Somewhere in the northern suburbs of a European capital there is a street that has never existed. It appears on a commercial map published in 1974, complete with a name, a length, and a plausible relationship to the roads around it. It was placed there deliberately. The practice has a name in the trade: a trap street. Cartography is expensive and copying is cheap, and a publisher who suspects a rival of copying has a difficult evidentiary problem. Two accurate maps of the same city will inevitably resemble each other; resemblance proves nothing. A fictional street, however, cannot arrive independently. If it appears in a competitor's edition, the inference is unavoidable. Similar devices exist elsewhere. Dictionaries have carried invented words, directories fictitious subscribers, and encyclopaedias entries for people who were never born. The logic is identical and, as a piece of reasoning, rather elegant: you cannot copyright a fact, but you can copyright the particular arrangement of facts, and an error is the one element of a reference work that is unambiguously yours. What makes the trap street interesting is what happened next. Maps are not merely descriptions. They are used, and use has consequences. Delivery drivers looked for the street. Councils received queries about it. At least twice, according to the cartographic literature, a fictitious road appeared in a later official document because a clerk, checking a commercial map, assumed it had simply been overlooked in his own records. A thing invented to prove copying had begun, faintly, to be copied by reality. This is less mystical than it sounds. Any sufficiently authoritative description tends to be treated as a source rather than a claim, and sources propagate. The mechanism is the same one that keeps a corrected statistic circulating for years after its retraction: not credulity exactly, but the reasonable assumption that somebody upstream has already checked. The legal position, when it was finally tested, proved unhelpful to the publishers. Courts in several jurisdictions have taken the view that a deliberate error is not protected by copyright at all, because copyright subsists in expression rather than in mistakes, and because a map that knowingly misleads its users is arguably defective as a product. The trap street survives more as folklore than as practice. But the underlying difficulty has not gone away; it has migrated. Digital mapping firms now embed traceable irregularities in their data, and the question of who owns a description of the physical world is litigated more often, and with far more money at stake, than it ever was in 1974. The street in the northern suburbs was removed from later editions. The area was developed in the 1990s and now contains a road running roughly where the fictional one was drawn. It has a different name. Nobody involved in the planning was aware of the coincidence, and there is no reason to think it was anything else.