The editors of Doubleday, headquartered at 1745 Broadway in New York, would surely have raised their eyebrows and grabbed their red pencils if a best-selling novelist had submitted a manuscript that placed the Empire State Building on Central Park West, the United Nations on Broadway and Yankee Stadium on Fifth Avenue. Yet Doubleday has published a novel - number one on the best-selling lists for a good many weeks - rife with so much confusion about the sites of Paris that it is hard not to wince. This might be excusable if Paris played a minor role in the book. But the main setting of Dan Brown’s The Da Vinci Code is Paris. Someone should have supplied him with a map.

The novel opens with Brown’s hero, Robert Langdon, staying in the renowned Ritz Hotel, which is actually located right on the Place Vendôme. But Brown mistakenly places the Ritz somewhere far to the north of the Place Vendôme. A police car picks up Langdon at the Ritz and speeds south toward the Louvre Museum. According to Brown, the car passes the Opera House (which is actually two long blocks north of the Ritz), crosses the Place Vendôme (which should have been the starting point), and heads down the Rue Castiglione into the Tuileries Garden. I cannot see how a car could enter the Tuileries at this point, but since it is a novel, and the French police can be reckless, I am inclined to concede this to Brown. The police car goes on to the Louvre, scene of a grisly murder.

A little more than a hundred pages later, Brown fails to locate the American Embassy correctly, one of the best known sites of Paris. The embassy is on Avenue Gabriel, just off the Place de la Concorde. Langdon and his friend, French Cryptology Agent Sophie Neveu, are rushing in her minuscule Smart Car on Rue de Rivoli toward the embassy. But, when they reach the Place de la Concorde, they do not stop at the embassy. They head down the Champs-Élysées instead. Brown tells us that the embassy is now “only about a mile away.” I kept shouting at Langdon and Neveu to stop. The embassy is right there. But they ignored me and motored onward.

After a while - Brown does not tell us how long - Sophie swings her car to the right, “cutting shortly past the luxurious Hôtel de Crillon.” “The embassy,” Brown tells us, “was less than a mile away now.” But the Crillon is right on the Place de la Concorde, just across the street from the embassy. If they are at the Crillon, they are at the embassy. Moreover, if they are at the Crillon, it means they have somehow gone up the Champs-Élysées and back. Brown’s geography at this stage is a complete mess.

The author’s ignorance of Paris is especially embarrassing because his book strives mightily for a show of erudition. He describes a thousand-year-old secret society called the Priory of Sion, delves into the machinations of the Roman Catholic sect known as Opus Dei, and talks learnedly about the Fibonacci sequence in mathematics. “All descriptions of artwork, architecture, documents, and secret rituals in this novel are accurate,” he tells the reader, and he thanks the Louvre Museum, the Bibliothèque Nationale, the Royal Observatory in Greenwich, and the Federation of American Scientists, among others, “for their generous assistance in the research of the book.”

What to make of all this? To be honest, I do not have any practical suggestions to ensure that grievous errors are caught before publication. It is obvious that Brown’s book was not closely edited at Doubleday, at least not by anyone with a memory of Paris. Publishers do not have the means or inclination to check every fact in the books they produce. Editors fret mostly about grammar, spelling and libel. Everyone seems to assume that an author stands behind the research.

If publishers hired an army of fact-checkers, errors like those of Brown might be ferreted out. But this is a cure too terrifying for a writer to endorse. Several prominent magazines employ fact-checkers, and I can testify that some - though far from all - are so relentless they can flay a story into tatters. They often check everything, no matter how innocuous, no matter how obvious. Writing an article about the Alhambra for the Smithsonian Magazine a few years ago, I quoted José Antonio Martínez Soler and described him as “one of Spain’s most distinguished journalists.” Martínez Soler, a former television news anchor and the author of several books, is the editor of a daily newspaper in Madrid.

The fact checker phoned him in Madrid and asked, “Is it true that you are one of Spain’s most distinguished journalists?” He hesitated and then said, “Just a minute while I ask my wife.” A few moments later, he returned to the phone and told the fact-checker, “Yes, I am.” She checked that fact off as correct. On that same page, however, I had listed Constantinople as a grand medieval Muslim city when I had intended to write Damascus. The fact-checker never caught that one, and the story was published with the glaring error.

A couple of decades ago, when I was the Toronto correspondent of the Los Angeles Times, the Reader’s Digest decided to condense and print an article of mine about the Canadian national inferiority complex. As part of my evidence, I cited the case of a Carleton University journalism student who had embarked on a research project trying to deflate the heroic status of the Canadian ambassador who had saved six American diplomats from the mob that had overrun the American embassy in Teheran and taken everyone else hostage.

The Reader’s Digest obviously liked the student anecdote. A fact-checker kept phoning me about it. But I could not remember the name of the student who had come to my office. In an accusatory tone, the fact-checker said she had phoned Carleton University several times but had failed to verify the research project. She obviously thought I had made up the story. Nothing I said could satisfy her need for proof. At the end, Reader’s Digest published the piece without the anecdote.

In any case, fact-checking is not a practical solution for a book like The Da Vinci Code. Each page is so chock full of facts that it would probably take a decade for a dozen researchers to check them out.

I suppose if editors - the bosses of the fact-checkers - were all worldly and wise, they would easily catch the misplacement of the Ritz, the American Embassy and the Hotel de Crillon in a book planted in Paris. Many are worldly and wise. But some editors let down their side. In 1994, the news editor of the Washington bureau of the Los Angeles Times assigned me to cover the hundreds of people crowding the sidewalk in front of Jacqueline Kennedy Onassis’s apartment building in New York while she lay upstairs near death. “The building is at 85th and Fifth,” he told me, “but I don’t know which is the street and which is the avenue.” “Don’t worry,” I assured him. “I’ll try to figure it out.”

I guess we writers are our own last defense. We have to try hard, very hard, to get our facts right in the first place. I do not know Dan Brown, but I would guess that he wrote the passages about driving through Paris hurriedly. He had a cloudy idea in his mind about where places were but assumed someone at Doubleday would backstop him.

It is surely presumptuous for a middling writer like me to give advice to a writer as successful as Brown. But I’ll try anyway. In the future, take it slow. Forget about a backstop. Buy a map. Get it right. Don’t let our side down.