After years of wrangling, a 2005 study on symmetry and dance published in Nature has been retracted. The paper reported that Jamaican men and women with more symmetric bodies were also better dancers. Because symmetry is seen as an indicator of genetic quality, this would bolster the theory that human dance evolved as a sexually selected courtship ritual. The retraction notice gives no reason.
Biologist Robert Trivers of Rutgers University in New Brunswick, New Jersey, one of the authors, says he became aware of problems with the data after publication. He accuses co-author William Brown of preselecting the dancers and changing the values on some of the dancers' measures of symmetry. When Trivers's early attempts to have the paper retracted failed, he wrote a short book about it called The Anatomy of a Fraud that he sent to researchers who cited the paper. "Everybody in the fraud network, that is the journal, the university, the fraudster, they all benefit by stringing the thing out," he said earlier this year.
In April, an investigation at Rutgers University finally concluded that "substantial (clear and convincing) evidence exists that research fraud has occurred in several areas" including "biased selection of subjects who were to be included in the symmetry/asymmetry comparison groups so as to artificially obtain desired results." At the time, Brown denied the accusations. He could not be reached for comment.
Even though the paper has now been retracted, Trivers says he is not entirely happy. "It took them 8 years after publication of the paper, and 5 after we submitted a retraction, and 4½ years after we published proof of fraud (later borne out by Rutgers' investigation) for them finally to ‘retract’ a paper now cited 136 times," he writes in an e-mail. "Journals see no upside in admitting fraud--especially intense in the U.K. because of their ‘upper-class protection’ (as they are called) libel laws."