Elsevier

Decision Support Systems

Volume 41, Issue 2, January 2006, Pages 358-379
Decision Support Systems

The generic/actual argument model of practical reasoning

https://doi.org/10.1016/j.dss.2004.07.004Get rights and content

Abstract

In this paper, we present a model of reasoning called the generic/actual argument model (GAAM). Reasoning within a discursive community can be represented with this model so that participant claims can be accommodated without recourse to combative metaphors such as attack or defeat. The model facilitates the comprehension of complex reasoning for humans as well as being a computational representation for machine modelling of reasoning. As such, the model naturally integrates machine inferences with human. The model has been the basis for the development of practical systems to support reasoning and deliberation in areas of law and organizational decision making. Here, we present a formal description of the model and identify some of its characteristics.

Introduction

In this paper, we present a formal description of the generic/actual argument model (GAAM) and develop from this some of its characteristics, practical advantages and disadvantages. The GAAM is intended as a model for decision support for decision making by individuals within a group or discursive community. It can be used by individuals without inference support, by individuals with varying degrees of inference support or as a fully computational system. The GAAM has been used to model reasoning in copyright law by Stranieri and Zeleznikow [46], predict judicial decisions regarding a property split following divorce by Stranieri et al. [44], support refugee status decision makers by Yearwood and Stranieri [54], facilitate interactive e-commerce by Yearwood et al. [56], implement multi-agent negotiation by Avery and Yearwood [4], and in determining eligibility for government funded legal aid by Stranieri and Zeleznikow [47]. Two shell programs that implement GAAM ideas are described in Stranieri and Zeleznikow [47] and Yearwood and Stanieri [55].

The GAAM was developed as a framework for modelling discretionary reasoning and has been used to develop practical decision support systems over the last 5 years. The objective of this paper is to provide a description of the GAAM in terms of:

  • identifying its basic set of propositions and how they are combined

  • identifying the elements that formally control or represent the structure of reasoning

  • its inference mechanisms and how propositions are derived

  • the extent to which derived propositions are valid and accepted

  • the way in which it supports discretionary decision making

  • setting out its capabilities as a non-dialectical model upon which a dialectical model can be built.

The remainder of this paper is organised as follows: Section 2 provides a brief review of Toulmin argument structures. Section 3 sets out how the elements of the GAAM relate to Toulmin argument structures and discusses inferences and the separation of inference from the structure of reasoning. Section 4 presents the GAAM more formally and in detail. Section 5 discusses some of the characteristics of the model, exploring deducibility and possible notions of argument strength and validity. Section 6 compares the model with other approaches.

Section snippets

Toulmin argument structures

Toulmin [48] concluded that most arguments, regardless of the domain, have a structure that consists of six basic invariants: claim, data, modality, rebuttal, warrant and backing. Every argument makes a claim based on some data. The argument in Fig. 1 is drawn from reasoning regarding refugee status according to the 1951 United Nations Convention relating to the Status of Refugees (as amended by the 1967 United Nations Protocol relating to the Status of Refugees) and relevant High Court of

The generic/actual argument model

Often reasoning occurs in the context of a small group of stakeholders involved in dialogue who would like to reach agreement on some issue. Whilst there is much anecdotal evidence for this it is also true that most organizations like to see a team approach to the solution of problems but are keen to have frameworks that permit a range of views. In general, we can distil the following characteristics of small group reasoning:

  • Membership of the discursive community is usually well defined

  • Members

Defining the GAAM

The GAAM is a means of specifying generic argument structures to model reasoning within a domain.

Syntax and deducibility

The propositions of a GAS for a domain modelled by the GAAM are the claims that can be formed within claim slots. So, this is the finite set of expressions of the form CpCvkCs. All logical connectives, if needed, are encoded in the inference procedures of inference slots. Propositions can only be combined if they occur together attached to inward arcs of the same inference slot. Their combination then has to be with any other claim slots that occur in the domain of this inference slot.

There is

Other approaches

Our work here has focused on the use of argumentation to structure reasoning (i.e. a non-dialectical emphasis) rather than on the use of argumentation to model discourse (i.e. a dialectical emphasis). Argumentation has usually been associated with defeasible reasoning and many have approached defeasible reasoning from the point of view of developing formal logics. For example, Nute [34] describes defeasible reasoning as:

When some new fact causes us to reject a prior conclusion, we will say that

Conclusion

In this paper, we have described the generic/actual argument model. The model derives exibility and power from: nodes whose generality is efficient in capturing many instance arguments that essentially have the same structure: a clear layout of the structure of reasoning, a clear delineation of inferences, capturing dialectical positions within a common structure. We have:

  • identified its basic set of propositions and how they are combined

  • identified the elements that formally control or represent

Acknowledgement

This research was supported by the Australian Research Council.

References (57)

  • M.D. Bayles

    Procedural Justice. Allocating to Individuals

    (1990)
  • Karl L. Branting

    A computational model of ratio decidendi

    Artificial Intelligence and Law: An International Journal

    (1994)
  • L. Karl Branting

    Reasoning with Rules and Precedents—A Computational Model of Legal Analysis

    (2000)
  • D. Carbogim et al.

    Argument-based applications to knowledge engineering

    Knowledge Engineering Review

    (2000)
  • G.C. Christie

    An essay on discretion

    Duke Law Journal

    (1986)
  • P. Clark, A model of argumentation and its application in a cooperative expert system, PhD thesis, Turing Institute....
  • P. Cohen

    Heuristic Reasoning About Uncertainty: An Artificial Intelligence Approach

    (1985)
  • Jeff Conklin et al.

    gIBIS: a hypertext tool for exploratory policy discussion

    ACM Transactions on Office Information Systems

    (1988 (October))
  • Judith P. Dick

    Conceptual retrieval and case law

  • Judith P. Dick. A conceptual, case-relation representation of text for intelligent retrieval, PhD thesis, Department of...
  • K. Engisch

    Logische Studien zur Gesetzesanwendung

    (1960)
  • A.M. Farley et al.

    Burden of proof in legal argumentation

  • John Fox

    Knowledge, decision making and uncertainty

  • John Fox et al.

    Arguing about beliefs and actions

  • Clark Glymour et al.

    Default reasoning and the logic of theory perturbation

  • T.F. Gordon

    The pleadings game: an exercise in computational dialectics

    Artificial Intelligence and Law

    (1995)
  • Thomas F. Gordon et al.

    The Zeno argumentation framework

  • Cited by (35)

    • That is your evidence?: Classifying stance in online political debate

      2012, Decision Support Systems
      Citation Excerpt :

      We also conduct an experiment to classify rebuttals, as a type of disagreement discourse relation, and show that we can identify rebuttals with 63% accuracy. There are several threads of related work that focuses on classifying a speaker's “side” or “stance” toward a debate topic in either formal or informal debate settings, such as congressional floor debates or in conversations from online forums and debate websites [3,34,38]. The research most strongly related to our own is that of Somasundaran and Wiebe [31,32], who also report results for automatically determining the stance of a debate participant in online forums.

    View all citing articles on Scopus
    View full text