back next next
Table of Contents
Index
Glossary










































Table of Contents
Index
Glossary
back next next

Next: Abstract research: logic for Up: Appendix: modularisation in other Previous: The development of new

  
Empirical research in the humanities: argumentation theory

[Van Eemeren et al., 1995] is an example of empirical work in the domain of argumentation theory. In this article, the authors report about empirical investigations on the performance of Dutch secondary education students in identifying unexpressed premises and argumentation schemes as defined in the theoretical framework of pragma-dialectics.

In argumentation, arguers aim to justify a standpoint by means of premises that they think are accepted by their interlocutors. Not all premises have to be made explicit. An example given in the article is: ``Amos is pig-headed, because he is a teacher.''; the premise that teachers are pig-headed is unexpressed. The unexpressed premise serves as a connecting principle between the explicit premise and the standpoint. There are different ways to connect a standpoint to premises. In other words, arguers can use different argumentation schemes. In pragma-dialectics, three main categories of argumentation schemes are distinguished. Following these argumentation schemes, arguers point out that there is a relation of concomitance, analogy or causality between standpoint and premise. In the example, there is a concomitance relation between the standpoint that Amos is pig-headed and the premise that he is a teacher; Amos's pig-headedness is `symptomatic' of him as a teacher. In the pragma-dialectical analysis of argumentative discourse, two of the analytical tasks are to make unexpressed premises explicit and to identify the argumentation scheme that is being employed. The question addressed in this article is whether or not ordinary language users can also identify them in practice.

The article consists of the following main sections: 1.Introduction, 2.Theoretical background, 3.Hypotheses, 4.Design, 5.Results and 6.Discussion. The information presented in this linear article may be recast in the following modules.

Positioning
  This work is embedded in the research programme of pragma-dialectics, regarding argumentative discourse. This embedding does not have to be made explicit in the text of the article, because it is published in a book providing a collection of articles, by various authors, on the pragma-dialectical approach to argumentation. In particular, these investigations are part of a project aimed at establishing a sustained connection between argumentation theory and argumentative practice. This project is only implicitly mentioned in the very short introduction.9.3 In a modular version of this article, the Situation module could be linked to mesoscopic or macroscopic modules about the research programme and the project.

The central problem is formulated in terms of the testing of precise hypotheses concerning the ability of students to identify unexpressed premises and argumentation schemes. The hypotheses, that are given in the section 3.Hypotheses, can be represented in a module Central problem.

Methods
  The authors have asked groups of students to identify, in two separate tests, unexpressed premises and argumentation schemes in multiple choice items. In the original article, the tests are described in section 4.Design, which includes subsections about the test format in general, about the two tests and about the test subjects.

In the domain of empirical discourse studies, a module Empirical methods could be defined as a counterpart of the module Experimental methods  in science. This module could be defined as a compound module containing the constituent modules Test format, Test samples and Test conditions. In the module Test format, the general design of the empirical test could be described - in this case, a paper-and-pencil test with multiple choice items constructed following a specified design method. This module may be linked to mesoscopic or macroscopic modules about test methods (e.g. multiple choice items, other questionnaires, or interviews). In a modular version of this article, the module Test samples could in its turn consist of two constituent modules, distinguished by the domain-oriented characterisation of the information as `unexpressed premises' and `argumentation schemes', each providing the series of multiple choice items that had been submitted to the test subjects. This would allow others to use the samples of multiple choice items in other experiments pertaining to these issues. The description of the test subjects and the situation of this particular experiment can be given in a constituent module Test conditions.

With respect to the theoretical framework, a full account of the pragma-dialectical model, including the idea of unexpressed premises and argumentation schemes, can be given in mesoscopic or macroscopic Theoretical methods modules and linked to this article.

Results
  The results can easily be represented in a complex Results, with separate constituent modules about unexpressed premises and about argumentation schemes. The finished test-papers of the test subjects form the most basic raw data of the experiment. The proportions of correct responses in the test subjects' output are calculated and presented in tables. These results are then used as input in an analysis of the variance of these results with particular fixed factors (e.g. the type of argumentation scheme, and the type of school of the test subjects). For this purpose, statistical methods (i.e. numerical methods) are used. In the linear article, these methods are referred to in section Results. In a modular version, a link can be made explicit to a mesoscopic or macroscopic module Statistical methods. In the analysis of variance, the significance, i.e. reliability, of the results is explicitly taken into account. The statistical methods are also used to determine the reliability of the test, which is given in section 4.Design.

Interpretation and Outcome
  In the section Discussion, it is discussed to what degree the results confirm the hypotheses. Because these hypotheses have already been formulated in detail, the discussion is relatively short. This tallies with the observation in [Buxton and Meadows, 1978, p.177] that the Discussion sections in social sciences in which explicitly formulated hypotheses are tested are shorter than the Discussions in natural sciences. The interpretation of the empirical results in the light of the hypotheses could be represented in a short Interpretation module.

Outcome
  The findings of this article are that the hypotheses are indeed confirmed. In the printed article, these conclusions are drawn in the section Discussion. In a modular version, they could be summarised in a module Findings. Since this work is relevant both in the academic context of the development of argumentation theory and in the practical context of teaching, some implications for the educational system are suggested. For this type of publication, the module Outcome could be defined as a compound of constituent modules Findings, Leads for further research and Practical implications.



Next: Abstract research: logic for Up: Appendix: modularisation in other Previous: The development of new