The Moon Is Always Half Lit

The Sun lights one hemisphere of the Moon. Not usually. Not on average. Continuously, for four and a half billion years, without a single interruption that was not an eclipse.

This is worth sitting with, because almost nobody pictures it that way. A crescent Moon looks like a Moon that has had most of its light taken away. It is not. It is a fully half-lit Moon seen nearly edge-on to its own daylight — you are looking at the boundary between its day and its night from almost alongside, so the lit part presents to you as a sliver.

The illuminated hemisphere is exactly as large at new Moon as at full. Nothing dims. Nothing is withheld. Nothing is recovering. The same amount of Moon is in the same amount of sunlight the entire month, and the entire month is a story about where you happen to be standing.

One refinement, because precision is a form of care. Slightly more than half is lit — about 50.2 percent. The Sun is not a point and it is much larger than the Moon, so the tangent lines that define the terminator close a little past the geometric halfway line. It is a small number and it is the honest one, and we show the arithmetic in the colophon rather than asking you to take it.

Three bodies are required to make a phase: the Sun, the Moon, and somebody looking.

Remove any one of them and the phase does not become something else — it stops existing. There is no fact of the matter about "what phase the Moon is in" that does not include an observer's position in it. The quantity astronomers actually compute is the angular separation between the Sun and the Moon as seen from where you are, and that number has your location baked into it as surely as it has the Moon's.

So the month-long procession of shapes is not the Moon doing anything. It is the Moon and the Earth carrying you around a lit object at a steadily changing angle, and your line of sight sweeping across a boundary that, on the Moon itself, has not moved at all.

Figure 1 is the standard diagram with one thing insisted upon: every Moon in the ring is drawn with the same lit half, facing the same way. That is not an artist's simplification. That is the finding.

Everyone has seen the chart: eight little Moons in a row, eight names, eight equal boxes.

The chart is a convention laid over a continuum, and it is lopsided in a way the drawing hides. The US Naval Observatory is precise about this. New Moon, First Quarter, Full Moon and Last Quarter are defined to occur when the excess of the Moon's apparent ecliptic longitude over the Sun's is 0°, 90°, 180° and 270° respectively — which makes them instants. Moments. They have no duration. The Moon is Full at a specific second and is not Full a second later.

The other four — waxing crescent, waxing gibbous, waning gibbous, waning crescent — are the entire rest of the month, about a week each.

Underneath the chart the illuminated fraction changes smoothly and without steps, every value between nothing and everything, all month, every month. There is no moment at which the Moon crosses from one kind of thing into another kind of thing. The boxes are ours. We drew them because a chart needs boxes, and then we started reading the boxes back as though we had found them.

This is not an argument against the names. They are useful, they are precise where they claim to be, and the USNO states their status plainly rather than hiding it. It is an argument about what happens when a naming convention outlives the page it was drawn on.

Because you are one arm of the angle, moving you changes the answer.

Take a single instant — one waxing crescent, one illuminated fraction, one Moon. In Edinburgh it is lit along the right-hand edge. In Melbourne, at the same moment, it is lit along the left. Near the equator it rides low as a bowl, lit along the bottom, and there is a lovely folk name for it: a wet moon, tipped up to hold the rain.

Nobody is wrong. Nobody is seeing an illusion. The Moon is not doing three different things. Your horizon is tilted differently relative to the plane the Sun and Moon lie in, so the same lit edge presents at a different roll angle — and since up is defined by where your feet are, the picture rotates with you.

If somebody handed you these three sketches and asked which was the real Moon, the question would be malformed — and yet it is exactly the question we ask about people constantly, in reports, in assessments, in the sentence but she seems fine with me.

The far side of the Moon receives exactly as much sunlight as the side we know.

It has days and nights of the same length. It has noon. And at new Moon — the one night of the month when we can see nothing of the Moon at all — the far side is in full daylight, the whole hemisphere blazing, entirely unwitnessed. The phrase "dark side" is not poetic licence. It is simply a mistake, and it names the wrong property: the far side is not dark, it is turned away.

We found out what was there on 7 October 1959. Luna 3 swung behind the Moon and took twenty-nine photographs of a hemisphere no one had ever seen, developed the film in a small automated darkroom it was carrying, scanned the negatives with a moving light beam, and radioed the result back across the vacuum. The pictures were grainy and covered about seventy percent of it.

What came back was not a shadow. It was a landscape — sunlit, crowded edge to edge with craters, and almost entirely lacking the great dark maria that make the familiar face familiar. It had been like that the whole time.

"The Moon always shows us the same face" is the second thing everyone knows about the Moon, and it is also an approximation that dissolves by about nine percent if you watch carefully.

We see roughly 59 percent of the lunar surface from Earth, not fifty. The Moon rocks, and it does so for two unrelated reasons that happen to compound:

So the lock is real but it is loose. It is a tendency the system settled into, not a bolt. The Moon is very slightly out of step with itself all month, and the reward for noticing is nine percent more world.

Stand on the Moon on the night we call new, look up, and Earth is full — a complete blue-white disc, four times as wide as the Moon looks to us, and far brighter.

This holds all the way round. The phases are complements. When we have a thin crescent, they have a fat gibbous Earth. When we have full Moon, they have new Earth, a black disc ringed in red by every sunrise and sunset on the planet at once. At every instant, the two phases sum to one whole.

