Why do people hold particular political beliefs? Why is there such a large and persistent divide, at least in the USA, on the same set of core issues? And why is communication so difficult between people holding opposing views?
In the mid-1990s – soon after the publication of Newt Gringrich’s Contract With America – George Lakoff, a cognitive linguist at UC Berkeley, found himself asking these very questions. Lakoff recognized that many people formed belief clusters; a particular set of beliefs is frequently associated with the prototypical “progressive” point of view, while another (often opposing) set of beliefs is associated with the prototypical “conservative” point of view. He sought to develop a framework to explain how these political ideologies emerge, and how people come to identify with one ideology or another. Central to this model was the notion of Conceptual Metaphor Theory.
Metaphors We Live By
In 1980, George Lakoff and Mark Johnson published Metaphors We Live By (Lakoff & Johnson, 1980). Previously, metaphor was primarily discussed in the context of literary analysis and poetry, and the study of metaphor focused on conscious, intentional metaphors. Lakoff and Johnson, however, argued that metaphor is actually a vital part of human language and cognition.
Metaphor, in their view, is the process of understanding one idea in terms of another; this often manifests in the form of borrowing terms from one domain and using them for another domain. They offer the example of conceptualizing ARGUMENTS in terms of WAR:
He shot down all your points.
My claims are indefensible.
His criticisms are right on target.
Here, elements from the ARGUMENT domain are described using language typically reserved for WAR. Lakoff and Johnson demonstrate that once you think about it, you realize that metaphor is extremely pervasive in everyday language. They identify a variety of common “metaphor families”, including: IDEAS are FOOD (“half-baked idea”, “I’m still digesting this article”), INTIMACY is WARMTH (“he’s a frigid person”, “she’s so warm”), UNDERSTANDING is VISION (“I see what you mean”, “The truth dawned on him”, “he became enlightened”), and much more.
Crucially, Lakoff and Johnson argue that these metaphors are actually systematic, and that they affect the way we think. That is, human cognition can be viewed (metaphorically, of course) as having been constructed from basic conceptual building blocks, which map onto embodied experiences, such as: moving through the world, picking things up, seeing other people, falling down, experiencing emotions, and so on. We understand these embodied experiences using particular cognitive machinery. According to Conceptual Metaphor Theory, we recruit that same cognitive machinery to understand higher-level, more abstract concepts.
Since the release of Metaphors We Live By, Conceptual Metaphor Theory has spread like wildfire, with research programs that are both varied and far-reaching. Computational linguists have developed tools to automatically identify metaphors in text (Stefanowitsch, 2004; Del Tridici, 2006); psychologists have found that metaphors in language influence opinions about crime (Thibodeau & Boroditsky, 2011), immigration (Charteris-Black, 2006), and even climate change (Flusberg et al, in press); and most relevantly, those in politics have begun to take note of the theory’s implications for understanding their constituents and framing political platforms.
Metaphors We Vote By
In 1996, George Lakoff released Moral Politics: How Liberals and Conservatives Think. In it, he attempts to answer the questions posed at the beginning: why do liberals and conservatives believe what they believe? And why do they have such a hard time communicating?
In answer to the first question, Lakoff argues that at the root of people’s political beliefs is a metaphor about the government (Lakoff, 1996). More specifically, people tend to view the NATION as a FAMILY, giving rise to language like:
We send our sons and daughters to war.
The “Founding Fathers”.
Consequently, people’s views about the ideal form of government are derived from their views about the ideal family structure. Lakoff presents two “idealized models” of the family, roughly corresponding to the progressive and conservative ideologies, respectively:
The nurturant parent view emphasizes empathy and communication. Individuals in the family should be free to pursue their own form of happiness, as long as their choices do not impede on someone else’s happiness. When it comes to raising children, the “nurturant parent” believes that children learn to act morally by observing other people acting morally (such as their parents), not through punishment. Here, immorality is acting selfishly and anti-socially; the world itself is not inherently just or moral, so everyone should work to improve the condition of others.
