Nobody has seen everything. Not you, not me, not any scientist. What we can do is check things that anyone can check, and trust the ones that many different people tried to break and couldn't. This page shows you how that works, and gives you experiments to do yourself.
"Have you been to every body of water?" No. But water boils at 100 °C at sea level in every kitchen, lab and mountain-top test anyone has tried (a little lower the higher you go, in a way we can predict). If someone found a lake where it didn't, they'd be famous, because science rewards people who find mistakes. Knowing doesn't mean seeing everything. It means the claim can be checked by anyone and has survived people trying to knock it down.
1. Your own eyes are good, but not perfect
"Did you see it with your own eyes?" sounds like the strongest question there is. But careful studies show that seeing, and remembering what we saw, is less reliable than it feels. That is true for everybody: atheists, believers and scientists.
In 1999, Daniel Simons and Christopher Chabris asked people to count how many times basketball players passed a ball. In the middle of the video, a person in a gorilla suit walked through, stopped, thumped its chest and left. About half of the viewers never saw the gorilla. They were looking right at it.1
In 1974, Elizabeth Loftus and John Palmer showed people films of car crashes. Those asked how fast the cars were going when they smashed into each other guessed about 41 mph. Those asked about cars that contacted each other guessed about 32 mph. Same film, different memory.2
This is exactly why science doesn't depend on one person's eyes. It uses instruments that don't get distracted, many observers, and written records that others can re-check. A claim that rests only on "I saw it" can be true, but it can't be tested the way a boiling-point or a shadow can.
Why does science not rely on a single person's eyewitness account?
- Because people always lie
- Because even honest people can miss things or remember them wrongly
- Because eyes can't see anything useful
2. Three levels of "I know this"
Everyone, including the most careful scientist, relies on other people for almost everything they know. You've never seen the inside of your own brain, or the country of Japan, but you believe both exist. The real question isn't "did you see it?" It's "how good is the reason to trust it?"
| Level | What it looks like | Example |
|---|---|---|
| 1. I tested it myself | Strongest for you. But only for what you can reach. | You drop a ball and watch it fall. |
| 2. Anyone can test it, and many did | Strong. Different people, countries and rivals got the same answer. | The speed of light, measured by many labs for over 150 years. |
| 3. Many different clues point the same way | Strong, even for things in the past you can't re-run. | A detective with fingerprints, DNA and a camera recording. |
| 4. One person said so | Weakest. It may be right, but you can't check it. | "My friend's cousin saw it." |
Claims of miracles are, by their nature, not repeatable on demand. That doesn't make them false, but it does put them in the weakest level, because there is nothing for a stranger to re-run. That's the reason scientists treat them differently from boiling water.
3. "But nobody saw the past happen"
True. Nobody watched the dinosaurs die or the Big Bang. But detectives convict people of crimes they didn't see, using clues that can't all be wrong at once. Science of the past works the same way: it makes a prediction about what we will find, then goes and looks. When Tiktaalik, the fish with wrists, was found in rocks of exactly the predicted age, in exactly the predicted place, that was a prediction that could have failed and didn't.7 See evolution and dating methods.
4. "Has the universe really been proven?"
This is a fair question, and the honest answer is more careful than a yes. Science doesn't "prove" things the way a maths problem is proved. It builds confidence, by checking many independent clues. For the Big Bang, there are several, and they were found by different people using different tools:
- The universe is expanding. Distant galaxies move away from us, and the farther away, the faster (Hubble, 1929).3
- Leftover glow. In 1948, scientists predicted that the early universe should have left a faint glow in every direction. In 1965 two engineers, Penzias and Wilson, found it by accident while trying to get rid of static in an antenna.4 A prediction that came true.
- The right mix of atoms. The Big Bang predicts that the universe's ordinary matter should be about three-quarters hydrogen and one-quarter helium. That's what we measure.
What we do not know: what, if anything, came "before", and why there's a universe at all. A good scientist says so out loud. Read more in What the Big Bang Really Says.
5. "How did the human body, and organs like the eye, even come to exist?"
This is the question that comes up most, and it deserves a real answer. In short: slowly, in many small steps, each one useful, over a very long time. The eye is a good example, because we can see every stage alive today:
- A patch of skin that senses light and dark (some single-celled life and flatworms).
- The patch sinks into a little cup, so it can tell which direction light comes from.
- The cup narrows into a pinhole, which makes a rough image (the nautilus has this).
- A clear covering and a lens form, and the image sharpens.
In 1994, Dan-Eric Nilsson and Susanne Pelger built a computer model starting with a flat light-sensing patch and let it change by tiny amounts, with deliberately slow, cautious assumptions. It reached a camera-style eye in about 364,000 generations, a blink in geological time.5 It's a model, not a fossil record (soft eyes rarely fossilize), but it shows nothing impossible is needed.
