Meaning What We Say, Saying What We Mean
Recently on social media, I posted this Slate article by Sam Kriss criticising Eric Garland’s popular “Game Theory” tweetstorm, which has been shared widely by those on the left who continue to feel shaken and demoralised by the results of the recent US presidential election. I generally try to avoid making too many political posts on Facebook, but I felt compelled to share Kriss’s article primarily because it succinctly lays out many of the same criticisms that came to my mind as I finally got around to reading the half-baked, meandering, 127-tweet thread after seeing it shared by many friends and colleagues. Like Kriss, I was and remain troubled to see so many people I respect and admire eagerly latch onto an obvious conspiracy theory simply because it apparently offers some kind of political catharsis.
And it is a transparent conspiracy theory. Glenn Greenwald is a Russian agent? It’s suspicious that he lives in Brazil? What a traitor, making a life in the country that legally recognised his marriage as opposed to the one that would deny his partner residency.
But I digress.
Not long after I posted the article, a colleague of mine posted to disagree with Kriss’s criticism. Oddly, however, the point she wanted to contest was almost entirely peripheral to his argument. Toward the end of the article, Kriss writes as a background statement that “[d]ecades of neoliberal policy disenfranchised people to the extent that Donald Trump could look like a savior.”
Knowing this colleague to be an avid Clinton fan (way back to the primaries), I couldn’t figure out why she disagreed with that particular statement. It was only later in the conversation that I realised that my colleague was not familiar with the common meaning of “neoliberal.” Having taken it to mean the same thing as “liberal,” she ended up viewing the whole piece as criticism from the right rather than from the left.
Julia Galef recently observed on Facebook that political discussions seem to be prone to an unusually high level of misunderstandings. She attributes it in part to the high number of bad arguments in political discourse affecting our expectations to the extent that we resolve ambiguities in ways that conform to the bad arguments we have seen rather than the arguments actually being made. I think this is true, and I think the problem is exacerbated by the use of terminology that easily lends itself to misunderstanding by having multiple, sometimes contradictory, meanings. The word “liberal” is a good example, meaning as it does different things in different political cultures: it can refer to the Classical liberalism of anyone from Adam Smith to John Stuart Mill, to the American left-liberalism of FDR or LBJ, or the current neoliberal orthodoxy of globalised market capitalism. Not to mention that outside politics can refer to both liberal education and the liberal arts, or even a liberal pinch of salt.
This recent Atlantic article by Conor Friedersdorf discusses a similar problem in the use of the term “white supremacy,” which has both a conventional meaning used by the general public and a technical meaning used primarily by academics and activists. As the article shows, these two distinct meanings can lead not only to confusion, but also to criticism and disagreement in which the two sides talk past one another due to fundamental misunderstanding of each other’s arguments.
The conventional/technical dichotomy of meaning exists throughout social justice language (e.g. privilege), since many SJ terms originated in an academic discourse that seems to positively delight in coming up with new technical definitions for familiar vocabulary. These kinds of words, along with technical jargon, from the twin pillars of Academic Obscurantism: the oft-criticised and frequently parodied tendency of scholars to write impenetrable, inaccessible prose, presumably because it makes their arguments sound more sophisticated and profound.
I have long been an advocate for the use of plain language in scholarship, by which I mean actively avoiding the use of jargon, technical redefinitions of common words, and unnecessarily complex or convoluted syntax whenever possible. I use the word “active” here for a reason: a commitment to plain language requires effort. As we become increasingly familiar with certain kinds of discourse, it becomes increasingly easy to forget what it is like to be an outsider to that discourse or to the community that uses it. As a consequence, we can end up writing or saying things whose meaning seems perfectly obvious to us, but are nonetheless either impenetrable or prone to misunderstanding by an audience without the same background knowledge. Working against this tendency requires care and vigilance, but the effort is worthwhile: not only does the use of plain language help to reduce misunderstanding and democratise knowledge, but I also find it helpful in clarifying my own thoughts and ensuring that there are no weak points in my arguments that might otherwise be hidden in the mist.
