There is a word on café menus everywhere that now means chocolate, used to mean a Yemeni port, and for about two hundred years meant "coffee good enough to have a name". Al-Makha is the port. It sits on the Yemeni coast of the Red Sea, and its coffee trade ended more than a hundred years ago. The word outlived the harbour by a wide margin, which is exactly the problem a Chicago court had to sort out in 1912.
Origin words on coffee are enforced as a ban on lying, not as protected geographic names. A US federal case in 1912 turned on whether Ethiopian coffee could be sold as "Longberry Mocha", and the outcome was not a ban on the word. The court settled on "Abyssinian Mocha", a qualifier attached to a name the seller was allowed to keep. That is still roughly where origin labelling sits. A regulator can act against a claim that is false. Almost nobody has defined what most origin words on a coffee bag are supposed to mean in the first place.
Mocha was ambiguous before anyone cheated with it
The confusion did not start with fraud. It started with logistics.
A bean grown on a Yemeni highland terrace changed hands several times before it saw salt water. It came down to the coastal plain and into the wholesale market at Bayt al-Faqih, and only then went onto camels for the last stretch out to Al-Makha or Hudaydah. Jonathan Morris traces the money behind that chain to Banyan merchants from Gujarat, who held the Yemeni credit networks and who, on his reading, did not merely finance the crop but effectively started it. So the word stamped on the bale named the exit point and said nothing at all about where the coffee had grown.
Then the exit point started handling two countries' coffee. In 1638 the Qasimi dynasty of Zaydi imams drove out the Ottomans, unified Yemen, and took Zeila on the African shore as well. One authority now sat on both sides of that water, so the two crops left through the same harbour, and buyers at the far end filed them under a single word. A port label was covering two origins nearly three centuries before a court was asked to referee it. When an American importer in 1912 argued that Abyssinian coffee could be sold as Mocha, he was not inventing a lie. He was describing a shipping practice that predated the United States.
The port itself then died. Britain seized Aden in 1839 and made it a free port in 1850, and Aden took the traffic. Al-Makha's approaches silted up on top of that. Morris passes on the local explanation, which is that American ships dumped their stone ballast in the channel and filled the space it left with coffee, and it is a satisfying story I would treat as likely rather than proven. By the time the word reached a courtroom, it referred to a place that no longer functioned.
Why 1912, and why the United States
Because there were only two words. William Ukers, writing in 1922 from inside the trade, describes an American market in which Java and Mocha were close to the whole consumer vocabulary for coffee until Brazilian volume arrived. A vocabulary that small is not a description system. It is a pair of quality signals, and quality signals with no definition behind them are an invitation.
The trade took the invitation. A bag sold as Mocha routinely had Bourbon Santos in it. A bag sold as Java routinely had Sumatran coffee in it. In Holland the counterfeiting became an industrial process with a name of its own, manipulated Java: buyers picked out the Santos beans that most resembled Java, then worked the surface, polishing and colouring until the resemblance held up on a merchant's desk. That ended in 1912, when American consuls stopped certifying invoices unless a written declaration of purity came with them. Faking one origin's appearance had become somebody's full-time occupation.
There was also a taste being faked. Months in a humid hold turned green coffee brown and took some of the sharpness out of the cup, and buyers of the period paid a premium for exactly that. So firms ran beans through steam to counterfeit it, and Ukers records a federal prosecution that put the practice on the wrong side of the 1906 Pure Food and Drugs Act, as adulteration and as misbranding. India's Monsooned Malabar is the surviving lawful version of the same flavour, made by leaving coffee out in the monsoon wind until the beans swell and go pale gold. One region industrialised the effect in the open. A whole trade faked it and got prosecuted.
The prosecutions were specific, with names and quantities attached. In one 1910s case an importer and a broker in New York took 84 bags of washed Caracas coffee, re-bagged it under P.A.L. Bogota, a known Colombian mark, and were convicted of conspiracy and fined $3,000 each. That figure is as of the decade it was levied in, the 1910s, and is worth many times the number today. The Bureau of Chemistry followed in February 1921 with a ruling that shut Coffea robusta out of the Java name entirely, including labels that only implied it. The New York Coffee Exchange had already stopped accepting robusta on delivery in 1912. Origin fraud in coffee was a documented, prosecuted offence with fines and bag counts more than a hundred years ago, which is worth remembering the next time origin transparency is described as a new concern.
