Yet, I would as a sceptic also understand that most of the time when we use metaphorical language we mean more, not less, than the language indicates. The biologist who speaks of a paramecium ‘deciding’ to go thataway for food probably means less than his imagery suggests–most biologists don’t consider paramecia to be capable of conscious choices and other actions, but only capable of automatic reactions and counterreactions. But such use of metaphor (though important) is relatively uncommon. More often, we mean more than the imagery suggests. Language is necessarily reductive, so we have to use similar words for multiple meanings. For example, by ‘reductive’ in the last sentence, I don’t mean that language makes real things smaller. I mean something more complex and nuanced than my language indicates. And I would also play fairly by not requiring that people somehow abandon metaphor and ‘talk plainly’. It can’t be done; the effort to do so results in choosing other metaphors (without realizing they are metaphors) which are often less efficient at helping the idea across than the original metaphor. On the other hand, sometimes it isn’t a bad idea to restate the contention using different imagery and then using a comparison of the two images to help correct and refine the perception of the idea I am trying to communicate. This same process takes place on a somewhat larger scale when analogies are used to help illustrate a previously developed argument.
(Footnote: I would, of course, be on the lookout for the so-called ‘argument from analogy’ where the analogies only illustrate a blanket assertion–the argument only being presumed to have been made–but I would also be careful not to fall off the horse on the other side and accuse someone of arguing by analogy simply because he happens to use a number of illustrative analogies.)