Program 10th eColloq

Monday FEB 24, 2014, 4 pm CET (Central European Time, e.g., Rome, Berlin, Stockholm)

4.00-4.10 Connect, Welcome

4.10-4:35 Gregor Betz (KIT, Stuttgart, Germany): Is there a method for reconstructing natural language arguments?

Abstract: One purpose of argumentation theory is to help us to better understand, analyse and assess natural language argumentation. A theory of argumentation (e.g., deductive logic, Dung’s AFs, or Walton’s argument schemes) typically provides a formal or informal model of argumentation. To assess a piece of natural language argumentation A by means of such a theory, we have to state A in terms of the corresponding model. This process is what we call “reconstruction”. I shall outline a method of argument reconstruction and illustrate it with a complex example. I maintain that reconstruction is a principally open-ended, iterative process _that cannot be separated_ from multi-dimensional evaluation of an argumentation. It’s definitely much more than mere text annotation. I shall briefly discuss implications for argumentation software.

4:35-4:50 Discussion

4:50-5:00 Break

5:00-5:25: Michael Hoffmann (Georgia Tech, GA, USA): Designing argument visualization software as cognitive tools

Abstract: Whatever the function or intended goal of arguing is—persuasion, conflict resolution, knowledge, etc.—arguments need to be represented. A discussion of the various ways in which we represent arguments should be informed by semiotics, the theory of signs and representations. In my own contributions to semiotics—which stand in the tradition of Charles Peirce—I argued that signs do not only have a representational function, but other functions as well. Most importantly, they have an epistemic function, meaning that we use signs to (a) refer to both concrete and abstract objects, (b) structure reality or sets of objects, and (c) hypostasize (reify or construct) abstract objects. Based on such an epistemic function of signs and representations in general, I would like to pursue two things. First, I will argue that representations of arguments also have an important epistemic function: they allow us to structure relations between reasons and conclusions and contribute, thus, to the possibility of knowledge. Second, and more importantly, I would like to show how a reflection on the epistemic function of argument representations can help to design software for the visualization of arguments. Whereas an epistemic approach to argument presentation perceives those presentations as means to structure inferential relations in a very general sense, I propose to use the term “cognitive tool” when argument visualization software is conceptualized as a means for more specific purposes such as learning, discovering, and collaboration. Based on a reflection on the Peircean notion of “diagrammatic reasoning” and his argument that diagrammatic reasoning is only possible by means of representational systems that have to be developed, I will argue that those argument visualization tools are most beneficial with regard to the cognitive purposes listed above in which the number of entities available to represent arguments—in particular the number of argument schemes—is reduced as far as possible. Like in an axiomatic system, my goal is to design software tools in which the largest number of possible representations can be derived from the smallest possible number of representational means.

5:25-5:40 Discussion

To participate as a discussant
 (n.b. “listening in” is fine, too), please register now.