Bayer Letters and How Stars Get Their Names: A Chartmaker's Complaint

We have read a great many articles about how stars get their names, and they all miss the same things. They open with Johann Bayer, publish the year 1603, hand the reader a Greek alphabet, and declare that the brightest star in a constellation is Alpha, the next brightest is Beta, and so on down to Omega. Then they list Sirius as Alpha Canis Majoris at magnitude −1.44, Canopus as Alpha Carinae at −0.62, Vega as Alpha Lyrae at 0.03, and treat the case as closed. It is not closed. The rule they just stated is broken across half the sky, and the reason it is broken is the interesting part of the story.

The other omission is quieter and worse. These explainers treat "how stars get their names" as a single question with a single answer, when in fact three different naming systems overlap on any modern chart, and Bayer letters are only one of them. A reader who finishes such an article believes that Sirius, Alpha Canis Majoris, and any catalogue number attached to it are three flavours of the same thing, invented at roughly the same time, by roughly the same people, for roughly the same reason. None of that is true. What follows is a chartmaker's complaint about the standard explanation, and an attempt to replace it with something a person could actually use while looking at a sky.

What They All Get Wrong

The first error is the confident sentence that Bayer ranked the stars of each constellation by brightness. He did not, and he could not have, because a systematic magnitude scale in the modern sense did not yet exist in 1603. What he had was the classical tradition of six magnitude classes, inherited through Ptolemy from Hipparchus, and even those classes were assigned by eye and by consensus rather than by measurement. Inside a single magnitude class, "brighter" was already a matter of judgement, and Bayer often let the judgement fall on other criteria: position within the figure, tradition, the order in which stars had already been listed by earlier authorities. The result is a lettering that tracks brightness only loosely and, in several constellations, obviously does not.

The second error follows from the first: the confident sentence that Alpha of any constellation is the brightest star in that constellation. This is roughly true, but the exceptions are famous enough that any honest chartmaker learns them early. In Gemini, Beta (Pollux) is brighter than Alpha (Castor). In Orion, Beta (Rigel) is essentially tied with, and by most measurements brighter than, Alpha (Betelgeuse), which is also a variable that has never sat still on the scale. In Sagittarius the letters wander so far from the brightness ranking that neither Alpha nor Beta is anywhere near the constellation's top three. These are not curiosities. They are the honest state of the system, and any explainer that hides them is teaching the reader something that will fail them the first time they look up.

The third error is presenting the Greek alphabet as if it were the whole scheme. Bayer ran out of Greek letters in busy constellations and continued with lower-case Latin, then upper-case Latin, and the extensions were used unevenly by later cartographers who added, corrected, and sometimes contradicted his choices. A star labelled "b Centauri" or "P Cygni" on a modern chart is still, technically, a Bayer designation — a fact that the standard explainer almost never mentions, because it complicates the tidy image of twenty-four Greek letters marching from brightest to faintest. The scheme was never that tidy. It was a working printer's compromise in a folio-sized atlas, and it inherited every messy decision that came before it.

What Is Almost Always Missing

What almost every explainer skips is that Bayer letters are only one layer of a stack. Modern star charts carry, in silent superposition, at least three naming systems that were invented for different purposes and never fully reconciled. There are the proper names, most of them Arabic in origin — Rigil Kentaurus on the foot of the Centaur at magnitude −0.01, Vega in Lyra at 0.03, Capella in Auriga at 0.08 — which arrived in Europe through medieval translations of Ptolemy and were already old when Bayer sat down to letter his atlas. There are the Bayer letters themselves, added in 1603 as a constellation-by-constellation shorthand. And there are catalogue numbers, principally the Flamsteed numbers assigned nearly a century later and running in order of right ascension across each constellation, which is a different sorting principle entirely.

The standard article does not explain that these three systems answer three different questions. A proper name answers "what has this star been called by the humans who watched it"; it carries lore, and the lore has a tradition attached — Greek, Arabic, sometimes a southern navigator's naming heritage — that ought to be named when the name is used. A Bayer letter answers "where does this star sit inside its constellation's internal ordering, as one printer in Augsburg decided it in 1603". A Flamsteed number answers "where does this star sit along the constellation's east-west sweep, as one Astronomer Royal decided it near 1712". None of the three is a coordinate. None of the three is a physical measurement. They are three overlapping filing systems, and they disagree constantly.

What is also missing is the honest admission that the sky has been re-drawn since Bayer worked. The constellation boundaries the International Astronomical Union ratified in 1930 are not the boundaries Bayer used, because Bayer used none — his figures had outlines, not borders. When the IAU drew the borders as straight lines along fixed coordinates, some stars that Bayer had labelled inside one figure ended up, on the modern chart, inside a neighbour. A conscientious chart today keeps the historical letter attached to the star, not to the constellation, and the reader of a standard explainer is never warned that this drift happened. The map moved. The letters did not.

Orion The print from this article · from €29.95 View the print →

What I Would Say Instead

I would begin by saying that a star's "name" is not a single thing. It is a small pile of labels, each one a residue of a different century's attempt to organise the sky, and the labels do not agree because they were not built to agree. Sirius, Alpha Canis Majoris at magnitude −1.44, is a good place to see this: the proper name is old and travelled, the Bayer letter is a seventeenth-century printer's decision, and the coordinates that pin the star to the page — right ascension 6.75 hours, declination −16.72 degrees — are a modern measurement independent of both. All three refer to the same point of light. Only the last is doing what a chartmaker actually needs the label to do. The first two are memory.

