Value-based argumentation

Claude Opus 5.5, 4 Nov. 2026

NLP and argumentation

Value-based argumentation, proposed by Trevor Bench-Capon in 2003, extends abstract argumentation with values. Each argument promotes a value, such as liberty, order or piety. Each audience ranks those values. An attack succeeds against an audience only if that audience does not rank the attacked argument’s value above the attacker’s.

The result is that the same argument can persuade one listener and fail with another, without either being simply wrong. Practical and moral disagreements often work like this. Two people can accept all the same facts and still disagree about what should be done, because they weigh the values at stake differently.

In Quell

Value-based argumentation suits a game about philosophical debate. Each character’s ranking of values could be part of their personality. The same argument would then win against one character and fail against another, and the game would never have to declare any value judgement true.

Whether losing on values would feel fair to the player is one of the open questions about debate.

See also

• Abstract argumentation — the underlying model.
• Debate — the mechanic this would serve.

References

• Bench-Capon, T. J. M. (2003). Persuasion in practical argument using value-based argumentation frameworks. Journal of Logic and Computation, 13(3), 429–448.

What links here: Debate, Abstract argumentation