What the court actually settled
The Chicago decision is easy to misread as a win for accuracy. It was narrower than that.
The court was not asked where the coffee grew and did not create a rule about where coffee called Mocha must come from. It was asked whether a particular phrase would mislead a buyer, and it answered by requiring a qualifier that carried both facts at once. "Abyssinian Mocha" tells you the coffee is Ethiopian and tells you it belongs to the Mocha type. "Longberry Mocha" implies Yemeni origin and hides the Ethiopian one.
That is a misbranding standard, not an appellation. French wine law says who may use the word Chablis and on what land. The 1912 coffee ruling says only that you may not use a word in a way that deceives. The difference matters, because a misbranding standard cannot protect a name, it can only punish a specific false statement about one. If nobody has defined what the name means, there is nothing precise to be false about.
What this will not do for you
The law here will not tell you the coffee is good. Nothing in a truth-in-labelling regime speaks to cup quality, and the 1912 outcome would have been identical for a bad Ethiopian lot and an excellent one.
It will not give you farm-level traceability either. "Single origin" is a trade term rather than a legal one. Jessica Easto draws the line by counting places: one for a single origin, two or more for a blend, with the blend existing so a roaster can keep the same cup on the shelf in every month of the year. Neither half of that distinction puts any size limit on the word location. A country is one location. So is a co-operative of four hundred growers, and so is a single washing station. All three can be printed on a bag lawfully, and only one of them tells you very much. This is the reason single origin means considerably less than most people assume.
It will not stop the word drifting further, either. Mocha is now simultaneously a Yemeni port, a botanical variety, a coffee type and an espresso drink with chocolate in it. Almost nothing about the drink connects to the origin of the word.
And there are exceptions, so I will not claim the field is uniformly unregulated. Hawaii has statutory blend-percentage disclosure for Kona, which is the strongest origin-labelling rule in the American coffee market, and the required minimum percentage has been amended since most coffee books were written. I am not quoting a figure I have not verified against the current statute. The general point stands: origin protection in coffee exists in a small number of jurisdictions that fought for it, and nowhere else.
The market name is the thing to understand
Kenneth Davids has a term that clears most of this up: the market name. These are the old trade labels, stencilled onto sacks and typed onto importers' lists, and most were already in circulation in the nineteenth century. What they point at is inconsistent. Some name the region where the coffee grew. Others name the harbour it left through, and Mocha names a harbour that stopped handling coffee generations ago.
A market name is a piece of trade shorthand with commercial weight and no legal definition. Mocha-Java, which has been around longer than any other blend still on sale, shows what that costs. The old proportions run about one third Yemeni Mocha to two thirds Java. The Yemeni third is in there for brightness and fruit, the Java for the weight underneath it. Davids points out that the Yemeni third is now commonly filled by dry-processed Ethiopian Harar, and that the blends doing this are not the careless ones. So a bag labelled Mocha-Java today may contain neither Yemeni coffee nor Indonesian coffee, and may still be honestly sold, because the name now describes a flavour intention rather than two places.
The 1912 court did not stop that. It could not have. It set the one rule the trade could actually enforce, which is that a name may not carry an implication the seller knows to be untrue.
What to look for instead
Ignore the romance words and read the operational ones. A specific producer, co-operative or washing station name is worth more than a country. A harvest year tells you whether you are drinking coffee that is in its season or well past it. The processing method tells you more about how the cup will taste than the origin does, which is why three words on the bag carry so much of the flavour information. A roaster who publishes what they paid and to whom has given you a claim that can be checked, and a farm name on its own has not.
One caveat on the history itself. The detail of the 1912 Chicago case reaches us through Ukers's 1922 account, which is a trade record written by an industry insider rather than a court reporter, and I have not read the docket. The outcome is consistently reported and the surrounding enforcement actions are well documented. The internal reasoning of the judgment is not something I can vouch for line by line.