In his brilliant new book, The Stuff of Thought: Language as a Window into Human Nature, Steven Pinker devotes a chapter to metaphor, and how our language is "saturated" with it. Leaving aside the fact that "saturate" could itself be categorized as an imprecise metaphor -- after all, language can presumably accommodate limitless metaphors -- there is, everywhere you look, heightened awareness of the extent to which our opinions, judgments and behavior are shaped by figurative linguistic concepts:
The September Harvard Business Review has an article, "Women and the Labyrinth of Leadership," suggesting that the phrase "glass ceiling" be replaced with the "glass labyrinth," if women are ever going to shatter said transparent structures. In New York, the hoary "political parties as parents" metaphor and recasts the Democrats as "Jodie Foster mommies" and the Republicans as "deadbeat daddies"; The Boston Globe wants people to stop thinking of America's defense against terrorism in terms of "a metaphor of endless war against an amorphous enemy"; and in Ethan Nadelmann's Foreign Policy cover story, he asserts that "futile rhetoric" regarding "the war on drugs" means that casualties in this unwinnable "war" are accepted by the public as collateral damage.
Nadelmann's assertion implies that the framing of drug law enforcement as "war" goes largely unrecognized; indeed, a central question of Pinker's is whether people are generally conscious of the metaphors used for communicating, or blind to them. One school of thought -- "the killjoy theory" -- proposes that most metaphors are indeed "dead," used so often and for so long that their figurative meaning (but not necessarily resonance) has drained away. Certainly, the war metaphor is pressed into service in so many contexts and in such oblique forms that it's rarely noticeable ("war on drugs" is taken as literal) even though, Pinker points out, the vast number of expressions conceptualizing argument as war shows that "the tacit metaphor must have been transparent to a large number of coiners and adopters for a very long time."
Obviously, an expression like glass ceiling is too blatantly metaphorical to go unnoticed, which is part of the point that the HBR essay makes: such a concrete and specific image to explain the dearth of women in powerful jobs has focused attention too narrowly on the notion of one barrier at the highest level, rather than on various obstacles along the way -- indicating that fresh, conspicuous metaphors, of which there are in fact plenty, can be as problematically influential as those we're oblivious to.
Still, Pinker reassures us that "metaphors can be tested on their predictions and scrutinized on their merits, including their fidelity to the structure of the world." The kind of vigilance exercised by the writers quoted above evidently does much to undermine the power of metaphor, especially as used by politicians. Not that vigilance is always necessary: while "witch hunt" is surely not too far from joining the "dead" group of metaphors, it's unlikely that too many people needed to parse Sen. Larry Craig's invocation of seventeenth century Salem and fifties McCarthyism to decide that he isn't currently a victim of one.