercase letters.
[40]
Once the mass of data is available, the advantage of the use of computer programs to analyze the data and allows for the creation of applications and viewing of the data. This does not mean that the analysis is automatic: it takes the researcher time to select which is the relevant data for a particular analysis in advance. Currently, Computer Science has developed automated techniques of data analysis, like Data Mining and linked Data6.
7.
Conclusion: remarks about E-Justice and Q-Justice in Brazil ^
[41]
Once the data mass of the judicial system is made available, it must be possible to extract qualitative information on the data, and especially if you can get knowledge of the correlations between them. Usually the mechanisms of data analysis used in empirical research on law in Brazil are intuitive and based on experience and sensitivity of the researcher. However, as is widely known, the increase in the number of disputes tends to grow exponentially in the coming years by various factors, such as increased income of the population and the expansion of the list of rights. Therefore, we have an increasing number of legal cases and, at the same time, an increase the complexity of the court system. Certainly the experience and sensitivity of the researcher in legal research is empirical and does not cease to be important, however, for the development of more refined analysis and for dealing with a massive amount of data, manual research finds its limits, and the researcher will need to use advanced techniques for computational analysis of the data. It is in this context that Data Mining can be used in empirical legal research.
[42]
Currently, with increasing data traffic from computers to mobile phones and devices such as tablets, Data Mining is increasingly employed as a technique for automatic analysis of information and knowledge generation, as the term denotes, it is to mine the data.
[43]
The CNJ, in recent reports mentioned earlier in this article, has produced knowledge generated from the judicial statistics. An important finding was the identification of the unit cost of the process in each state systems of justice of the Brazilian federation. Besides these, many other data can be extracted through various techniques of data mining.
[44]
With these techniques, a new field of analysis is opened for empirical research in the legal field. A recent example of applications of data mining techniques can be found in the project The Supreme Court in Numbers.7 The very CNJ, in the National Committee for the Management of Judicial Information Technology, has an internal Business Intelligence, such as the use of techniques similar to Data Mining.
[45]
Another recent breakthrough in computing is cloud computing. The Electronic Judical Process can also develop through its adoption. Many tech analysts predict that the next generation of computers and the internet is based on the «cloud». The term cloud computing refers to the centralization of certain functions such as storage, memory, processing and software on central servers, which communicate with the user via the internet.
[46]
One of the major obstacles in the implementation of an Electronic Process throughout the judicial system in Brazil is the problem of interoperability. In recent years, virtually every branch of the judiciary created and deployed its own system of electronic process, and the great difficulty for end users and advocates is having to deal with numerous systems, each with its own peculiarities, which do not communicate with each other.
[47]
Cloud computing may be a solution to the problem of interoperability of computerized procedures. With its adoption, there would be no incompatibility issues, for example, between different courts. Software and hardware of the personal computer user have practically become secondary; its only function becomes providing access to the Internet and the system itself. Another advantage relates to the software update, which can be done automatically and without user intervention. There is also a decrease in costs, as personal computers do not require individual maintenance software, the user does not necessarily need to pay a license to use the operating system.
[48]
From the viewpoint of Q-law, the advantages are numerous. A Court may be fully connected, and the system of data collection may become automatic, because there will be no division among computerized procedural systems. Some researchers of the U.S. judicial system coined the term Smart Courts to the Courts that employ the techniques of cloud computing.8 In addition to data collection being automatic, the courts may rely on applications that can provide charts and pivot tables, updated in real time, for the diagnosis and measurement of its functioning.9
[49]
Throughout this panorama, the question necessarily arises from the control of information. Today, large companies like Google and Facebook are facing proceedings in regards to the use of the personal data of its users. There is no denying that relations of power and hierarchy permeate decisions on the adoption of these new technologies and systems and how a system of an organ should (or should not) be linked to another.
[50]
The purpose of this study was, first, to expose to an international audience of researchers, whether from the fields of law or social sciences in general, to the current system of data collection and preparation of indexes of the judicial system in Brazil. Gradually, with the current system and the establishment of a historical series increasing thematic and annual reports, legal databases tend to grow and allow a more refined analysis of the functioning of the judicial system in Brazil. The current system is also in a permanent process of modernization, toward an automatic collection of data and standardization of information on procedures. In the near future, we will have legal databases with more consistent information. And finally, Data Mining and Cloud Computing may or may not be adopted in the Electronic Judicial system in Brazil, a new perspective on Q-Justice may be offered, with massive production of data, which however must always be subjected to a rigorous judgment on the feasibility of its disclosure, and consequently, on what use the actors involved in the Brazilian judicial system will do with it.
8.
