Right -- which is related to thinking in percentages (I almost wanted to say thinking in probabilities, but it's not quite that... it's more fundamental.)
For instance, the age-old question of nature vs nurture. In debates, the answer always into 100% one or the other which makes for a good fight to watch, but in real life we know it's both, and with differing percentages in each situation.
Wading into a slightly more controversial category -- gender pay gaps. Is the pay gap for the same position 100% consistently due to gender? (Twitter will have you believe "yes" but the answer is more complex). Most of us know that pay gaps are due to a series of factors of which structural discrimination based on gender is only one (a significant one, and one that needs to be rectified). But there are other factors like job performance, compensation negotiation skills, visibility of work (results don't always speak for themselves, there's often a need to sell), all factors occurring in different percentages in individuals/groups. Understanding these other factors and the proportions to which they occur and help in devising practical interventions to address them in order to have sustainable equity over the long term.
If we can only switch our mindset from binary categories to multinomial ones, and assign percentages to the contributions of each, we can make moves toward finding solutions instead of just being angry at each other, which doesn't really lead to lasting change.
Let's say pay gap is 5% "anti-black racism" when it comes to a black man being paid in software development. Pay gap is also 40% job performance, 20% visibility, 15% negotiations, 10% unknown, 5% underpaying due to being black.
What happens if visibility is 10% antiblackness, due to a manager have just a small amount of bias when delegating visible work? Does this increase the 5% antiblackness by the 10% of 40% (to like 5.4%?)
EDIT: I'm not trying to be socially just in this analysis, so we can also replace being black with being openly conservative for example, or being from a different college, or not liking the same sport or same beer as one's peers. Pretty much anything can fit into this, I'm just trying to analyze this viewpoint.
y ~ x_1 + x_2 + ... + x_1 * x_2 + x_2 * x_3 ...
etc. where you have interaction terms like x_1 * x_2 (say x_1 = being part of a non-dominant-culture, x_2 = gender) and so on. Then you can see stacking issues show up, e.g. if you're a woman you're -40% and if you're a woman of color you're -45% etc. (numbers are made up of course).
It's possible to get into more rigorous modeling methods like Bayesian networks (DAGs) and hierarchical regression models (where the coefficients are dependent) but I think the added rigor/refinement in most cases is unnecessary because it often washes out due to the inherent uncertainty in the data and in most cases does not add value to the goal at hand: that is to understand the major contributors to an issue in order to design an intervention.
If more folks could bring themselves to think in simple percentages, we would have moved the needle toward that goal.
For instance we know that fields dominated by women typically pay less than fields dominated by men. It's a factor in the pay gap overall.
Now, we could say that women chose those fields and those fields are just less desirable to the market and all that jazz.
However, there is something else that has been noted. When a field becomes dominated by women, the average pay decreases and when a field becomes dominated by men, the average pay increases. So the average pay of a field is also influenced by the group that dominates it.
Women aren't getting paid less because they take jobs like teaching and nursing. Teaching and nursing pay less because women dominate those fields.
So there are always factors upon factors upon factors and nothing exists in a vacuum. Visibility, negotiation, perceived job performance, and even the unknown are all influenced by the person being black.
Corollary 1: you can simplify your model of reality by dropping some of the weaker effects, and replacing a host of them by a simple constant. You can even just look at the single most important factor and ignore the rest when making estimates.
Corollary 2: You should always remember that your simplified model is simplified, and realize how much it is simplified. You should expect complexity of detail, and be ready to let go of the neat and simple view if you want a more correct view.
I think thinking in percentages is a good first-order approximation/heuristic to most problems. There's so much uncertainty in our data anyway so even just conceptually breaking down the issue to percentages of contribution would already substantially help us think more clearly with the goal being to identify the top contributing factors to address.
Adding Grey is good, unless you only think in Grey because that is only 1 color, and 1 color is less than 2.
We should all think in at least 3 colors. Which requires a basic understanding of statistics.
But what about thinking in colors that are actual colors? Like... at least the primaries. What type of education would be required to throw in some Blue or maybe Red?
"Black and white" can just as easily be "blue and red" thinking. Any binary choice works. "Gray" is just supposed to represent seeing things in a more analog fashion. You could replace it with "purple".
The problem arises when all you do is point out the gradients rather than make a decision based on the information given.
So the idea isn't to "introduce more colors" but to finally make a decision. "Yes, this is grey. But it's a really, really, really dark grey. It's mostly black with a tiny bit of white."
A good live example is gambling. You're often forced to make decisions based on incomplete information where every decision has some upside and some downside. You can't sit there hemming and hawing.
"A slippery slope; and so it all began..."
Ironically, dismissing any criticism of Trump as simply this knee-jerk reaction, the people using this phrase quite often make the same mistake they are trying to accuse others of.
but this depends merely on the prioritization of the factors that is assigned.
People who are benefited to an extent by Trump's impact/policies/actions have him at a light grey.
The next thing it leads to is a demonization of the group in light grey by the group in dark grey - instead of white and black.
why not the groups just agree to disagree and move on?
Some people abandon black and white and replace it with a single shade of grey [1].
In order to have effective grey thinking, I think you need at least a basic understanding of statistics. This is also a great defense against others who seek to bamboozle you.
___
1. https://www.lesswrong.com/posts/dLJv2CoRCgeC2mPgj/the-fallac...