Two dots, and the physics that put them there
Red-eye isn't a flaw in an old photo. It's a light beside the lens and a wide pupil in a dark room, and the geometry repeats identically every time.
Two silhouettes, side by side, on flat black. No faces — just the outline of a head and shoulders, twice. Where the eyes would be, four small hard dots of red, and nothing else in the image is bright.
Nobody posed for those dots. A light did.
What is actually printed here?
Two blank silhouettes and four red dots, nothing more.
There's no face to recognise and no scene behind either figure — just the shape a lot of indoor photographs took on for years, reduced to the one part of it nobody controlled. The dots sit exactly where a flash puts them: dead centre, one per eye, the same red every time.
Why do the two dots happen at all?
A light sits beside the lens, the room is dark, and the pupil is open wide.
In low light the eye opens as far as it can to gather what light there is. A flash mounted an inch or two from the lens fires straight down that open pupil, catches the blood vessels lining the back of the eye, and bounces almost straight back out along the path it came in on — so the lens records red instead of black. Move the light further from the lens and the angle breaks; that's the entire difference between a flash that does this and one that doesn't.
Wasn't this just a mistake?
Not one anybody could have avoided with that light in that position.
For as long as a small flash sat bolted an inch from a small lens — the same light source doing its work in the frame it also blows out behind the subject — the geometry stayed fixed and the result stayed identical. What people actually did about it came after the fact: a fine red pen sold specifically to draw over the two dots on a finished print, by hand, print by print.
Why does it always hit both eyes together?
Because both pupils face the same light at the same instant.
There's nothing selective about it — not a coin-flip per face, a consequence of geometry that applies to every eye pointed roughly at the camera when the flash fires. One figure gets two dots. A room full of people looking up gets two dots each, which is most of why a flash photograph of a full table reads the way it does, corner burned with the date the print carries in its own corner and every face in it lit the same hard way.
Whose face can this be?
Whoever's it is, rendered down to a silhouette, the question mostly answers itself.
A blank outline carries the geometry without carrying anyone's actual likeness, so the design doesn't need a real photograph behind it and doesn't take on the permission a real one would. A red pen and a real face are two different projects — putting colour back into an old print is the one that genuinely needs the original and somebody's say-so; this one only needs the shape.
Does it work for more than two people?
Yes — the dots scale with the row, not with the idea.
A third or fourth silhouette adds another pair, same red, same size. What changes is how the row reads at a glance: two figures look like a couple, four look like a family line-up, and the design doesn't have to say which — the dots do the same job at any length.
Does it print well?
About as simply as anything on the site.
Flat black ground, flat silhouette, four small dots in one bright ink — no gradient anywhere, no detail below the weave's floor to worry about. Near-black or true black garments hold the contrast best, and the red is the entire extra cost of the design; everything else is the shirt's own colour showing through.
How do I get one made?
Describe the row at JustOG — how many figures, and whether the dots sit exactly centred or slightly off, the way a real flash sometimes catches them.
Pick a direction, drag the crop frame, see it composited on the real garment, and it's made to order and shipped.
Designs other people have published are in the shop.
Four dots, red against black, and nobody in the room did anything wrong.