3rd Workshop on
Computational Models of Natural Argument

The workshop intends to recognise and consolidate the critical mass that research in the field overlapping Argumentation Theory and Artificial Intelligence has developed in recent years. Potentials for exploitation of literature in the philosophical theory of argumentation, in informal logic, in dialectics, have been recognised relatively recently by researchers in artificial intelligence, but already fruits of such cross fertilisation are beginning to ripen. Research in negotiation protocols for multi-agent systems, showing higher level of sophistication, including explicit treatment of autonomy, delegation and trust, is probably the most evident and recent result of such collaboration, but important issues have also been raised in the fields of Artificial Intelligence and Law, in Computational Linguistics, in Computer Supported Collaborative Work, in Artificial Intelligence and Education, together with classical research in knowledge representation and reasoning. Benefits have not been unilateral for AI, as proved by various collaborations with those interested in the philosophy of argument, and by the increasing presence of AI scholars in classical argumentation theory events.

The workshop that we are proposing here focuses itself on answering one particular core question: how can "natural" argumentation be modelled? The need for more "naturalness" has been raised on several occasions, and several interpretations have been given to the attribute. Especially when the human is involved in the loop, arguments which are, for instance, expressed in natural language are definitely more appropriate. But the use of human languages is not the only way to bring the reasoning process closer to its users. Naturalness may involve the use of means which are more immediate than language to illustrate a point, such as graphics or multimedia. Naturalness can also relate to the preference for one particular style of reasoning as opposed to another, to structure complex arguments. Or to the use of more sophisticated rhetorical devices, interacting at various layers of abstraction. Or the exploitation of "extra-rational" characteristics of the audience, taking into account emotions and affective factors. In perhaps not so extreme cases, natural argumentation may also lead to the relaxation of the hypothesis of collaborativeness (or indeed sincerity) among arguing agents, and therefore to the need of not excluding a priori the ability to "protect" oneself by means of deceptive behaviour.

CMNA 2003 is the third workshop in the series, and will be held at IJCAI-03 in Acapulco, Mexico, the first having been held at ICCS in 2001 in San Francisco, and the second at ECAI in 2002 in Lyon.

Areas of Interest:

• The characteristics of "natural" arguments: ontological aspects and cognitive issues.
• The use of models from informal logic and argumentation theory, and in particular, approaches to specific schools of thought developed in informal logic and argumentation.
• The linguistic characteristics of natural argumentation, including discourse markers, sentence format, referring expressions, and style. Empirical work based on corpora looking at these topics would be especially welcomed.
• The generation of linguistic natural argument: techniques in natural language generation to account for argumentation and persuasion (including techniques for handling discourse goals and structure, speaker/hearer models, content selection, etc.)
• Rhetoric and affect: the role of emotions, personalities, etc. in argumentation.
• Computational approaches to the similarities and differences between dialogic and monologic argumentation.
• Issues of domain specificity, and in particular, the independence of argumentation techniques from the domain of application.
• Language dependence and multilingual approaches to argumentation.
• The ethical implications of implemented systems demonstrating, for example, licentiousness and deceipt.
• Applications of argumentation based systems, including, for example, the pedagogical, health-related, political, and promotional.
• Argumentation in multi-agent systems, especially with respect to protocols enabling negotiation of delegation, and trust.
• The methods to better convey the structure of complex argument, including their summarisation.
• Natural argumentation and media: visual arguments, speech, multi- mediality.
• Evaluative arguments and their application in AI systems (such as decision support and advice giving).
• Non monotonic, defeasible and uncertain argumentation.
• Computer supported collaborative argumentation.
• Computer supported teaching of argumentation and critical thinking skills
• The building of computational resources such as online corpora related to argumentation

The workshop is therefore specifically aimed at gathering together scholars from diverse disciplines and backgrounds to establish the current state of the art in the interdisciplinary area overlapping the philosophy or argumentation and AI, and to provide a forum in which to develop new cross-disciplinary collaborations and new lines of inquiry.

Submission Instructions

The workshop encourages submissions in two categories:

• Long papers, either reporting on completed work or offering a polemic discussion on a burning issue (up to 6000 words)
• Short papers describing work in progress (up to 3000 words)

Authors of papers accepted to CMNA 2001, CMNA 2002 and CMNA 2003 will be invited to revise and resubmit their manuscripts for publication in a special issue of the International Journal of Intelligent Systems.

The workshop again aims to stimulate discussion, and in service of this ideal, our submission deadline is a little early to allow pre-workshop discussion. Details of this process will be sent to authors of accepted papers.

Note: Participants are expected to register for the main IJCAI conference in addition to the workshop.

Important Dates and Deadlines

• Deadline for the submission of long papers: 3 March 2003
• Deadline for the submission of short papers: 24 March 2003
• Notification for long papers: 17 March
• Notification for short papers: 1 April
• Deadline for the receipt of camera-ready papers: 12 May 2003

Organizing Committee

Chris Reed (Chair)
Department of Applied Computing
University of Dundee
Dundee DD1 4HN
Scotland
chris@computing.dundee.ac.uk
http://www.computing.dundee.ac.uk/staff/creed/

Floriana Grasso
Department of Computer Science
University of Liverpool
Liverpool L69 7ZF
England
floriana@csc.liv.ac.uk
http://www.csc.liv.ac.uk/~floriana/

Giuseppe Carenini
Department of Computer Science
University of British Columbia
Vancouver, B.C. Canada V6T 1Z4
carenini@cs.ubc.ca
http://www.cs.ubc.ca/~carenini/

Program Committee

Leila Amgoud, IRIT (Institut de Recherche en Informatique de Toulouse)

Trevor Bench-Capon, University of Liverpool

Alison Cawsey, Heriot-Watt University, Edinburgh

Fiorella de Rosis, University of Bari

Rino Falcone, Institute of Cognitive Science and Technology - CNR, Rome

John Fox, Advanced Computational Lab, Cancer Research UK, London

Nancy Green, University of North Carolina Greensboro

Peter McBurney, University of Liverpool

David Moore, Leeds Metropolitan University

Simon Parsons, City University of New York

Henry Prakken, University of Utrecht

Ehud Reiter, University of Aberdeen