A two-year-old Pentagon program that assigns social scientists to work with military units in Iraq and Afghanistan has come under sharp criticism from a panel of anthropologists who argue that the undertaking is dangerous, unethical and unscholarly.

The committee, which released the report on Thursday at the annual meeting of the American Anthropological Association, the discipline's largest professional group, has been studying the program since its inception in 2007.

The panel concluded that the Pentagon program, called the "Human Terrain System," has two conflicting goals: counterinsurgency and research. Collecting data in the context of war, where coercion and offensive tactics are always potentially present, "can no longer be considered a legitimate professional exercise of anthropology," the report says.

The idea that the military should have a deeper understanding of the cultures and societies in which it operates is one that both academics and Defense Department officials support. How to accomplish that goal is the question.

Commanders in the field have reported that the advisers helped reduce the number of combat operations and enabled units to focus more on nonmilitary needs like local health care and education. Defense Secretary Robert M. Gates has said, "The net effect of these efforts is often less violence across the board and fewer hardships and civilian deaths."

The panel's criticisms are not surprising; the association's executive board had previously expressed serious reservations about the program. Still, it assigned an internal committee to look at the Human Terrain System in more depth. Although political scientists, sociologists, area studies specialists and linguists are also involved in the program, the panel said it focused only on anthropologists.

The report cited insufficient training to prepare scholars for work in the field, concern about confidentiality and obtaining informed consent from the local population, and the possibility that collected research could be used to select military targets. Scholars are supposed to refuse to hand over any data they suspect will be used for choosing targets.

Many people who were interviewed for the report requested anonymity, so it is impossible to assess specific charges of unethical or flawed practices.

Some criticism of the program has also come from inside the Defense Department, from those who produce similar types of sociological and cultural assessments. In the March-April issue of Military Review, published by the Army, Maj. Ben Connable of the Marines wrote that the Human Terrain System approach is a quick-fix policy that "is inconsistent with standing doctrine and ignores recent improvements in military cultural capabilities."

Social scientists working for the Terrain System program have been asked, for example, to answer questions like: How do poor sanitation, health and educational services affect local support for insurgents? Why have young men been forced to leave the country to look for work? Why might children throw rocks at soldiers? How can jobs be created? Are people scared to vote in elections? And what projects should be financed?

Over the past couple of years the Pentagon has actively recruited academics to give advice on how to supplement its military policy. One goal in Afghanistan, for example, has been to strengthen the central government and civilian institutions so that it can counter the Taliban's influence. Gen. Stanley A. McChrystal, the senior allied commander in Afghanistan, said at the end of August that "our strategy cannot be focused on seizing terrain or destroying insurgent forces; our objective must be the population." He added that "gaining their support will require a better understanding of the people's choices and needs."