Why use metaphor analysis to teach argument, rather than formal deductive logic?

Metaphor

Metaphors and models are structured by similarity relations and pattern recognition, whereas formal logic is structured by relations of identity and difference and by definitions in terms of necessary and sufficient conditions for category membership.

Metaphors, models, gestalts, images and schemas, generate entailments because the way the whole is interpreted functionally and structurally constrains the interpretation of the parts.

Models and metaphors are functioning wholes which include causal properties.

Objective categories

The relevance of formal logic for natural language reasoning depends on the idea that ideas and objects can be sorted into categories based on shared objective properties and characteristics.

While objectivist categorisation and formal logic systems may suit the binary code of digital computers well, it turns out that language is an analogue system which does not work the way objectivist categorisation expects it to.

Logic & Reason

Lakoff shows that the container schema and the part whole schema can map onto the properties of formal logic.¹

This means that when the features of the world relevant to a particular problem or issue can be treated as if they map onto the container and part-whole schemas, then formal logic can be used to reason about the situation. However when the relevant features of a situation do not fit the container and part-whole schemas, formal logic may not be useful as a guide to reasoning about the situation.

In our reasoning we rely on different inferential and operational rules or heuristics according to specific contexts, contents, or schemas.  Success at such inferential tasks is dependent on causal and logical analysis of the problem coinciding.²

Because action requires us to get our causes right, our inferential practice gives precedence to causal schemas over formal logical analysis. 

Logical vs causal reasoning

The truth functions of formal logic do not match the truth functions of causal inferences.  To the extent that causal reasoning is an important part of argumentation, formal logic cannot provide guidelines for developing or critiquing such arguments.

Because models and metaphors can incorporate causal factors, reasoning with models and metaphors can be used to construct effective arguments in relation to causal inference.

Conclusion

In order to arrive at and defend our use of a particular heuristic in a specific context we need to understand how conceptual metaphors work, and how to analyse their adequacy for the context at hand.

Arguments based on metaphor analysis are oriented towards achieving the 'aha' reaction of art or scientific discovery – 'that is an interesting way of looking at it' or 'I hadn't looked at it like that before'.

This contrasts with the objectivist assumptions of formal logic, which are oriented towards 'winning' arguments – 'you are wrong' or 'I am right''.

Metaphor helps you recruit allies, whereas in order to 'win' with logic you are forced to create enemies. Argument based on metaphor is a much more effective communication strategy.

Notes:

1 Lakoff, G. 1987  Women, Fire and Dangerous Things: What Categories Reveal about the Mind, The University of Chicago press, Chicago., p.365-7, 456-9.

2 Johnson-Laird & Wason, 1977 (Eds.) Thinking.  Cambridge University Press, Cambridge, p.193.