The IMPACT Project: Policy Argument Modeling & Text Analysis

A number of legal informatics scholars and institutions are involved in the new EU research project called IMPACT: Integrated Method for Policy Making Using Argument Modelling and Computer Assisted Text Analysis.

The goal of the project is to apply those computing methods to “facilitate deliberations about policy at a conceptual, language-independent level.” Many of the policy debates targeted by the project will inform law-making processes.

The project also seeks to develop applications and tools to assist officials and citizens in deliberating about policies, including “tools for reconstructing arguments from data resources distributed throughout the Internet.” The tools will be designed as Rich Internet Applications (RIAs) or widgets, so that they can be integrated into a variety of systems.

The tools are to be integrated into an open source content management system. The project developers plan to make publicly available an Application Programming Interface (API) for services required by the tools.

Legal informatics institutions involved in the project include:

For more information, please see the project’s Website.

Leave a Comment


Leave a Reply

Srinidhi Boray

This is cool. There is something more to it. There is something called “cartesian dilemma” owing to which the subjective ideas and thoughts (like legislations and policies) find it very difficult and sometimes impossible to translate ( or better transform) into cohesive set of objective actions that align in a proper structured way to address holistically the policy established.

At the policy level the ideas also have the problem of being”empirical”, so they might not transform into correct logical constructs. So somewhere in the journey of translation the rationality is lost.

It will be awesome if argument modeling is employed to “facilitate deliberations about policy at a conceptual, language-independent level.”

Example for this is Enterprise Architecture that attempts to translate strategies and missions that is driven by legislations and policies into proper arguments. Unfortunately, since the existence of empirical notions are not properly understood, people tend to make plans that is not design mind oriented which argument modeling attempts to establish. Some such issue was discussed by Mary Parker Follett nearing hundred years back. And, yet we have many policies and legislations that have not transformed in more productive instruments for the society.

Rob Richards

Mr. Boray suggests that the IMPACT project cannot succeed, because neither policies, nor laws, nor arguments about policies or laws, can be modeled by computer programs. This assertion seems inconsistent with a great deal of research that has occurred in this area in the past three decades — see, e.g., the conference literature linked here: and the journal literature linked here: — and with a large number of systems and applications that have been produced as a result of that research — see, e.g., the list of projects linked here: . Mr. Boray might wish to examine the prior research, systems, and applications on which the IMPACT project leaders plan to base their research: . Mr. Boray might also wish to examine the many applications currently available for modelling law-related arguments, many of which are mentioned by Dr. Floris Bex in his recent post: . Mr. Boray might also wish to examine the many legal knowledge representation resources currently available, many of which are mentioned by Dr. Núria Casellas in her recent post: .

Srinidhi Boray

Mr. Rob

On the contrary I am suggesting otherwise. That the endeavor for programmatically evaluating the policies by applying modeling is a “good thing”. In fact I am suggesting in support of your work owing to the fact that it overcomes “cartesian dilemma”.


Srinidhi Boray

What is modeled is the “argument” abstracted – a design logical construct for the policy.

It appears to me that in the first discernment of the “narration” of my first comment was not abstracted into logical argument construct, before attempting to understand what is being said.

An “argument” a logical construct has potential to overcome the many interpretations that simple linear narration can introduce. While doing so, since argument itself is Cartesian in nature, it exposes the “cartesian dilemma” experienced. This issue further along amplifies when “ontologies” are sought that seem to be modified taxonomy. Ontology and Taxonomy are two different things. But many times applied to achieve similar results. Taxanomy is more “data” centric and to incorporate behavior that includes “process” Ontology needs to be sought. In the journey of seeking a true representation in terms of a model / or set of models, somehow the in-completeness for the subject’s context representation will be experienced. Despite this fact “arguments” lend in better rationalizing the polices for ensuring that they are better tactically realized and implemented.

Thanks for the links, it was nice reading them.