The Word That Wasn't There

On page 771 of a dictionary, between a small kangaroo and a furnace for refining gold, there was a word that no one had ever used. It was printed in lowercase, given a pronunciation, assigned a part of speech, and furnished with a definition: dord, noun, physics and chemistry, density.

The word was not obscure. It was nonexistent.

For several years, dord sat among the legitimate inhabitants of Webster’s New International Dictionary, Second Edition. It looked official because it had passed through the machinery that makes words look official. A printer had set it. An editor had supplied its pronunciation. A proofreader had seen it. A book had preserved it. That was enough to give a mistake the posture of a fact.

The slip of paper that changed shape

The story begins in 1931, with Austin M. Patterson, Merriam’s chemistry editor. He was not inventing a term. He was recording an abbreviation. In the notation used by the dictionary office, a capital D or lowercase d could stand for density. Patterson typed the information on a small editorial slip: “D or d,” followed by the explanation “density.”

A dictionary office was a physical system before it was a literary one. Information lived on slips of paper, and those slips had to reach the right piles, editors, printers, and proofreaders. At some point, Patterson’s note escaped the file for abbreviations and entered the alphabetic stream of words. The spaces in “D or d” disappeared. The sequence became Dord.

That alone would not have made a dictionary entry. Someone still had to treat the new shape as a word. The initial letter was changed to lowercase, a pronunciation was added, and the entry was given the grammatical label “noun.” Each ordinary editorial action made the original error harder to recognize. The mark no longer looked like a note about notation. It looked like a term that had been waiting to be defined.

When the second edition appeared in 1934, it defined dord as “density” in physics and chemistry. It offered no etymology and no examples of the word in use. Those absences were clues. A real technical term should leave traces in the papers and books of the people who use it. Dord left none.

The dictionary itself was an imposing object, produced with the labor of hundreds of editors and consultants. Its scale made the error more instructive, not less. A huge reference work does not eliminate human judgment. It gives human judgment a long chain of handoffs through which a small misunderstanding can travel.

A word becomes real by being used

The modern dictionary is often imagined as a court. A word appears before the editors, who decide whether to admit it into the language. The actual process is closer to natural history. Editors observe words in books, newspapers, magazines, and electronic publications. They collect examples, compare meanings, and look for patterns of use. The evidence comes first; the entry records what the evidence shows.

That is why dord was a ghost word. It had the appearance of language without the life of language. It was not a rare word, an old word, a regional word, or a technical word whose meaning had escaped the general reader. It had no established usage at all. Its only habitat was the dictionary.

The phrase “ghost word” was introduced into lexicography by W. W. Skeat in 1886 for forms produced by the mistakes of printers, scribes, or editors. Allen Walker Read later distinguished such words from several neighboring creatures. A nonce word is made for a particular occasion, perhaps by a poet or novelist. A newly coined word may be unfamiliar but still have a speaker, a context, and an intention behind it. A ghost word has a different origin: it is an error that has acquired the silhouette of vocabulary.

Some ghosts can become real. Read pointed to forms such as mumpsimus, which began in a misreading but was later used deliberately, and gravy, whose history may involve a corrupted reading that settled into English. Once people use a form knowingly and repeatedly, its origin becomes less important than its life. Language is not a museum in which every object must have been manufactured correctly. Usage can adopt a mistake and make it its own.

Dord never received that rescue. No physicist needed it. No chemist wrote a paper with it. No speaker carried it out of the dictionary and into the world.

The editor who noticed the silence

On February 28, 1939, an editor named Edward Artin noticed that dord had no etymology. That was not merely a missing footnote. It was a missing history. He searched for the word’s evidence and found nothing. The office records led back to Patterson’s slip, where the supposed word dissolved into the phrase “D or d.”

The correction was ordered with an urgency that sounds almost theatrical: “plate change / imperative / urgent.” The printer was told to remove the entry. The space it left behind was repaired by expanding the neighboring entry for a Doré furnace, turning a short definition into a longer grammatical phrase. Even the erasure had to be engineered. A dictionary page is a tightly packed physical object; delete a word and the blank must go somewhere.

The discovery happened in 1939, but the correction did not instantly clean every copy. Changes to the printing plate took time, and copies bearing the old entry continued to circulate. The word was removed from later printings, yet dord remained visible in some books into the 1940s. The ghost was discovered before it was fully exorcised.

In 1954, Philip Babcock Gove published “The History of ‘Dord’” in American Speech, turning an embarrassing office accident into a public account. The paper did not pretend that dictionaries were infallible. It showed instead how an error moved through a process built to make errors unlikely. The very detail that makes the story funny, the transformation of “D or d” into dord, also reveals the seriousness of editorial work: every symbol has to remain attached to its intended category.

The authority of a printed mistake

There is a temptation to make dord a story about careless proofreaders. That would be satisfying and wrong. The error required no single villain. It needed a file system, a change in how abbreviations were organized, a lost or re-created slip, a plausible-looking string of letters, and several people encountering the result at different stages. Each person could make a reasonable local decision while the whole process drifted away from the original meaning.

The dictionary then supplied the final illusion. Once printed, dord looked as though the language had created it. A reader could reasonably assume that a large reference book knew something they did not. The authority of the page concealed the absence of evidence beneath it.

That is the small warning inside the joke. A reference work is not the source of a word’s existence. It is a record of evidence about how people speak and write. When the record is mistaken, its typography can lend the mistake a borrowed reality.

Today dord is not a word for density. It is a word for the moment a filing error puts on a suit, receives a pronunciation, and walks into the dictionary as if it belongs there. Its brief life is over, but the lesson remains useful wherever information passes from one human system to another: the most convincing falsehood may be the one that has been formatted correctly.