In order to preserve the desired multidimensionality of the collection of legal documents and to examine their evolution over time, legislative corpora should be modelled as dynamic document networks20,21,22,23,24,25,26. In particular, because legal documents are carefully organized and interconnected, their structure provides a more direct insight into their content and dynamics than their language: networks respect the conscious design choices of document authors and bypass some of the ambiguity issues that natural language-based approaches inherently face. In this article, we therefore develop an informed data model for legislative corpora that captures the wealth of legislative data for research through social physics. We use our data model to analyze the evolution of federal laws in the United States and Germany. Here we see a huge increase in legal complexity depending on the volume, interconnectivity and hierarchical structure of legislation in both countries – proof that the highly industrialized countries we study are trying to control behavior by constructing increasingly complex legal norms. In search of the sources of growth we observe, we use graph aggregation techniques to find the legal issues that contribute most to the increase in complexity and track their evolution over time. If we derive our information on the content of legislative documents directly from the deliberate structural decisions of their authors, we find that the main driver of law growth in the United States and Germany is the expansion of the welfare state, which is supported by an expansion of the fiscal state. Beyond this overall picture, our methodology also allows for finer discoveries – for example, we find that during our observation period, the regulation of natural resources in the United States has shifted from exploitation to conservation. In this way, we achieve results that would be difficult (if not impossible) to achieve using approaches that use only the natural language of legislative documents while minimizing the number of subjective judgments. Our work highlights the potential of legal network data and document network analysis for the study of the interaction between law and society when viewed through the prism of complex adaptive systems (CAS) 17,27,28,29,30,31, and opens up new avenues of research for the interdisciplinary scientific community. In our settings, cluster families are sets of chapters, books, or laws that are closely related by cross-references or (almost) a textual identity over time. As such, they roughly correspond to legal issues. For more information on how we report these topics, see Section 5.1 of the IS.
Cabrelli, D. & Siems, M. M. Convergence, Legal Origins and Transplants in Comparative Corporate Law: A Case-Based and Quantitative Analysis. On the. J. Comp. Law 63, 109-153 (2015). Integrating documents from the executive and judicial branches of government into our datasets could help us examine how different parts of the legal system interact. How can the development of a legislative network be compared to that of a network of by-laws, a network of implementing regulations or a network of judicial decisions? In which areas of law is development dictated by the executive or the judiciary and not by the legislature? What does this tell us about the distribution of power between the different branches of government? From this discussion, he draws conclusions on the question of social order in relation to the development of law.
Unger explains how the disintegration of the community goes hand in hand with a tendency towards increasing specialization in the division of labor of a society, a tendency responsible for the separation of state and society. [14] This process undermines the foundations of a comprehensive set of shared beliefs. [15] The legal system emerges with the pluralism of interest groups[16] and the use of higher divine or universal law as a norm to justify and criticize the positive law of a state. [17] The rule of law is a response to the decline of order in society. [18] But when belief in a higher law declines, the legal order is threatened. In contemporary history, the synthesis of the natural rights that underpin the ideal of the legal order has been reversed, so that the coherence of the ideal of the rule of law is compromised. [19] A first simple way to reorganize the U.S. Code is to aggregate it at the chapter level and not at the title level. This is particularly convenient because the number of chapters of the US Code is comparable to the number of individual laws in Germany, which we only divide into smaller units if they contain several books (a common feature of large German codifications such as the Civil Code [BGB] and the German Commercial Code [HGB]). The node link diagrams of the quotient graphs corresponding to this reorganization for the United States and Germany in 1994 and 2018 are presented in Fig. 4. In these diagrams, nodes have the same color if they belong to the same cluster family.
Overall, cluster families are sets of clusters (a cluster is a set of nodes), typically from different snapshots that contain many of the same, similar, or related rules (see .