It was the last straw. That’s what broke the camel’s back. While factually accurate, that statement isn’t any kind of truth. Let’s dig into facts, truth, and beyond.
Well, water
Where does water come from? Well, in fact, it comes from the faucet. Giving this answer to an adult would feel like we’re avoiding the question. It’s factual, it’s truthful, but the scope of this truth is really very limited.
City water comes from some reservoir somewhere, and it’s piped all the way to our faucet. Well-water comes from the ground, but we honestly can’t be more specific or technical without flipping over to Google. Both of those statements contain oversimplified knowns and acknowledged unknowns.
It seems to us that “it comes from the faucet” conveys a truth so limited in scope that it feels dishonest to the point of being untrue. On the totem poll of facts, truth, and beyond, this ranks pretty low. We like to call things like this fact but not truth.
Speaking in truths takes time and paragraphs. It takes thought, nuance, resources, and most of all, attention span. Speaking in facts takes really very little. You only need one example to call it a fact, and it can be conveyed in the width of a news headline. With the remaining but acknowledged unknowns, we feel comfortable calling the above broader statement a truth. On the spectrum of facts, truth, and beyond, our reservoir explanation is much better than pointing to the faucet.
Summarizing facts
Statistics is a useful field, and we’re lost without it. The challenge with statistics, often, is in knowing exactly what the statistic is giving us.
“Lies, damned lies, and statistics”
Popularized by Mark Twain
The usual procedure is to collect data (various observations) in an appropriate way (random and representative of a population) in order to reduce them to some kind of summarization (e.g. average) or conclusion (e.g. our expensive product is beneficial).
We see lots of ways statistics are misused. Data can be collected in various less-than-ideal ways (e.g. not random, not representative). Examples abound where data were completely fabricated in order to give the desired result. The summarization can be based on biased estimators or miscommunicated with misleading charts. Even the conclusions can be incorrect. It’s standard to require a p-value smaller than 5%, so even if everything is done correctly, then such a p-value gives the wrong conclusion literally 5% of the time if in reality the conclusion shouldn’t have been made (called a type-II error).
Maybe the most misleading aspect comes from the fact that negative results rarely get published. Given that type-II errors happen 5% of the time (when the result should be negative), then it’s entirely conceivable that half of the published literature in certain fields could actually be untrue. Science is big on reproducibility, and with infinite time and resources, other independently funded research groups would reproduce everything. Unfortunately, many statistical results are only needed long enough to cross some kind of one-time hurdle. Maybe worst of all, industry groups will often fund research just long enough to get the first positive result.
We’re lost without statistics. We don’t mean to bash it, really. The field of statistics is fantastic, but the practical implementation is really much harder. Researchers work in teams, and when they produce a result, they want it peer reviewed and reproduced elsewhere. Even then, when everyone agrees, the wording of the result has to be very careful.
Moving upward in facts, truth, and beyond
A single observation brought to your attention somehow is known as anecdotal evidence. It’s a single fact in isolation. It’s not hard to find a woman that’s taller than a man. Such pairings exist.
Many observations collected in an appropriate and random way can give something more. Done appropriately, a statistical sample can be used to find a truth. For the most part, it’s true that men are taller than women. It’s entirely true that the population average height of men is higher than the population average height of women. As we expand the scope of the statement, we move up the facts, truth, and beyond hierarchy.
Aside from misusing statistics on accident, misusing statistics on purpose, getting type-I errors, getting type-II errors, measuring the wrong thing, or asking the wrong question, statistics bridges the divide between fact and truth in many application domains. The softer sciences all need statistics to uncover the foundational truths in their fields, and without it, they only have a jumble of useless facts.
Building truth from facts is risky
We saw above that locating a shorter man and a taller woman gives us a fact. It’s a fact that such pairings exist. We also saw that it’s a truth that men are on average taller than women.
The scope of what the truth describes is much larger.
If we considered every pairing of men and women across the planet, we would have roughly 4 billion of each sex and approximately that number squared pairings. Some percentage of those pairings (let’s squint and say 20%) would have women taller than men. 80% of the facts would be in agreement in one direction while 20% of the facts would be in agreement in the other way. We have contradictory facts.
How can you assemble a bunch of disagreeing facts to produce a truth? Very carefully.
Given the available facts, we do our best to consider all models that explain them. The models with the fewest assumptions are the most likely, and those form our leading hypothesis. The scientific method attempts to disprove all untrue models, but as history shows time and time again, it’s an iterative process. Generally speaking at any given moment, we collectively believe what will eventually be understood as untrue or incomplete. Producing truth from facts is very hard. Producing truth from self-contradicting facts requires statistics in addition to all these other difficulties.
The proper model is our truth.
Pushing the limit of facts, truth, and beyond
If the difference between fact and truth is the scope of the statements, then what’s beyond truth? How much further can we broaden the scope? How much further up the facts, truth, and beyond hierarchy can any subject really go?
Trickle truth
Water comes from the faucet. Before that, the pipes brought it from the city reservoir. Previously, the rain deposited everything there, but only after it evaporated from somewhere else. And on and on we go. Long before that, the individual hydrogen and oxygen atoms came together to form a water molecule. The hydrogen atom was produced as part of the big bang, but much later the oxygen atom was discarded as spent nuclear fuel from either our sun or some other star. The subatomic particles producing those atoms pop in and out of existence all the time, so you might say they’re always recreated many many times per second.
Tall tales
Individual women can be taller than men. The average man is taller than the average woman. In reality, we’re all the same non-physical energy occupying virtual space-time suits for the sake of experiencing this physical format. None of us have height when we exist outside space.
The beyond part
If there’s any goal to this site, it’s to get us all to think in terms of broader truths. Expand our scope of awareness. Think in universal truths and beyond-universal truths. When necessary, challenge assumptions.
“Whenever you’re trying to understand anything that is as huge as all of the Universe, or as huge as all eternity, all you have to do is bring it back to something simple that you do understand, and ask the questions and apply them. And then you can understand the Whole. Everything that’s true of the Whole is true of the individual.”
Abraham, excerpted from Tucson, AZ on 2/20/01
Producing truths from facts is a fool’s errand. Sure, we get it right sometimes, but we conveniently forget all the times we get it wrong. We make the argument that it’s easier to begin with the universal truths and work your way down. The only challenge there is wrapping your arms around the universal truths.
