The quote is ancient. The wording isn't.

THE OBSTACLE IS THE WAY, cast in raised bronze capitals across two lines. Verdigris has settled into every recess; the faces of the letters are rubbed pale and bright where a hand keeps passing over them.

Underneath, in far smaller letters, one more word: ATTRIBUTED.

That word is doing more work than the line above it.

Why does this kind of quote get mangled so often?

Because it wasn't written to be quotable in the first place.

The Stoic writing that ends up on shirts mostly began as private notebooks and letters between practitioners — a man arguing with himself about how to get through a bad week, not composing sentences for an audience. Turning that into something readable in three seconds means somebody has to cut it down, smooth the syntax, and choose a rhythm in English that the original language simply didn't have.

Most of what circulates as "the Stoics said" is exactly that: a compression, sometimes several compressions deep, each pass moving a little further from the notebook it started in.

What's the difference between a paraphrase and a fabrication?

A paraphrase still traces back to something. A fabrication doesn't trace back to anything at all.

A tight, punchy compression of a real passage is ordinary editorial work — every popular translation does some version of it, and a reader who goes looking can usually find the longer original underneath. A fabricated line has no underneath. It's a sentence somebody wrote because it sounds like the sort of thing that philosopher would say, and it started travelling with his name attached because nobody checked. Both end up looking identical on a shirt. Telling them apart is the whole difference between quoting and inventing.

So what's the honest way to use the material?

Pick one of three moves, and each one is legitimate.

Name the actual work the line is drawn from, if it's genuinely known — a title under the quote does real work. Or leave the line as it is and admit, plainly, that the phrasing is a compression rather than something transcribed word for word, which is what ATTRIBUTED does here. Or set the philosopher's name aside entirely and write a sentence of your own, in your own voice, about the same idea. Passing off a fabrication as a transcript is the one move that isn't on that list.

Who owns a line like this?

Two different things, and they have different owners.

The ancient idea itself is old enough to be common property — nobody living wrote it, and no estate holds it. But the specific, polished English sentence a reader recognises usually isn't the ancient text. It's a modern translator's phrasing, chosen word by word, decades or centuries after the original — and a translation is its own copyrighted work with a living or recently living author behind it. Lift a well-known modern rendering onto a shirt and the thing being quoted is the translator, not the philosopher. This is the layer almost nobody accounts for, which is exactly the general shape of the licensing question quoting anybody raises — it just arrives here through a second author nobody notices is standing behind the first one.

Does the format work outside the Stoics?

Anywhere the same gap exists between an old idea and its modern English wording.

Scripture in translation, folk sayings passed through several languages, historical figures quoted from letters nobody but a specialist has read in the original — all of it carries the same two layers, an old source and a modern renderer. It's a close cousin of the layering problem a genuinely bilingual print runs into, just arriving from the direction of time instead of the direction of language. The honest habits are identical either way: know which layer you're actually quoting, and say so.

Isn't ATTRIBUTED just hedging — a way of not committing to the line?

It reads the opposite way once you notice it's cast into permanent metal rather than left off.

Leaving a disputed line unmarked is the version that avoids commitment — it lets the reader assume a certainty nobody actually has. Casting the qualifier into the same bronze, at a size the plaque can carry, states the actual level of confidence and stands behind it. A confident-looking line with no hedge isn't more honest. It's just quieter about the same problem.

Does it print well?

Very, because a plaque is naturally a two-tone object.

Raised lettering on bronze photographs as a flat warm metal tone with a darker, cooler tone pooled in the recesses — no gradient needed, just two solid colours doing the job light and shadow would do on the real object. The letterforms want to stay wide and evenly spaced rather than tight, the way cast lettering actually is, so the print doesn't have to resolve fine gaps between strokes. Charcoal, bone and dark green all carry it well, and it prints fine with a palette this short — two inks does the job.

How do I get one made?

Describe the line at JustOG — the words, whether you want ATTRIBUTED cast beneath it or a named source instead, and the material it should look cut or cast into. 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.

A plaque turns that shade of green after years of being touched in exactly the same spot.