Last winter, when University of Pennsylvania sociologist Elijah Anderson first heard a colleague's summary of her forthcoming book, he thought he was hearing an echo of his own work.

Today that echo has turned into uproar, a dispute that has set scholar against scholar and led researchers at some of the nation's most prestigious colleges to choose sides.

The charge? "Conceptual plagiarism."

Neither Anderson nor his supporters claim that fellow Penn sociologist Kathryn Edin copied his writings in her new book, "Promises I Can Keep: Why Poor Women Put Motherhood Before Marriage," written with St. Joseph's University professor Maria Kefalas.

Instead they accuse the authors of appropriating concepts and ideas that sprang from Anderson's research, particularly those developed in his 1999 book, "Code of the Street: Decency, Violence, and the Moral Life of the Inner City."

"There's serious under-citation," Anderson says.

The controversy, first reported in the university's Daily Pennsylvanian newspaper, raises issues that reach beyond the halls of academia. In an age of file-sharing and open-source programming, when everything from movies to music can be captured at the touch of a button and when the Internet serves as a worldwide echo chamber, who truly owns an idea?

The stakes are enormous as countries move ever further from an industrial economy toward one where intellectual property is the thing of real value. And unlike the verbatim theft of written words -- a precise, quantifiable sort of plagiarism -- determining the provenance of an idea can be a mushy, mucking affair.

More than a dozen professors at schools including Princeton and Harvard have come to the defense of Edin and Kefalas, calling the charge of conceptual plagiarism "absurd." In response, Anderson compiled a list of 22 similarities between the books -- both examine motherhood and marriage in the inner city -- posting the comparison on the Penn Almanac, a university Web site. Anderson also released a statement saying the two authors "misled readers" by "repeating the distinctive ideas, findings, explanations and terms of Code without citing the source."

Edin declined to be interviewed for this article. Kefalas, who faced questions about attribution in her previous book -- questions that were resolved when she posted a list of corrections on the St. Joseph's Web site -- declined to discuss the controversy at Penn. But in an e-mail, she said she respected Anderson's work and had cited him more than any other ethnographer in "Promises." "The fact that we study some of the same issues makes some degree of conceptual overlap inevitable," she wrote, "but in no way did we appropriate Dr. Anderson's work."

On Tuesday, a letter signed by 110 scholars who support Anderson appeared on the Almanac, the latest salvo in the dispute.

"The use of ideas is one of the most problematic areas," says John P. Lesko, an English professor at Saginaw Valley State University in Michigan, where he runs the Famous Plagiarists Research Project. "How do you prove for sure that your idea has been taken?"

Who owns an idea? That question might best be answered with another: How does a clownfish smile?

In 2003, the Walt Disney Co. released an animated feature about a lovable lost clownfish, "Finding Nemo" -- and French children's author Franck Le Calvez thought he recognized his own story. He had written an outline for a movie, never made, and a book, "Pierrot Le Poisson Clown."

Le Calvez accused Disney of stealing his fish. Disney insisted "Nemo" was its own creation.

In court, a French judge found meaningful differences between the characters. When Nemo smiles, the judge said, he shows his teeth, like a person. But when Pierrot smiles, he keeps his lips together, akin to a dolphin. Further, the judge noted, Pierrot is a deep orange while Nemo is closer to red.

Le Calvez lost the case. In April, he returned to court and lost again. This time, the court ruled that Le Calvez had known about "Nemo" and had fraudulent intentions when he registered his character. He was ordered to pay about $73,000 in damages.

To people who study plagiarism, the case illustrates the complexity of trying to assign credit for overlapping ideas -- and how serious money can ride on the outcome. "Finding Nemo" earned Disney $340 million in box office receipts.

Big money usually isn't an issue in academic and scientific publishing, which embraces a tradition of attribution and citation. But even those fields are subject to new pressures. As colleges are run more like businesses, with administrators pushing faculty members to produce not just scholarship but revenue, the credit for groundbreaking ideas has become the currency of the market -- and the key to securing research grants and book contracts.

Further complicating matters: Plagiarism is an ethical concept, not a legal term. The police don't arrest people for plagiarism. Nobody goes to jail. Punishment is meted out in the form of damaged reputations and lost jobs.

Copyright law offers limited protection for a writer's words and none at all for the ideas expressed by those words.

"You can't copyright an idea," says Philadelphia licensing lawyer Peter T. Wakiyama. "That's Intellectual Property 101."

The difference between borrowing and not, between attribution and none, is the reason why few people have heard of Florence Deeks. And why everyone has heard of H.G. Wells.

In 1927, the year Deeks took him to court, Wells was the estimable author of popular science-fiction novels including "The Invisible Man" and "The War of the Worlds." But his wealth was secured by a best-selling nonfiction book, "The Outline of History."

Deeks, an amateur historian, accused Wells of stealing her unpublished manuscript, "The Web of the World's Romance," and lifting its ideas, its analysis and parts of its text. Wells, she said, even repeated her mistakes.

For instance, Deeks erroneously referred to Hatshepsut, the ancient Egyptian pharaoh, as "Hatasu." So did Wells. She gave the wrong date for the start of the Holy Roman Empire. Wells did, too.

Deeks sought $500,000 in damages -- the equivalent of $5.2 million today. But she had two things going against her: One, she was a woman. Two, she was a feminist. An unmarried feminist. In other words, a troublemaker.

Male-dominated courts in Canada and England sided with Wells, who maintained that he'd completed "Outline" with notes taken while writing other books and the help of prominent historians.

Deeks wasn't vindicated until 2002, when professor A.B. McKillop of Carleton University in Canada published his authoritative book "The Spinster and the Prophet: H.G. Wells, Florence Deeks, and the Case of the Plagiarized Text." McKillop documented how Deeks's manuscript found its way to Wells via his North American publisher.

By then, of course, it was too late for Deeks. After losing in court, she tried to interest publishers in a revised version of "The Web of the World's Romance" but found no takers.

"Your book would be subject to comparison with Wells' 'Outline of History,' " wrote an editor at Little, Brown. "For that reason I think you will have difficulty in securing a publisher."

At Penn, the next steps are uncertain. Neither side has filed an official complaint, which would trigger a more formal inquiry by the university.

The dispute could have ended last summer, in fact, when Anderson and Edin met to discuss their books, eventually reaching a confidential agreement.

But in August, the affair became the talk of the American Sociological Association conference, held in Philadelphia. Afterward, disturbed by what he saw as official silence, Penn professor emeritus Harold Bershady sent an e-mail to the department faculty, making the charge of "conceptual plagiarism."

When his e-mail was leaked to the Daily Pennsylvanian, the furor ignited publicly.

"I still do not understand why the department remains mum," Bershady says. "The longer it goes, the worse it gets."

Anderson says he's not sure what will happen.

"I didn't and still don't impute malice or sinister motivation on their parts," he says. "I simply maintain that prior scholarly work should be properly acknowledged."

Elijah Anderson of the University of Pennsylvania says a new book about motherhood in low-income areas does not credit his research. More than a dozen professors have come to the defense of Kathryn Edin, left, and Maria Kefalas, calling the charge of conceptual plagiarism "absurd."