Legal Complexity

In today's world, designers of legal systems face an unprecedented challenge: “to craft simple laws that produce desired ends, but not to pursue simplicity so far as to undermine those ends” [1].

What is complexity in legal texts, how does it arise and how can we reduce it? Legal scholars and lawmakers have acknowledged the issues connected to complex legal systems, and attempts have been made to localise and to control it [2,3], but no reliable objective definition or measure is known. In the absence of any quantitative framework or accepted definition, legal scholars have mostly “appealed to everyday, intuitive definitions of complexity”, when they did not go as far as simply averring, “I know it when I read it” [4].

The law can be understood as network of documents that are connected by different kinds of relations. The first one is ``inclusion'', which is the property that one document is part of another document in the same way as a section in a text is part of a chapter. The other is ``reference'', which occurs whenever one document points the reader to another document as the location of a specific piece of information. Moreover, documents can be created, amended and removed, and the result is a complex network that changes over time. Borrowing tools from Network Theory and Complexity Science, we wish to put forward an innovative way to represent and visualise each Act and its “surroundings” as the complex web of inter-relations just described.

We explore different approaches to quantifying aspects of complexity for legal networks. Notable examples are i) ``algorithmic complexity'' [5], being the length of the smallest computer program whose output is the considered system, or an encoding of it and ii) ``knowledge acquisition cost'', the effort required to understand a certain portion of the law, which includes following up on references and previous version to understand the detailed context. Further questions include the stability of these measures: How long does it take for a well designed (low complexity) system to devolve into an opaque system (high complexity)?



  • Dept. of Mathematics King's College
    London Strand London WC2R 2LS
    United Kingdom

  • +44 020 7848 2864