A question with a long, settled history
Some people today say that Earth is flat. It is worth stating plainly what the situation is: there is no genuine scientific debate on this point. The shape of Earth has been measured many times, by many independent methods, over more than two thousand years, and all of them agree that it is very close to a sphere.
The idea that people in the past believed in a flat Earth is also less simple than it sounds. Greek thinkers argued for a spherical Earth well before the common era, and educated people in medieval Europe generally knew Earth was round. The useful question is not who believed what, but what observations support the round-Earth conclusion. Many of them you can check yourself.
Observations you can make from the ground
Ships and coastlines. A ship sailing away does not just shrink; its hull disappears first and the masts last, as if it were going over a hill. On a flat surface, a ship would simply get smaller until it was too small to see. Using binoculars or a telescope can bring back the upper parts of the ship, but not the part hidden below the horizon.
The shadow in lunar eclipses. During a lunar eclipse, Earth’s shadow crosses the Moon, and its edge is always curved. A flat disk would, from some angles, cast an elongated or straight-edged shadow. Only a sphere casts a round shadow from every direction.
Different stars in different places. Travel north or south and the sky changes. Some constellations visible in the far south never rise for observers in the far north, and the height of the pole star above the horizon changes with latitude. On a flat plane, everyone would see roughly the same stars at once. The same effect is why certain seasonal constellations are visible from some latitudes and not others.
Time zones and daylight. When it is noon in one place, it is night somewhere else. A single Sun lighting a flat plane would produce a very different pattern of day and night than the one we observe. A spherical Earth rotating once a day explains the pattern simply, including the way sunrise reaches different cities in sequence.
Measuring Earth’s size
Around the third century BCE, the Greek scholar Eratosthenes compared the length of shadows at two Egyptian cities at the same time of year. The difference in angle, together with the distance between the cities, let him estimate the circumference of a round Earth. His exact figure is difficult to convert because ancient units are uncertain, but the method was sound and the result was in the right general range.
The same logic works in reverse. If Earth were flat, shadows measured at different places would imply different distances to the Sun that do not fit one another. On a sphere with a very distant Sun, the measurements agree.
Travel and technology
Ships and aircraft have circled the planet in different directions, and routes in the Southern Hemisphere, such as long flights between continents there, fit a spherical layout and are hard to explain on common flat maps. Navigation systems rely on this too: GPS navigation works by timing signals from satellites orbiting a spherical Earth, and the results are accurate enough for everyday use.
Photographs and video from space show the planet as a globe, and they come from many independent sources: different countries, companies and ordinary amateur groups tracking satellites. A person with modest equipment can watch the International Space Station cross the sky on a predictable schedule. Reaching it depends on rocket orbit calculations that assume a roughly spherical planet.
Why does it look flat?
Earth is large. Its diameter is roughly 12,700 kilometers, so the curvature over a small area is tiny. Standing in a field, you cover only a very small part of the sphere, and the ground appears level for the same reason a small patch of a basketball’s surface would seem flat to an ant. The curvature is real but only visible over distances of tens or hundreds of kilometers, or from high altitude.
Not a perfect sphere
Earth is, strictly, very slightly flattened. Its rotation makes it bulge around the equator, so the diameter across the equator is larger than the pole-to-pole distance by roughly 40 kilometers. That is a small fraction of the whole, which is why “a sphere” is a very good approximation. This flattening was itself predicted by physical reasoning and confirmed by measurement, which is the kind of check a good scientific model should pass.
Handling disagreement
People come to doubt mainstream science for many reasons, including mistrust of institutions or a wish to rely on personal experience. Those motivations deserve a respectful hearing. The best response is usually not to argue about who is trustworthy but to invite observation: look at a ship leaving harbor, watch a lunar eclipse, compare the stars at different latitudes, or track the Sun’s path through the year. Each of these is a test, and the round-Earth model passes all of them.
Science is strongest when it welcomes such tests. The shape of Earth is a case where the tests have been run so many times, in so many ways, that the answer is considered established.