#24: Part 2: Standard English Grammar is a Tool for White Supremacy

I can't stop thinking about AI; so much so that I spiral about metacognition, worrying about whether my thinking is my thinking or if it is influenced by AI. And I know that it isn't so much an if, but how much.

So when I read a review of Leslie Valiant's book The Importance of Being Educable, I felt validated. Valiant worked on computational AI, and in his book, he contrasts AI's ways of knowing with humans' ways of knowing (sense perception, memories, imagination, etc.). Whereas AI is an amalgamation, humans have the ability to be choosier. Valiant writes that our discerning brains, “combine pieces of knowledge gained years apart” into “theories of considerable complexity that have many and disparate parts.” 

While reading it, I couldn't help but participate in the process that Valiant outlines. My mind wandered to two things that happened at the beginning of the pandemic. In 2020, I took an online course from Harvard for educators called Visible Thinking. Its intention was to teach teachers how to teach students how to think better. 

In the 1990s, education and cognitive researchers at Harvard wanted to figure out if there was a way to distinguish the skills that were being used when students were participating in “good thinking.” They gave students a number of tasks and they found that there are three elements that need to be working together whenever “good thinking” is happening: 

1. Ability: People have to have the skills necessary in order to complete the task or thinking process. This means that the problem that they are solving, approaching, or considering has to be solvable with the skills that they have. 

2. Inclination: In short, it is motivation. One must be inclined to use the ability or skill. I also think that being disciplined to experiment or attempt is factored into this. People have to have the drive to "leap" into a solution and take action in order to know if their thinking is right.

3. Sensitivity: On top of having a skill and being inclined to use that skill, people need to sense or be sensitive to the times when it is appropriate to use that skill. Sensing works beyond our five senses and is also rooted in the ability to react to a situation appropriately.

This made me think about a conversation that I had with a friend in 2020 about language and the ways that words like "sensitivity" and "fragile" were often used synonymously even though the meanings are very different. My friend argued that people who wore sensitivity like a badge of honor saw themselves as being morally superior, but were actually weak. 

If we believe the Harvard researchers, sensitivity is a weakness. But in the long explanation, it is (of course) a bit more complicated than just the weakness that my friend implied. In the research, the element that was most often missing from kids’ thinking was “sensitivity.” Students did not know when to use the skills that they had at the opportune time. If this carries into adulthood, then people are not usually good thinkers when they are given a conflict. They do not know which "thinking tool" to draw from their arsenal, even if they are motivated or inclined to solve the problem.

This does not validate my friend (or maybe, partially validates him), but it requires us to rethink what it means to be sensitive. Sensitive means that we are using our senses to react based on the amount of information that our senses collect for us. People need to wonder: is that what people are really doing when they claim that they are "sensitive" or are they being reactionary? What we think of as sensitivity may actually just be reactionary because people who claim that they are “sensitive,” should also know that when they use that word, it implies that they are discerning enough to choose the moments to be mad or angry or upset. Truly sensitive people are more motivated and disciplined to draw from a more dynamic set of skills and abilities than to react with their first emotional response. When confronted by negative feelings or a problem to solve, they would be able to sense when to ask more questions or consider another perspective or gather more evidence, or just sit the fuck back and observe in order to find complexity and, hopefully, clarity in their own thoughts. 

Valiant claims that when we pull from different contexts and moments to inform our actions, we are "educable" (not to be confused with intelligence). This educability is the differentiating factor between humans and computers. It, arguably, takes more time. If people practice being educable then I suspect that people could learn to trust their own thoughts and become much less reliant on a computer (and all the tools that we use) to do the thinking for us. As we become more reliant on technology (which often implies we are relying on others who are disconnected from our context) to tell us what to do, we diminish the thinking that it takes to arrive at those decisions and become increasingly dependent on technology to continue to do that work. 

It’s also what is so hard about education and speaks to the way that teachers are often diminished in what they do. We cannot exact learning to a formula and make it less mysterious, and I think society would value teachers more if we could neatly draw the line between our explicit teaching methods and procedures and outcomes. Teachers are being held to a standard that we hold technology to and true learning (not just reproducing knowledge) is far messier than we can show with a model, a formula, or a linear process. A worry that I have from learning from Valiant is that we have spent so much time and resources dedicated to making AI think like a human when we don’t really know the exact science of how humans actually do this process. From this, our reliance on technology implies that there is a desire to model our thinking from a computer. Yet, I’m not sure if we can even know what we want AI to do for us if we still haven’t figured out what learning is.