Inventing magnitude: how bright is a star 

Position tells you where a star is. It says nothing about the thing that strikes anyone who looks up: some stars are blazing and some are barely there. To make that an entry in a table you need a scale, and a scale requires deciding how many steps the eye can honestly distinguish.

The answer that came down from antiquity was six. The brightest stars were of the first grade, the faintest visible ones of the sixth, with four grades between. In the Almagest, Ptolemy lists every star in his catalog with such a grade, sometimes adding a qualifier for a star judged slightly greater or less than its class. Tradition, and Ptolemy's own dependence on earlier work, attribute the scheme to Hipparchus, though because his catalog is lost the attribution rests on later testimony rather than a surviving text.

The scale is a piece of practical psychology as much as astronomy. It does not measure light; it sorts appearances into ranks a trained observer can reproduce. That makes it robust. Two people with no instruments, working centuries apart, can broadly agree on whether a star is second or fourth grade, which is precisely what a catalog intended for posterity needs.

It also turned out to match how human perception works. The eye responds to ratios rather than differences in brightness, so equal steps of apparent grade correspond to roughly equal multiplying factors of light. In 1856 Norman Pogson made this explicit, defining the scale so that five magnitudes equal a factor of one hundred in brightness, which puts each single step at a factor of about two and a half. The ancient grades and the modern definition line up reasonably well, which is a striking result for a system built by eye.

The one lasting inconvenience is the direction. Because first grade meant best, the numbers run the wrong way: smaller magnitude means brighter object. Extend the scale with measurement and it goes negative for the most brilliant objects. Sirius sits at about minus one and a half, Vega near zero, the faintest stars visible from a dark site around six, and modern telescopes reach far beyond thirty. Every astronomy student meets this backwards arithmetic and every one of them is inheriting a decision made on Rhodes.

It is also why brightness estimates in an old catalog are scientifically useful and not merely decorative. A star recorded at second grade that now looks fourth is a question worth asking, and a star listed where nothing bright stands today is a puzzle for someone. Hipparchus, if Pliny is right about the new star, built the scale for exactly that purpose.

The change he actually detected, however, was not in a single star. It was in the entire sky at once.