Surface generalizations: A grammar antifa warning

Woody Guthrie is playing a guitar that bears a sign that says "this machine kills fascists." The title of the article appears, "Surface generalizations: A grammar antifa warning."
Click the image to be taken to the full Instagram post.

Every once in a while, I can’t help but nerd out about linguistics and its relevance to everything. In this article, I will share two distinctions from theoretical linguistics that can help us avoid the trap of relying on surface generalizations in our thinking about mental health and social change.

My undergraduate training in linguistics, a discipline that straddles the conventional divisions between the humanities, social sciences, and natural sciences, heavily influences the way I look at the world. My professor Jim McCloskey described the task of the theoretical linguist in the following way: we are attempting to create a descriptive model for how language is organized in the finite organ of the human brain that can account for the infinite creative potential that we each carry inside of us. This infinite creative potential is evidenced by the observation that there is no limit to the number of unique sentences that one can produce in any given language, and there is no upper limit to the word length of a sentence in any given language. Even if in actuality we are unable to create never ending sentences, or a never ending list of unique sentences in a language, it is only for lack of time and lifeforce, not for lack of grammatical engineering.

In order to fit infinite creative potential into a finite organic space, the descriptive “rules” (or generalizations) that make up a language’s grammar must be finite in number and must be capable of recursion (eg, a sentence must be able to contain another sentence). The requirement to have a finite number of rules behooves the theoretical linguist to look for generalizations and patterns that can efficiently describe a myriad of linguistic behaviors (utterances).

So how is this relevant to mental health or social change? Just like human language, the way our minds work and the way society works are very complex topics of study, and the quality of our understanding of them will necessarily depend on the quality of thinking that we bring to them. Noticing patterns and making generalizations about natural phenomena is a part of the scientific process, and is a natural human behavior, but not all generalizations are created equal.

As a theoretical linguist, if I were to say that English works like this and Spanish works like that, and Mandarin in this other way, these three generalizations might accurately reflect the three languages’ surface differences, but it doesn’t adequately explain how an infant can easily learn any of these three languages depending on the social context of their upbringing. If each language has a separate set of rules, requiring a distinct neurological framework for language acquisition, does that mean that every human infant is born with all of these neurological frameworks already installed for every possible human language? That’s a lot of mental real estate, especially when you consider the very real but untestable possibility that a prehistoric human infant raised in the 21st century almost certainly would have no trouble acquiring a language that didn’t even exist at the time of its birth. The only possible explanation is that there are deeper similarities among languages than what we are able to see on the surface, and that human beings require only one neurological framework for language in order to be able to learn any human language. Sometimes the generalizations we are able to make based on surface observations are not as accurate as generalizations that are based on a deeper analysis.

In mental health, an example of a surface generalization is that women experience depression at twice the rate that men do. This might be a very true statistic, on the surface. But what is the consequence of taking this surface generalization to be evidence of a deeper truth? Some people might conclude that this says something about women’s resilience or mental strength compared to men’s. This conclusion could then have an impact on how much respect men and women receive in society. A deeper analysis would look for other explanatory factors for the perceived differences in mental health between men and women. For instance, the influence of patriarchy.

There is no shortage of examples of surface generalizations being used to justify an unjust societal status quo. Just google “scientific racism” to get a taste. Would-be scientists and anthropologists and sociologists like Heather McDonald of the Manhattan Institute love to use (often cherry-picked) surface generalizations to draw conclusions about the supposedly higher rates of criminality among certain racial groups. The underlying message they are sending is that racial groups are biologically distinct, and therefore rankable in terms of value. A deeper analysis of the data would look at the very real impact that racism, war, genocide, enslavement, and other forms of historical oppression can have on human development over time.

So here are the two distinctions. First, descriptivism is more helpful than prescriptivism (grammar Nazism). Unless you are an elitist who believes that certain people are better than other people, it doesn’t make sense to tell people how they should be. What makes sense (and is more loving) is to do one’s best to understand how people actually are and work with that. The second distinction is “simpler is better,” and you get to simpler by collecting more data. If your description of humanity is chock full of exceptions and contradictions, then you probably need more data.


Nauser Bear is a licensed marriage and family therapist in California (LMFT#123099; licensed with the BBS as Nicholas Jon Reynolds). To work with Nauser, send an email to, call (510) 394-5373, or schedule a free 20-minute consultation by clicking here.

Disclaimer: This blog and comments on it do not constitute medical or mental health advice. If you are in need of support and live in California, contact me to set up a consult to see if working together is a fit. Otherwise, seek mental health support in your area. If you are in danger of hurting yourself or someone else, please call 911 or the National Suicide Prevention hotline, which is available 24/7, at 800-273-8255.

2 Replies to “Surface generalizations: A grammar antifa warning”

Leave a Reply

%d bloggers like this: