1. Half of the Moon Is Always Lit, Whatever Shape We See
NASA puts it plainly: the Moon makes no light of its own, and "moonlight" is reflected sunlight. The Sun always lights half of the Moon, and the other half stays dark. Like Earth, the Moon has a day side and a night side[1].
So a crescent is not a Moon with a piece missing. It is a fully round Moon of which we can see only a thin slice of the lit half. NASA describes the waning (shrinking) phases the same way: the lit side seems to shrink, but as the Moon orbits Earth, the part of its lit half that we can see simply gets smaller[1].
That already rules out the shadow idea in a rough way. But a good test needs a second question: where is Earth's shadow, and what happens when the Moon actually enters it?
2. How the Angle Between the Sun and the Moon Sets the Phase
Imagine looking down on the Sun, Earth, and Moon from far above. The Moon circles Earth, and the Sun's light always comes from the same side.
- New Moon. The Moon is in the same part of the sky as the Sun. Its lit side faces the Sun and its night side faces us, so we see almost nothing. A new Moon also rises and sets with the Sun, so it is up during the day but lost in the glare[1].
- Full Moon. The Moon is on the opposite side of Earth from the Sun, and we see its whole day side. A full Moon rises around sunset and sets around sunrise[1].
- Quarter Moons. Between the two, the Sun and the Moon are about a right angle apart as seen from Earth. We then see half of the lit half. That is why astronomers call it a "quarter" Moon, even though it looks like a half disc: it is a quarter of the whole Moon's surface that we see lit[1].
In between come the crescents and the gibbous Moons (the bulging shapes just before and after full). In all, NASA lists eight named phases, and the cycle repeats about every 29.5 days[1].
Notice what the full Moon's position means: it sits on the far side of Earth from the Sun, which is the direction in which Earth's shadow points. So why doesn't every full Moon fall into that shadow?
3. Why Isn't There a Lunar Eclipse at Every Full Moon?
If Earth's shadow made the phases, we would see a dark disc creeping over the Moon every month. In reality, a lunar eclipse needs the Moon to be near one of the two points where its orbit crosses the plane of Earth's own path around the Sun, and it needs to be a full Moon at that moment. NASA's eclipse catalog says exactly that: lunar eclipses can occur when the Moon is near one of its two orbital nodes during the full Moon phase[2].
The reason this is rare is that the Moon's orbit is tilted. NASA's Moon phases page gives the tilt as about 5 degrees[1]. That is small, but at the distance of the Moon it is enough for most full Moons to pass above or below Earth's shadow. A simple way to picture it is two hula hoops, one tilted slightly against the other: they cross at only two points.
The numbers match the idea. Over the 5,000 years from 2000 BCE to 3000 CE, NASA's catalog counts a minimum of two and a maximum of five lunar eclipses in a calendar year, and in 70.8 percent of years there were just two[2]. A year has about twelve full Moons, so most of them are eclipse-free, yet the Moon still goes through its whole cycle of phases every month.
This is the strongest evidence against the shadow idea. The phases happen every month, including months with no eclipse at all, so they cannot be Earth's shadow. A real eclipse is a rare, different event, and it is the only time Earth's shadow touches the Moon. How long does the phase cycle itself take?
4. Why a Moon Month Is Longer Than the Moon's Trip Around Earth
NASA gives two numbers that look like they should be the same. The Moon travels once around Earth in about 27.3 days (27.322, to be precise), yet it takes about 29.5 days to go from one new Moon to the next[1].
The gap is not in NASA's page as a calculation, so here is a short explanation that is our own reasoning. A phase depends on where the Moon is compared with the Sun, not compared with the stars. While the Moon circles Earth once, Earth also moves along its own path around the Sun, about one twelfth of the way around in a month, roughly 30 degrees. The Sun now sits in a slightly different direction in the sky, so the Moon has to travel a bit farther to line up with it again.
We can check the size of that extra bit. The Moon covers about 360 degrees in 27.3 days, which is roughly 13 degrees a day. Catching up the extra 30 degrees takes about 30 ÷ 13 ≈ 2.3 days, which is close to the actual difference of 29.5 minus 27.3, about 2.2 days. This is a rough check we made ourselves, not a figure from a source.
So the cycle of phases is tied to the Sun, not to the stars, and that is why it takes about 29.5 days. What about the half of the Moon we never see?
5. Does the Far Side of the Moon Have Phases Too?
The Moon spins exactly once each time it orbits Earth, so nearly the same face always points toward us. NASA calls this tidal locking, and it means most of the far side is never seen from Earth[1].
Put that together with the half-lit Moon, and a surprise follows. At new Moon, NASA says, the lit side faces the Sun and the night side faces Earth[1]. Since the near side is the night side then, the far side must be in sunshine. Reasoning this way (the combination is ours, not a sentence from NASA), the "dark side of the Moon" is not a place that is always dark. It is just a side we never see. It has days and nights like every other part of the Moon.
6. How an Ancient Greek Used a Half Moon to Try to Measure the Sun's Distance
Phases are not only pretty. Aristarchus, a Greek astronomer born around 310 BCE on the island of Samos, noticed that when the Moon is exactly half lit, the Sun, Moon, and Earth form a triangle with a right angle at the Moon. If you can measure the angle at Earth between the Sun and the Moon, geometry tells you how much farther the Sun is than the Moon[3].
The Linda Hall Library, which holds an early printed edition of his one surviving book, reports that Aristarchus took this angle to be 87 degrees and concluded that the Sun is 18 to 20 times farther away than the Moon. The Sun is in fact about 400 times farther. The real angle is about 89 degrees 50 minutes, almost exactly a right angle, and the library calls that an impossible measurement to make back then, "or even today"[3].
Why does a 3 degree slip matter so much? Near a right angle, a tiny change in the angle changes the answer enormously. Using our own quick check with a calculator, 1 divided by the cosine of 87 degrees is about 19, which matches the 18 to 20 that he found. The method was sound. The hard part was measuring the angle by eye.
7. How to Check the Sun-Moon Angle Yourself
- Make a model. In a darkened room, put a lamp on a table and hold a ball (a tennis ball is good) at arm's length. Stand with your back to the lamp's glare, so the lamp is beside or behind you. Turn slowly in a circle while holding the ball steady. The lit part of the ball goes from thin to half to full to thin again, even though the ball itself never changes. Do not stare at the bare bulb.
- Look at the real Moon. On an evening when the Moon is up, notice which side is lit. It should face toward where the Sun has set, even though the Sun is out of sight. Never look at the Sun directly, and never use binoculars or a telescope anywhere near it.
- Keep a Moon diary. Draw the shape at the same time each evening for a week and note when it rises. NASA's phase page explains that the Moon rises a little later each day[1]. Does your diary agree?
- Read the original idea. The Linda Hall Library page about Aristarchus (source 3) shows pictures from the 1572 edition of his book. You can try drawing the right triangle yourself.