Currently there is a revival of the study of dialectical argumentation in the artificial intelligence community. There are good reasons why: First, the notions of argument and counterargument shed new light on nonmonotonic reasoning. Second, the process character of dialectical argumentation inspires new computational techniques. In a recent important paper, Dung [1] has studied the relations of (unstructured) arguments and their counterarguments in terms of admissible sets. He has investigated the relations between several types of extensions of argumentation theories. In this paper, we propose a model of the stages of argumentation, related to that of Verheij [2, 3]. Each stage is characterized by the arguments that have been taken into account and by the status of these arguments, either undefeated or defeated. This stage approach provides better understanding of the process of argumentation, because sequences of stages can be interpreted as lines of argumentation. The stage approach also gives naturally rise to two new types of extensions. Their definitions formalize the idea that as many arguments are taken into account as possible. We show the connections with Dung’s work and give a number of examples. It turns out that the argumentation stage approach generalizes the admissible set approach. The main conclusion of the paper is that the argumentation stage approach can give more insight in the procedural nature of dialectical argumentation than the admissible set approach.
[1]
A. Campbell,et al.
Progress in Artificial Intelligence
,
1995,
Lecture Notes in Computer Science.
[2]
Judith Wusteman.
Philosophy and AI: Essays at the Interface
,
1992
.
[3]
Phan Minh Dung,et al.
On the Acceptability of Arguments and its Fundamental Role in Nonmonotonic Reasoning, Logic Programming and n-Person Games
,
1995,
Artif. Intell..
[4]
John L. Pollock,et al.
Justification and Defeat
,
1994,
Artif. Intell..
[5]
R. P. Loui.
Ampliative Inference, Computation, and Dialectic
,
1989
.
[6]
Thomas F. Gordon,et al.
Pleadings game - an artificial intelligence model of procedural justice
,
1995
.
[7]
Bart Verheij,et al.
The Influence of Defeated Arguments in Defeasible Argumentation
,
1995,
WOCFAI.
[8]
Ronald Prescott Loui.
The Workshop on Computational Dialectics
,
1995,
AI Mag..
[9]
Bart Verheij,et al.
Arguments and Defeat in Argument-Based Nonmonotonic Reasoning
,
1995,
EPIA.
[10]
Nicholas Rescher,et al.
Dialectics: A Controversy-Oriented Approach to the Theory of Knowledge
,
1977
.