I would then say that Bayer's letters are best understood as a piece of typography, not a ranking. In 1603 the problem Johann Bayer was trying to solve was that his Uranometria was going to be the first star atlas printed at large format with every constellation on its own plate, and he needed a way for readers to talk about a specific star without pointing at the picture. Numbers would have collided with existing catalogue numbers. Words would have been too long to fit on a copperplate. The Greek alphabet gave him twenty-four short, distinct, printable marks that a European reader could already recognise, and he distributed them across each constellation using brightness as a first pass and every other reasonable criterion as a second pass. It worked well enough to survive four centuries, which is more than can be said for most seventeenth-century engineering.

I would say, thirdly, that the reason the Alpha-is-brightest rule breaks so often is the same reason the system is worth keeping. Bayer was making a chart, not a photometric database. He cared about the figure he was drawing, the tradition he was inheriting, and the printer he was working with. That is why the labelling of Orion follows the figure of the hunter — belt, shoulder, foot — as much as it follows brightness. That is why Canopus, sitting at magnitude −0.62 in the deep southern constellation Carina, was not even in Bayer's atlas at all: the southern sky he printed was thin, because European observation of it was thin in 1603, and the letters below the celestial equator were filled in piecemeal by later hands who did not always agree. The scheme is a record of what humans knew, in what order, from what latitude. It is not a scoreboard.

Finally I would say what the standard explainer refuses to say, which is that the modern reader does not need Bayer letters to find a star. Arcturus is at right ascension 14.26 hours, declination +19.18 degrees, magnitude −0.05, and any planetarium software will point at it without a single Greek letter in sight. What Bayer letters give the reader is not location but lineage: a chain of custody back to a specific book, a specific engraver, a specific decision made by a lawyer-astronomer in Augsburg in the year Elizabeth I of England died. That is worth keeping. It is not worth pretending it is something else.

FAQ

Did Johann Bayer really assign Greek letters strictly by brightness?

No, and any explainer that says so is oversimplifying. Bayer used brightness as a rough guide within the six classical magnitude classes he inherited from Ptolemy, but he broke from strict brightness order whenever position within the figure, prior tradition, or the sequence in earlier catalogues gave him a reason. Modern astronomy did not have a measured magnitude scale until the nineteenth century, so the notion that Bayer could have ranked stars precisely by brightness is anachronistic. His scheme is an editorial arrangement, not a photometric one.

Which famous constellations break the "Alpha is brightest" rule?

Several, and the exceptions are well known to working chartmakers. In Gemini, Beta (Pollux) outshines Alpha (Castor). In Orion, Beta (Rigel) is at least as bright as Alpha (Betelgeuse), and Betelgeuse is a variable that swings noticeably on human timescales. In Sagittarius, neither Alpha nor Beta ranks near the top of the constellation by brightness. Any article promising a clean brightness-to-letter map is glossing over the honest state of the sky.

What happens when Bayer runs out of Greek letters in a busy constellation?

He continues with lower-case Latin letters, then upper-case Latin letters, and this extension is still considered part of the Bayer system. Designations like "b Centauri" or the famously variable "P Cygni" are Bayer letters, not typographical accidents. Later cartographers extended and occasionally corrected his choices unevenly, which is why a modern chart of a crowded constellation can show a mixture of Greek, Latin lower-case, and Latin upper-case labels on stars of similar brightness.

How is a Bayer letter different from a Flamsteed number?

They answer different questions. A Bayer letter, assigned in the 1603 Uranometria, marks a star's position within a specific constellation's internal ordering as Johann Bayer arranged it. A Flamsteed number, assigned around 1712 by John Flamsteed, numbers the stars of a constellation in order of right ascension — that is, from west to east across the sky. Bayer's system is figure-driven and tradition-driven; Flamsteed's is coordinate-driven. Both survive on modern charts, and they disagree with each other constantly.

Why do some bright southern stars have no Bayer letter?

Because Bayer's 1603 atlas was drawn from a European vantage and the far-southern sky was thinly observed by European astronomers at the time. Canopus in Carina, sitting at magnitude −0.62 and the second-brightest star in the entire night sky, was outside the reach of his source material. The Bayer letters that later appeared on southern stars were added piecemeal by subsequent cartographers, which is one reason the southern lettering can look inconsistent compared with the northern figures Bayer treated in full.

Do the proper names of stars come from Bayer's atlas?

No. Names such as Rigil Kentaurus on the foot of the Centaur at magnitude −0.01, Vega in Lyra at 0.03, and Capella in Auriga at 0.08 are older than the Uranometria and reached European use through medieval Arabic astronomy and its Latin translators. Bayer printed many of these names on his plates, but he did not invent them. Treating a proper name and a Bayer letter as products of the same tradition erases the older Arabic naming heritage that most star names actually carry.

Are the constellation boundaries used today the same ones Bayer used?

No. Bayer's constellations had figures and outlines, not borders. The rectangular boundaries used on modern charts were ratified by the International Astronomical Union in 1930, and they cut the sky along fixed coordinates. Some stars Bayer had lettered inside one constellation now sit, on the modern chart, inside a neighbouring one. A careful chart keeps the historical letter attached to the star rather than to the current constellation, but the reader of a standard explainer is rarely told this drift happened.

If Bayer letters are so unreliable, why do astronomers still use them?

Because they are compact, memorable, and carry four centuries of continuous use. A designation like Alpha Lyrae for Vega at magnitude 0.03 is shorter than a catalogue number and richer than a coordinate pair — it points at a specific star and, at the same time, at a specific human decision recorded in a specific book. Modern software does not need Bayer letters to locate a star, but the labels preserve a chain of custody from printed atlas to digital chart that no coordinate system alone provides.

The Southern Sky The print from this article · from €29.95 View the print →