Lakoff's Women, Fire, and Dangerous Things is a treasure trove of linguistic examples and a carefully developed model of cognition argued on the basis of semantics. His experientialism places the human act of cognition in the center; his brilliantly presented result is that cognition is vitally dependent on metaphor, which he defines as a mapping of conceptual structures from one domain onto another--a result of particular relevance to literature.
Lakoff's approach of developing a general model of cognition on the basis of semantics has certain inherent weaknesses, in spite of its spectacular results. It cannot be taken for granted that semantic categories accurately represent cognitive domains--language may have access only to the output of other cognitive modules, and their domain-specificity may be partly elided by linguistic categories. For this reason, evidence for cognitive domains must be sought and demonstrated independently of language. However, semantics can be utilized as a way of generating hypotheses about domain-specificity, which can then be independently verified. The presence of cognitive domains also raises the question of how these came about. Lakoff assumes they flow out of our physical constitution and the nature of the world; a more precise way of speaking about this is evolutionary psychology. A consideration of the environment in which humans evolved would permit us to map the source domain onto a proper domain, and thus generate a more detailed list of properties and entailments. Such historical considerations would also allow us to provide principled answers to which domains do not have their own preconceptual structure: namely, domains that either did not exist in the ancestral environment (agriculture, most forms of technology, civilization) or domains that were not available or significant to survival (microbiology, quantum physics, chemistry). Moreover, Lakoff's notion of metaphor as a mapping from one cognitive domain to another as "one of the great imaginative triumphs of the human mind" has been echoed by the British paleo- anthropologist Steven Mithen (1996), who has suggested that the transition from Neanderthal man to Cro Magnon is marked precisely by the ability to "switch cognitive frames": the paleolithic blossoming in art may be correlated with the ability to think metaphorically. Lakoff's proposal on metaphor is a particularly pregnant one for literary studies-- not because ordinary speech is mainly literal (it clearly is not), but because literature is a deliberate forefronting of linguistic devices, a cultivation of special effects. Clarifying what is the source and proper domains of a metaphor promises to throw light on the way meaning is constructed in reading. As for cultural studies, Lakoff's work opens up for the possibility of tracing political and social patterns back to an underlying set of metaphors (see Lakoff and Johnson (1980), Brown (1988), and Boyer (1988)). It should also be noted that Leonard Talmy, whose early work clearly belonged to the Lakovian paradigm (see the next link), more recently has developed this approach explicitly in the direction of evolved cognitive structures (see Talmy (1995)).