The strict father view emphasizes discipline and moral character. Children learn to act morally through “tough love” and a system of rewards and punishment. Parents – particularly the father, in the more extreme or patriarchal view – are essentially arbiters of morality, and so children must learn to obedient to the parents. Contrary to the nurturant parent view, the strict father view holds that the world, though perhaps difficult, is fundamentally just: people get what they deserve. If you work hard, you succeed; if you fail, it is likely due to laziness. Parents should not spoil their children; if they do, the children will never learn to take care of themselves.
Thus, in answer to the second question – why can’t people get along? – Lakoff argues that the lack of communication and reconciliation is due to a fundamental contrast in world views. Liberals and conservatives navigate the world and make sense of what they see with different conceptual systems.
So is the Nation Truly a Family?
Lakoff’s model of “family politics” has enjoyed considerable circulation. In particular, several Democratic campaigns have used Moral Politics – and Lakoff’s later book, Don’t Think of An Elephant – as “political Bibles” of a sort. Conceptual Metaphor Theory has trickled into the discourse of sociology, political science, and even medicine. This type of impact extends far beyond what your typical linguist generally expects to achieve. So the question must be asked: is Lakoff’s model of politics in the USA “right”?
First, a brief caveat (scientists love caveats). There’s a famous saying in science that goes: “All models are wrong, but some are useful”. The origin of the saying is generally attributed to George Box, a 20th-century statistician. Box’s point is that one’s concern with a model should not be that the model is “true”, since any model of a phenomena is, by definition, a reduction of that phenomena – and thus is not the “whole truth” of the phenomena. Only the phenomena itself is the “whole truth”. In other words: the map is not the territory. Thus, when we develop models, our primary concern should be whether they provide any sort of useful insights into the phenomena that could not be obtained without the model.
Back to Lakoff’s model. Does it provide us with insight into the fractalization of American politics?
I think so. It’s certainly true that the prototypical perspectives of liberals and conservatives diverge rather radically when it comes to the “ideal” size and influence of the government. Lakoff’s explanation of this divergence in terms of differing views on child development and morality is pretty convincing to me. A liberal supports social programs like universal healthcare and “safety nets”, while a conservative argues that they disincentivize hard work and encourage over-reliance on the government. If you look at the metaphors used by certain conservatives, you’ll see these family references emerging:
Sucking on the teat of big government.
They just want to institute a “Nanny” government.
And, like many good models, this framework naturally begets more questions, such as:
1. How exactly do these idealized views of the family get formed in the first place?
2. Can the conceptualization of FAMILY itself be explained in terms of an even more fundamental experience? Would such an explanation be useful?
3. Do some people have different views on ideal family structure and ideal government structure? That is, are there extreme “nurturant parents” with equally extreme conservative views?
On the other hand, that doesn’t mean that the model is without flaws. One very immediate and obvious criticism is that the model’s framing is biased towards the progressive point of view; Lakoff is openly progressive, and it’s hard to read his model without picking up on an implicit criticism of the conservative perspective. In the interest of full disclosure, I also consider myself a “liberal”, but I’d be more comfortable with a less obviously biased model.
The other criticism is a little subtler. Lakoff’s model makes an assumption that pretty much all contemporary discourse about politics makes: the political spectrum is fundamentally two-dimensional, with progressives on one end and conservatives at another end. People can fall anywhere along this spectrum, or even the middle – “biconceptuals”, as Lakoff calls them – but their attitudes can still be explained by the two prototypical views. This is a reasonable argument, but it fails to explain several important political views. Namely, it’s not clear to me where the libertarian platform fits along this spectrum. Sure, a faith in laissez-faire capitalism, and the inverse distaste for big government, seems to go hand-in-hand with the strict father view that welfare programs only serve to encourage laziness. However, libertarians tend to have much more open views than conservatives regarding social freedoms, such as the legalization of drugs and sexual orientation; this seems more in line with the “nurturant parent” belief that individuals should be free to pursue their own freedoms, as long as they don’t impinge on others’ freedoms.
