Women, Fire, and Dangerous Things is a great book. It was George Lakoff's summation of everything known through 1987 on the subject of mental categorization, and how the conceptual organization of the mind is expressed through metaphors. Lakoff adduces dozens of examples from linguistics, cognitive psychology and anthropology to demonstrate the psychological reality of his theory, and SC regards it to be a reasonable first approximation of a True Theory of Mind.
The essential data structure for Lakoff's theory is a device called the Idealized Cognitive Model (ICM). An ICM is a concept (or cluster of related concepts) that define our knowledge of a category. At the center of an ICM are those features which most strongly characterize the category. Thus, to loosely borrow one of Lakoff's examples, a bird has two wings, can fly, has a beak, etc. Members of a category which most strongly fit the definition imposed by the ICM are called prototypes. A prototypical bird would be something like a sparrow or a robin.
Not all members of a category are prototypical, though. Some members deviate in just one or two ways, and some are barely recognizable as members of the same type. A chicken, for example, is less characteristic of our mental prototype of a bird -- it can't fly all that far, and the proportions of the body are rather different than those of sparrows or robins. Compared to ostriches, though, chickens are positively prototypical. What this means in practice is that: 1) categories are graded, and 2) the organization of mental models is radial (think of spokes on a wheel, but with some spokes longer than others).
The argumentation in Women, Fire, and Dangerous Things enriches the model considerably, but Lakoff doesn't assume much more than the basic model presented above, and so we'll stick to that. He does give a typology of prototypes, though (p. 8-11), and we'll list them here, since they'll be relevant later on. All examples provided here are paraphrases of Lakoff. Central subcategories are the core types to be adhered to/deviated from by real entities. Thus, if one category is "political thinkers", central subcategories are liberals and conservatives. Typical cases are what we usually think of; i.e., robins and sparrows when someone says "bird". Ideal cases are (possibly abstract) exemplars that we measure real objects against; in the present works, Lakoff offers alternate conceptions of ideal parents, ideal citizens, etc. according to the different worldviews in play. Anti-ideal cases aren't merely the negation of the ideals; they represent the demonology (Lakoff's term) for a given worldview. Stereotypes are properties associated with groups on the basis of myths or well-known individual examples; Lakoff gives "drunken Irishmen" and "industrious Japanese" as examples of stereotypes. Salient exemplars are memorable examples used to draw analogies to categories; they're some of the specific individuals that give rise to stereotypes. Finally, essential prototypes are bundles of features we associate with categories; they help us create folk theories of the "essences" of things (which shouldn't be understood as necessarily literally true). Beaks are essential to birds; hardness is essential to rocks.
This mental machinery suggests that human beings do best at reasoning by analogy, which is exactly how Lakoff claims we go about much, if not all, of our thinking. Lakoff is a committed philosophical materialist, and extends that with particular claims that our reasoning is "experiential"; we create and extend our categories, and create analogies, based on our actual experience (as opposed to having them innately, and living through things trumps reading about them). SC doesn't share or endorse all of Lakoff's commitments in this regard, but for present purposes, we'll stipulate that analogical reasoning is about the best way available to make use of the categorical architecture presented above.
So how does this apply to politics? Lakoff postulates that our political thought is primarily derived from analogies to our most strongly familiar models of group dynamics -- thinking about families. Lakoff asserts that the core metaphor underlying all political thought, regardless of your specific views, is "Nation As Family". This is a strong claim, and Lakoff precedes it with a presentation of the ways in which moral reasoning is not metaphorical. Moral reasoning is held to be experientially grounded (pp. 41-3), largely through various dichotomies. It's better to be healthy than sick; rich instead of poor; beautiful rather than ugly, etc. All this falls under a category of "well-being", and it's our thinking about well-being that leads to our moral systems, which in turn produces our political thought.
The well-being category is not metaphorical, but the way we think about it is. Specifically, Lakoff holds that we think about it in terms of accounting -- "keeping the moral books" (p. 44). The metaphors that Lakoff invokes are "Well-Being As Wealth" and "Moral Action As Financial Transaction" (p. 47). These terms are at least superficially plausible; Lakoff provides examples of reciprocity (i.e., "I owe you one"), retribution ("You'll pay for this"), revenge ("I'll get you, Gadget!" [that's not Lakoff's, is it? -- ed.]), and so forth. The key concept about morality is that our accounting metaphor includes notions of moral credit and debt. Doing good things builds up your own moral credit; having good things done for you builds up your own debt. Doing bad things also builds up debt.
One powerful example of morality as accounting is Martin Luther King's "I Have a Dream Speech" (p. 57). King speaks of cashing a check, signing a promissory note, a bad check. The notion of moral bookkeeping like this covers our conception of rights and obligations. When we talk about rights, the financial metaphor helps us understand that there are corresponding obligations on someone else's part to facilitate them. Thus, Lakoff writes, "if you have a right to an education, someone has a duty to provide it".
The financial metaphor also helps us understand moral notions of fairness. There are many different models of fairness (p. 60), like "one child, one cookie", or "one person, one raffle ticket". The model of fairness we assume to be correct is dictated in no small part by our understanding of how the moral books have been kept, and Lakoff ascribes differences in political views to the ways in which we draw analogies from financial thinking to specific cases of possible fairness/unfairness. The reader would be correct in guessing that this most significantly underpins Lakoff's explanation of how we come to our views on affirmative action and taxation.
Lakoff ties this discussion back to the "Nation as Family" notion by arguing that we formulate our views about raising our children by reasoning about how the moral books are kept, and what it will take for our children to successfully interact with other people according to those principles. Government policies represent similar problems to those we face in raising our children, and thus, we draw on our experience of raising a family to decide how we want other citizens to be treated, and how we want our nation as a whole to interact with other nations.
This last part is something that your host is not entirely willing to concede to Lakoff. One can readily accept that we draw analogies between child-rearing and government based on the same principles that we use in order to draw analogies in other parts of life. But as we will see in subsequent sections of this review, Lakoff assigns an almost Freudian primacy to the explanatory power of his theory, and frequently claims to be puzzled by apparently counterintuitive behaviors by both liberals and conservatives, which can only be resolved by appealing to his theory. This wouldn't necessarily be a problem -- if we could be sure that Lakoff's models of either conservative or liberal thought represented their objects accurately. As we will demonstrate later, not only is this frequently not the case, but there are often alternative explanations for political positions which do not require Lakoff's theory to be invoked. In the next section of our review, we'll discuss some of the systematic methodological flaws that plague Lakoff's attempt to construct models of political thought.