References ^
[1] Coelho, Flávio Codeço/Souza, Renato Rocha/Cerdeira, Pablo de Camargo. Talk Information Mining and Visualization of a Large Volume of Legal Texts. EuroSciPy – Annual European Conference for Scientists using Python. Paris, 25 a 28 de agosto de 2011. Abstract available at http://www.euroscipy.org/talk/4182
[2] European judicial systems Edition 2010 (data 2008): Efficiency and quality of justice – European Commission for the Efficiency of Justice (CEPEJ). French edition: Les systèmes européens judiciaires. Edition 2010. ISBN 978-92-6986-0 Available at http://www.coe.int/t/dghl/cooperation/cepej/default_en.asp
[3] Estudos sobre os sistemas de coleta de dados: CNJ / Conselho Nacional de Justiça. Série CNJ Acadêmico, n. 4. Departamento de Pesquisas Judiciárias. Brasília: CNJ, agosto 2010. 51 p.
[4] Keilitz, Ingo. Smart Courts: Performance Dashboards and Business Intelligence. In: Future Trends in State Courts 2010. C. Flango, A. McDowell, C. Campbell, and N. Kauder. Williamsburg, VA: NCSC – National Center for State Courts, 2010. p. 72-79. ISBN: 0-89656-274-3. Available at http://www.ncsconline.org/d_kis/trends/index.html
Cesar Antonio Serbena
Professor of Philosophy of Law, Law School, Federal University of Paraná-Brazil
Santos Andrade Square 50, 80020-300 Curitiba, BR
cserbena@gmail.com; http://www.ejustica.ufpr.br
Maurício Dalri Timm do Valle
Professor of Tax Law at Unicuritiba Faculty-Brazil, PhD candidate in Law at Federal University of Paraná-Brazil
Santos Andrade Square 50, 80020-300 Curitiba, BR
mauricio_do_valle@hotmail.com; http://www.ejustica.ufpr.br
- 1 Areader interested in more details of the reports of the European Union and the United States can refer respectively to: European Judicial Systems Edition 2010 (data 2008): Efficiency and quality of justice – European Commission for the Efficiency of Justice – CEPEJ and the report of the NCSC – National Center for State Courts. The report of the European Commission for the Efficiency of Justice, 2010 and 2008 data, is available on the Council of Europe (http://www.coe.int/t/dghl/cooperation/cepej/default_en.asp), the NCSC report is available on its official page on the web (www.ncsc.org).
- 2 Source:Economist Intelligence Unit – The Economist Magazine.
- 3 For more details see Department of Judicial Research at http://www.cnj.jus.br.
- 4 For the exposition of item 1 of this article, we have based on research on the CNJ study Studies on data collection systems.
- 5 The report in The High Numbers is available at http://www.supremoemnumeros.com.br/i-relatorio-abril2011-o-multiplo-supremo/.
- 6 Betweenthe time of writing and completing this article, the CNJ provided through Ordinance n. 216 of December 19, 2012, the entire database that underlies the report «Justice in Numbers» of the year 2009 for free consultation on the web-site at http://www.cnj.jus.br/programas-de-a-a-z/eficiencia-modernizacao-etransparencia/pj-justica-em-numeros/2013-01-04-19-13-21.
- 7 ResearchersCodeço Flavio Coelho, Renato Souza Rocha and Pablo Camargo Cerdeira developed two applications on the database processes the STF. In the first application, the work Talk Information Mining and Visualization of a Large Volume of Legal Texts (http://www.euroscipy.org/talk/4182), they developed an animation that shows the first test plot of the law according to the Brazilian Supreme. It lists the laws cited in more than 1.2 million decisions. When two laws are mentioned together in the same decision they form a link between them, and when each new decision in which they appear together is found this link is strengthened. With this we can see how the ministers of the Supreme Court associate depicted in space. Graph Relationship between laws and Flavio Coelho Pablo Cerdeira, description and video available at http://www.youtube.com/watch?v=zbHXj3ANbqE. The second application used a clustering technique based on several variables: the main branches are the states of the federation, the first branches are the types of decisions (monocratic, presidency, full etc.), The second branches represent procedural classes (aggravations instrument, ADIs etc..) and (the circles at the end) represents the amount of decisions. Thus it becomes possible to see which states are growing fastest in each period, and identify each type of decision in each class proceedings. This large amount of information is almost impossible to be perceived in traditional views of graphs. Description Pablo Cerdeira in http://www.supremoemnumeros.com.br. The view held by Flavio C. Coelho, is available at http://www.youtube.com/watch?v=IHnJyfNoEgg&feature=player_embedded.
- 8 Cf. Ingo Keilitz, Smart Courts: Performance Dashboards and Business Intelligence.
- 9 Anexample of dynamic graphs is the system used by the state courts of Utah, the Utah Courts Performance Measures, which can be accessed at http://www.utcourts.gov/courtools/.
