How much money should the grieving families of the Washington area sniper victims receive? Is $5,000 enough? How about $50,000? Or how about nothing at all?

In the aftermath of the shootings, a local nonprofit organization appealed for donations to help families victimized by the attacks. After several weeks, the group had raised $50,000 -- to the dismay of its spokeswoman, who observed that "$50,000 divided among 13 families isn't a lot of money." In fact, it is only a minute fraction of the $1.6 billion mobilized for the families of those who died in the Sept. 11 attacks. Even so, this latest rattling of the tin cup raises uncomfortable questions about whether we should be so quick to seek charitable resources for every new group of victims that violence and terror create.

The desire to compensate bereaved families no doubt is a well-meaning expression of compassion and caring. But considered more fully, the use of donations as gifts for victims of violence is not the best use of private funds. First, it is charity, not philanthropy. Second, it supports the dubious idea that whenever bad things happen to people, they should be compensated for it.

Charity and philanthropy are not the same thing. Before America had organized philanthropy, there was private charity, with modest intentions and a simple formula: By aiding those in need, givers simply sought to ease suffering and provide humane help without rendering judgment on the recipients.

Then came philanthropy. Seeing flaws in the Victorian model of almsgiving to the poor, Andrew Carnegie and John D. Rockefeller, the progenitors of the new "scientific philanthropy" in the United States at the turn of the last century, sought to change the principles of giving. At the center of their philosophy were two powerful notions. The first was that giving could be a positive force when it helped people find ways to help themselves. The second idea was that philanthropy needed to aim at root causes of problems. While charity sought to repair the past by providing temporary help in the present, philanthropy would create permanent solutions for the future. Both ideas had a profound influence on the development of philanthropy over the past century. They inspired a new vision of giving in which impact and results are key considerations.

Philanthropy has enjoyed the upper hand lately, in large part because it promises greater and more lasting effects than charity. Many donors have come to expect results from their giving. So it is curious to see a new class of victims defined as entitled to receive money solely because of a tragedy they have endured. Funds are solicited and given away based not on their expected impact, or even on actual need, but rather on the desire of strangers to express their compassion for a loss experienced by others.

There's nothing new about using charity to help people in crisis; what is new is the idea of using it to compensate those affected by sudden, dramatic acts of violence. Unlike aid for victims of disease, famine and natural disasters, where the focus is on helping those in dire straits rebuild their lives, compensation for victims of violence carries a different set of rules. First, it tends to operate on the mistaken assumption that victims who are part of a "class," or visible group -- the sniper victims, for example -- are more important than those who die alone on the streets every day. Second, it does not target those who are neediest, but assumes that all members of the group are equally entitled. There is no way to ensure that the most deserving victims receive what they need, or that those who can fend for themselves do not receive a financial windfall.

In the case of the Sept. 11 attacks, early relief efforts were aimed at everyone who had lost family members as a result of domestic terrorism. But that sympathetic approach brought problems and conflicts. Groups collecting donations for the D.C. area sniper victims and others would be well-advised to reflect on them.

Given the size and complexity of the private philanthropic response to Sept. 11, it was inevitable that controversies and criticisms would arise. One concern was efficiency: The process of releasing funds appeared to be cumbersome and disorganized. Another concern was effectiveness: whether the money could be better used for longer-term preparedness and the needs of the broader communities around the disaster sites. Finally, there were concerns about equity. They included worries about the perceived overcompensation of survivors of police and fire personnel killed in the attacks, the duplication of efforts undertaken by the federal government to compensate victims, and the possibility that support for non-disaster charities, particularly cultural institutions, would suffer. None of those concerns was ever fully resolved. And they are likely to be raised again and again if charity continues to embrace crime or terror victims as automatic targets of aid. While, in principle, the charitable impulse should be cultivated, such victim compensation is likely to remain difficult to execute well. That does not mean victims of violence need to be left empty-handed. It may well be that government is simply in a better position to address at least the efficiency and equity concerns.

When it comes to efficiency, government is far better suited to distribute funds than a multiplicity of charities acting without coordination and common standards. Given that terror and violence are ultimately breaches in national security and personal safety, government appears to be in a far more logical position to assume responsibility for helping victims, and to do so in a way that is direct and effective.

When it comes to equity and the issue of whether more formal standards of awarding money should be used, having a small army of independent charities raising funds will make it hard to achieve fairness across cases and incidents. Though it is not clear what such standards might look like, victims of tragedies may well begin to demand parity in treatment. The victims of the Oklahoma City bombing, in fact, think they were every bit as much the victims of terrorism as those of Sept. 11, yet there was no comparable outpouring of support. Demands for parity would put charities in a difficult position, because they do not have access to equal resources for every incident. Not so for government.

What, then, is the right choice for someone who, as we say, just wants to help? One option is to focus on donating to nonprofits that work to address the underlying issues connected to a particular crime or tragedy. Another is to donate to a cause or group that a particular victim cared about. Though that kind of giving does not have the emotional appeal of sending money to victims, it does allow donors to translate their compassion into concrete and directed action.

As charities attempt to assert a role in cases of mass violence, we must realize that, while there is certainly a place for gifts inspired by caring and empathy, this may not be the best place for it to manifest itself. Before embracing such an enlarged role for charity, we must confront more directly some of the uncomfortable questions that this kind of giving raises.

The risks of not doing so are great. We may end up creating a culture in which giving becomes too focused on injustices and tragedies, rather than on creating opportunities and innovations. With government ready to become involved in providing terrorism insurance and compensation, the charity world should reevaluate its current direction in favor of its traditional work: helping those whom government does not, and whose material needs are greatest. Peter Frumkin is a senior fellow at the New America Foundation. He teaches at the Hauser Center for Nonprofit Organizations at Harvard University, and is the author of "On Being Nonprofit" (Harvard University Press).