Wrestling With Ambiguity
The Bane of My Existence
I think a lot about what it means to have a belief about the world, and how we can better express those beliefs in ways that most accurately reflect our own knowledge. One of the ways you can do this is to express beliefs not as "I believe in X", or "I don't believe in X", but rather "I believe in X with 75% credence", where the 75% represents the strength of your belief, or the degree to which you believe the proposition X is true. The motivation behind this is that thinking in terms of these probabilities allows us to make more nuanced claims about the world that account for uncertainty, whilst also providing systematic ways to update our beliefs in light of new evidence.
I don't know anyone who thinks strictly in terms of true and false, but I also haven't many people who use probabilities to express their certainty in everyday conversation. Most people seem to live somewhere in between, where they have some internal model that gives them a fuzzy sense of how much they believe in something, but don't feel a need to come up with a concrete number. Now, you might ask:
Why is any of this important?
First, I think it's important to realize that human language is riddled with ambiguity. If I walk into a pizzeria and ask for a pizza with "pepperoni and mushrooms or olives," there are a total of four different ways you could interpret my pizza of choice, depending on where you place the parentheses and whether the "or" is inclusive/exclusive. Similarly, if I were to ask you the question "is college a scam?" or "should the U.S. intervene in foreign conflicts?", everyone is going to interpret these questions differently, it's not actually that helpful to reason about these types of ambiguous statements without a shared language for discussing them.
Second, I think the nature of this ambiguity is often much more semantic than ideological. Using the previous examples, it's not actually clear what we mean by the word "scam," or which "foreign conflicts" we're referring to, because there's no way to prove or disprove such statements since we can bend the words to mean whatever we want. The underlying sentiment behind the word "scam" might be that college is too expensive, that college perpetuates social inequality, that college fails to prepare students to enter the workforce, or that lots of people end up working jobs that don't require a college degree. But is it that we disagree about the existence of the scam itself, or the exact nature of the scam? Or are we talking about two totally different things? The people deserve to know!
Third, even when we do take the time to specify what we actually mean by the original question, it remains unclear how much credence we should assign to each proposed claim. Suppose we're trying to compare two statements about whether college is a scam: "college is too expensive for the majority of Americans," and "lots of people end up working jobs that don't require a college degree." Which argument should we take more seriously? For the former, we can measure things like how much attendance costs have risen over the past 10 years, how much debt the average graduate has 5 years after graduating, or the lifetime earnings from people with college degrees. Informed by these statistics, I might assign ~70% credence to this claim. For the latter, we might reason that while there are lots of people working jobs that don't require a degree, many of these people probably still benefited from having a college degree to begin with (since employers find this desirable). Knowing this, I might assign ~40% credence to this.
The beauty of this is that you and I can now engage in this systematic process of truth-seeking, which involves (1) starting with some claim about which we both hold beliefs, (2) bringing new evidence to the table, and (3) updating our beliefs accordingly. My credence might go up, or down, or stay the same, and that's okay—so long as I am completely detached from the status of being right. We're all just gradually updating our beliefs, hopefully converging on this elusive yet tantalizing concept that is truth.
The main takeaway is that we should strive to use precise language whenever we can, whether that's through disambiguating words or expressing certainty through probabilities. I'd much rather spend my time wrestling with ideas that we can think about concretely, measure quantitatively, and are open to falsification, rather than with vague words that shift in meaning to fit any and all perspectives.