Which means the light is never missing. It is only ever distributed differently between two vantage points. Nobody in this system is having less; somebody is having it elsewhere, and the arithmetic is exact.

This zine is about light. The other half of the relationship — nearness, tides, and what a body close by does for another body's rhythm — belongs to No. 17, The Nearest Body, and is deliberately not re-walked here.

Find a thin crescent a few days after new Moon, on a clear evening, and look at the part that should be nothing. It is not nothing. The whole round disc is faintly there, ash-grey, the dark part visible.

That glow is our light. Sunlight strikes Earth, bounces off cloud and ocean and ice, crosses to the Moon, and comes back to your eye. English has an old name for the sight: the old Moon in the new Moon's arms.

Leonardo da Vinci worked out the mechanism around 1506–09 and drew it in the Codex Leicester, and — being Leonardo — never published it, so it did nobody any good for a century. He also got it slightly wrong: he thought the light came off Earth's oceans, where in fact most of it comes off clouds. The first published explanation is credited to Michael Mästlin.

Mästlin taught at Tübingen, and among his students was Johannes Kepler — who, a few years later, ran out of vocabulary while drawing eclipse geometry and had to invent a word for the region that is neither full shadow nor full light. He called it penumbra. That story is No. 51.

And note what earthshine actually requires. The dark limb is visible because something else is lit and facing it. Nothing about the Moon changed to make it appear. It became visible because of what was turned toward it.

The Moon completes one full orbit of the Earth — a whole circuit, 360 degrees, measured against the fixed stars — in 27.32 days. That is the sidereal month, and it is the orbit the Moon is actually on.

But it is not back to new Moon. Not close. Because while the Moon went round, Earth moved too, about 27 degrees along its own orbit around the Sun, and "new Moon" is defined by the alignment with the Sun. The target slid forward. The Moon has to keep going — past a full circle, roughly 29 degrees further, about two more days — to catch an arrangement that did not stay put.

That gives the synodic month: 29.53 days. It is the month everyone actually lives by. It is the one calendars are built from, and it is longer than the orbit it is made of.

There is a whole second piece in this — 29.53 days will not divide 365.24 days, and every calendar humans have built is a patch over that refusal. That is a chain for another day, and it is not this zine.

Here is the whole of it: waxing and waning are words about the geometry. They are not words about the Moon, and nothing in the Moon does them.

We are described in phases constantly. Verbal on Tuesday and not on Thursday. Fine at home and struggling at school, or the reverse, which is usually taken as more damning. Capable in the assessment room and not in the corridor outside it. Someone writes down the sliver they were standing at the right angle to see, and it enters the record as how much there is.

Those are not two quantities of person. Nothing was withheld in the first case and released in the second. The whole hemisphere was lit the entire time. What changed was the angle, the conditions, and who was doing the looking.

And the observer is in the measurement — that is the part that keeps getting dropped. But she seems fine with me is not a fact about her. It is a report of one angle, offered as though it were the object. So is he never speaks. So is every profile assembled from people who each saw one edge and none of whom were asked where they were standing.

The spiky profile that No. 21 is about is exactly this, seen from inside: not a person who fluctuates in amount, but a whole person meeting differently-angled conditions and being measured as though the conditions were neutral.

The previous spread is an argument. Here is what it looks like from inside one school career — Ryan Boren, writing in 2022 about being read at both extremes at once.

He is hyperlexic and Autistic. School did not know he could read, because he would not read aloud: situational mutism and exposure anxiety, in his words, “locked my lips.” Then one day, in a quiet corner with what he calls a semblance of psychological safety, he read a book cover to cover to his teacher in a whisper, and asked: “Can I go now?”

Note what actually changed, because it was nothing about him. The reading was there the whole time — the entire lit hemisphere, at full brightness, before the quiet corner and after it. What moved was the room. And he names the variable exactly:

Then the sentence this whole zine has been building toward without knowing it. He describes going “from being a full time r-word to a part time one, depending on who was in the room with me and if I felt safe enough speak.”

Full time to part time. That is a phase, recorded in the vocabulary of proportion — as though there were an amount of him, and it had gone up. And depending on who was in the room with me is not a metaphor for the observer's position in the angle. It is the observer's position in the angle, stated plainly by the person being measured.

He goes further than we would have dared, and he is right to: “Autistic people are psychological safety barometers.” Which inverts the whole measurement. A barometer's reading is a fact about the air around it. If the apparent phase tracks the safety of the room, then the instrument was measuring the room the entire time — and the room wrote the result down as a fact about the child.

There is an easy, wrong version of this zine, and it is one sentence away from the right one. The easy version says: since the Moon is always half lit, the difference is imaginary, so nothing is really harder on Thursday than it was on Tuesday.

That is false, and it would be a cruel thing to hand anybody. A crescent genuinely delivers less light to you. Measurably. A full Moon can light a path home; a crescent cannot. The geometry is not a trick and its consequences are entirely real for whoever is standing under it.

The correction we are actually making is narrower and holds up better: the difference is located in the arrangement, not in the body. Not in the Moon, which never changed. In the angle, the conditions, the room, the lighting, the hour, and the position of whoever is writing it down. That is exactly where an arrangement can be altered — and exactly where a diagnosis cannot reach.

This is the same move No. 51 makes with shadow, and we are borrowing it rather than re-deriving it: an umbra is a relation between a source, an obstruction and a place, never a property of the ground it falls on. Same here. A phase is a relation between a source, a body and an observer, never a property of the body.