Honest limits: scientists are still working out the details of many organs. "We're still finding out" is not the same as "it can't be explained". See Evolution and Common Ancestor.
6. Try it yourself: eight tests
Here are things you can check with your own eyes and simple stuff. Try one this week. If you get a different answer from the one we say, that's worth knowing. Tell us on the contact page.
Around 240 BC, Eratosthenes did this with a stick. On the same day, at local solar noon (when the Sun is highest), you and a friend in a town a long way north or south each stand a stick straight up and measure the angle of its shadow. The difference in angles is how much the ground curves between you. Then:
circumference = distance between you × 360 ÷ difference in angle
Try it with his numbers, or your own:
Result: about 40,000 km. The real number is about 40,000 km.
A flat Earth can't give this result: on a flat Earth with a far-away Sun, shadows would match everywhere. See Flat Earth vs. Science.
On a clear night, find Jupiter (a very bright "star" that doesn't twinkle). Steady ordinary binoculars (7×50 or 10×50) on a wall or a tripod. You'll see up to four tiny dots in a line next to it. Look again the next night: they have moved. Galileo saw this in 1610 and it showed that not everything circles the Earth.
On a dark, moonless night away from city lights, find the Andromeda Galaxy, a faint fuzzy smudge. A free sky app will point you to it. It's about 2.5 million light-years away, so the light reaching your eye left it about 2.5 million years ago, long before our species existed (about 300,000 years old).
Take the turntable out of a microwave (or stop it spinning), put in a flat plate of chocolate or marshmallows, and heat for 15 to 20 seconds, until a couple of spots start to melt. Don't eat it. Measure the distance between two neighbouring melted spots. That's half the wavelength of the microwaves. Most microwaves run at 2,450 MHz (check the label on the back).
Result: about 299,000 km per second. The accepted value is 299,792 km per second.
Speed = wavelength × frequency, and wavelength = 2 × the gap. Expect to be within about 5 to 10%. Only run a microwave with food in it.
The International Space Station is about 400 km up and circles the Earth roughly every 90 minutes. On NASA's free Spot the Station page, enter your town and it tells you when a very bright "star" will glide across your sky without blinking. That's real people, up there, right now.
Mash one strawberry in a zip bag with a pinch of salt and a little water. Strain it, add a teaspoon of dish soap, then slowly pour cold rubbing alcohol down the side of the glass. White, stringy, cloudy stuff rises to the top. That's DNA, the same molecule in every living thing, including you.
Individual atoms decay at random, but big groups decay on a very predictable schedule. That's why radiometric dating works. Here, 100 dice stand for 100 atoms. Each round, every die rolls; any that land on 1 "decay" and are removed.
Atoms left: 100. Round 0.
Expect about ⅚ left after each round, so after about 4 rounds roughly half remain (that's the half-life). Run it twice: single results differ, the pattern doesn't.
Search "your state geological survey fossils". Many areas have public sites, road cuts and quarries where seashells, trilobites or plant fossils can be found, often sitting in layers, in the same order, that scientists predicted. Always follow local rules and ask permission on private land.
7. A note on tests nobody can run
Some questions can't be tested with a stick or a microwave, such as whether a god exists, what happens after death, or what makes a life meaningful. For those, it's fair to say "this isn't something evidence can settle", and people can hold different views. The goal of this site is narrower: for claims that do make checkable predictions about the world (the age of the Earth, a global flood, how species arose), look at what the checks actually show.
8. Watch it explained
Claims that anyone can test, and that have survived many tries by rivals, are the ones we have the best reason to trust. Eyewitness memory, though honest, is easily changed.
"If you didn't see it with your own eyes, you can't really know it."
Many believers say, fairly, that scientists also take things on trust, and that a personal experience of God or of a healing can feel as real as any measurement. That's true: all of us trust other people for most of what we know. The reply is that trust works best when it can be checked. A kitchen test can be repeated by a stranger; a private experience, however sincere, can't. That doesn't make the person dishonest. It just means the two kinds of knowing are different, and it's worth knowing which one you're using. See why we believe and healing and prayer.
Sources
- Simons & Chabris, "Gorillas in our midst: sustained inattentional blindness for dynamic events", Perception 28 (1999).
- Loftus & Palmer, "Reconstruction of automobile destruction", Journal of Verbal Learning and Verbal Behavior 13 (1974).
- Hubble, "A relation between distance and radial velocity among extra-galactic nebulae", PNAS 15 (1929).
- Penzias & Wilson, "A measurement of excess antenna temperature at 4080 Mc/s", Astrophysical Journal 142 (1965). Alpher & Herman predicted the background radiation in 1948.
- Nilsson & Pelger, "A pessimistic estimate of the time required for an eye to evolve", Proceedings of the Royal Society B 256 (1994).
- NASA, Spot the Station.
- Daeschler, Shubin & Jenkins, "A Devonian tetrapod-like fish and the evolution of the tetrapod body plan", Nature 440 (2006).