Of the two pillars of obscurantism, I definitely think that multivalent (or polysemic) words of the conventional/technical type pose the greatest risk for misunderstanding. Technical jargon and other terms of art may hinder accessibility, but this is a problem that is for the most part easily solved with a dictionary or a Google search. The problem with conventional/technical multivalent terms is that they are familiar, and so they can be misunderstood without any awareness or evidence that a misunderstanding has taken place. Both the speaker and the listener find their different interpretations of the utterance to be perfectly clear and obvious.
Polysemic words don’t even require a second person to cause confusion: this is what philosophers and fallacy enthusiasts call equivocation. Equivocation is what happens when someone argues, for example, that feathers can’t be dark because they are light. While pretty much anyone can see the problem with that argument (because the two definitions of “light” are so different), equivocation becomes harder to see–and therefore more of a problem–as the multiple definitions approach one another in meaning. That’s how we get to “if murder is defined as killing a human, and a fetus is human, then abortion is murder.”
In addition to being prone to misunderstanding, polysemic terms also require vigilance because of their potential for abuse. Deliberate equivocation can be employed as part of a rhetorical bait-and-switch, a practice sometimes called the motte-and-bailey doctrine after the medieval castle design. A medieval motte and bailey was basically a well fortified stone or wooden keep (the motte) set within a larger, less fortified courtyard (the bailey). In the event of an attack, defenders would retreat from the bailey into the the motte, because it was easier to defend. Analogously, a person using the motte-and-bailey doctrine will make a broad or controversial claim using one meaning of a term, but upon receiving criticism will “retreat” to a narrower or less controversial argument employing a different meaning of the term.
This is the kind of argument creationists and certain kinds of theologians are making when they say “God created the universe” (by which they mean the God of the Bible), but then once confronted with scientific evidence retreat to a narrower definition of God as a “prime mover” or something even more vague. It’s also the kind of argument “dictionary atheists” are making when they claim that contemporary Atheism is not a belief system, worldview, or set of values, and refers only to a lack of belief in gods.
The particular danger in using polysemic terms in a politically or emotionally charged context is that when equivocation occurs it can be very difficult to tell whether that equivocation was an unintentional error in logic or a deliberate attempt to mislead via the motte-and-bailey doctrine. This is exacerbated by the fact that in those kinds of contexts the principle of charity is very unlikely to be invoked and ideological opponents are already primed to expect bad or misleading arguments. What started out as an error or miscommunication can therefore easily spiral into accusations of mendacity, bad faith, or worse.
I’d like to close this post with a short list of best practices for communication and argument that I try my best to follow, and that I hope others will try to follow as well. Obviously not everyone in the world is equally amenable to the idea of persuasion through careful, constructive debate, but for those who are:
- Carefully consider your audience before using anything other than plain language.
- If you are writing something on the internet, your audience is “the general public.” Even if you think you’re talking to a specific person or group, if anyone in the world can read or hear it then you ought to try to make it accessible to them, too–which means using plain language.
- Know thyself. You know better than anyone what circles you tend to run in, and therefore what kind of specialised terminology or shorthand you are likely to use without thinking.
- Define your terms. This can be as simple as a hyperlink, or a brief gloss in plain language the first time you use a term.
- Reread. Once you’ve written something, run through it again for points of potential misunderstanding.
- Consider paraphrase. Sometimes avoiding a loaded term can actually make things clearer, both for you and for your audience.
- Ask others to define their terms. When in doubt or disagreement, try to make sure you both agree on what’s being said.
- Reformulate. When questioning or disagreeing with someone, try to restate what they have said according to the above rules to ensure that’s what they’re saying. Maybe even steelman it if you’re feeling especially charitable. Note this is not the same as putting words in their mouth or accusing them of hiding a(n unsavoury) premise.
- If all else fails, find a cat. Petting a cat is very likely to improve the situation.