It also fails to explain a subset of “progressive” beliefs. For example, I favor strong government regulation and social programs because I don’t have faith that the free market will take care of these problems. To me, a faith in the free market overlooks the fact that bad things can happen when many individuals are out – with perfectly good intentions – to make a profit for themselves, without broader societal aims in mind. In a way, this is a somewhat pessimistic view of human nature. Instead, I think that the government should provide a regulatory system to make sure basic needs are met and people have access to opportunities. This is probably closer to the “nurturant parent” view than the “strict father” view, but I think the “nurturant parent” view offers a more optimistic perspective on what society would look like, if humans were left unconstrained.
I’ve tried to summarize both Conceptual Metaphor Theory and Lakoff’s model of family politics. A more serious discussion of his political model would apply it to an observable phenomena, such as contemporary discourse. Regardless of your stance on Lakoff’s model of politics, I hope that the discussion of metaphors and their pervasiveness in everyday life and language use has made you consider just how it is you conceptualize the world.
Lakoff, G., & Johnson, M. (1980). Conceptual metaphor in everyday language. The Journal of Philosophy. http://doi.org/10.2307/2025464
Thibodeau, Paul; Boroditsky, L. (2011). Metaphors We Think With: The Role of Metaphor in Reasoning. PLoS ONE (2011), 89(3), 493–510. http://doi.org/10.1371/journal.pone.0016782
Charteris-Black, J. (2006). Britain as a container: immigration metaphors in the 2005 election campaign. Discourse & Society, 17(5), 563-581.
Flusberg, S. J., Matlock, T., & Thibodeau, P. H. (In Press) Metaphors for the war (or race) against climate change.
Hubel, D. H.; Wiesel, T. N. (1959) Receptive Fields of Single Neurones in the Cat’s Striate Cortex. J. Physiol.
Del Tredici, M., Nissim, M., & Zaninello, A. (2016). Tracing metaphors in time through self-distance in vector spaces, 1–6. Retrieved from http://arxiv.org/abs/1611.03279
Stefanowitsch, A. (2004). Happiness in English and German: A metaphorical-pattern analysis. Language, Culture, and Mind, 137–149.
Lakoff, George. (1996) Moral Politics: What Conservatives Know That Liberals Don’t. University of Chicago Press.
 Cartoon taken from http://wearechange.org/new-study-elephants-dont-care-income-inequality-donkeys/. The cartoon happens to be very relevant to this blog post, given its concern with progressive/conservative conflict, and the fact that its humor is derived from what linguists call a “conceptual blend” (as many political cartoons are) – that is, the “boxing” space is combined with the “political disagreement” space, which is also combined with the “animals as stand-ins for political platforms” space. Conceptual Blending Theory was developed as a formalized offshoot of Conceptual Metaphor Theory, so I found it a fitting mention.
 Note that in the scientific literature, domains or “frames”, as they are called, are typically capitalized, as they are here. I experienced some hesitation regarding whether to follow this custom (fearing that it would alienate some readers), but decided in favor of following the established notation.
 A frequently-cited example is the metaphor QUANTITY is VERTICALITY, also called “More is Up”. This metaphor is so engrained in language and culture that it hardly seems metaphorical at all: “stock prices plummeted/skyrocketed”, “that’s a very high number”, and so on. The VERTICALITY domain is probably very embodied, considering there’s evidence that neurons in V1 (the primary visual cortex) fire selectively according to the verticality or orientation of stimuli in their receptive fields (Hubel & Wiesel, 1959), and language about VERTICALITY is used to describe a number of other domains, including ABSTRACTNESS (“high-level”) and MORALITY (“that’s a low blow”). As a pretty much caveat, however, note that the explanations for the neural mechanics of metaphor are mostly conjectural. They’re grounded in knowledge about the brain, of course, but research investigating the neural basis of metaphor is only just beginning.
 He’s primarily speaking of mathematical models, but I believe that the point can be extended to philosophical frameworks more generally. In fact, I intend, in a later post, to draw an analogy between statistical models and forms of literary theory, in that both attempt to understand something about the world or a set of data (or literary text) through a reduction of some sort, with the aim of providing insight.
 Though it’s hard to determine whether this phenomenon in society is itself a function of its categorization.
 Which, I should note, goes quite a long way – probably the furthest in contemporary science – in explaining the origins and structure of high-level human cognition.