At present, 57 percent of Georgia's high school graduates are HOPE-eligible scholars. The B-average standard is a phony. Course work's not the same, F's effectively don't count and B's are cheap. Eventually, end-of-course tests will reveal whether students are getting it.

In the meantime, some objective measure, such as the SAT, is needed. Otherwise, HOPE's not a merit-based scholarship. It's just another welfare-state entitlement.

A reasonable standard for a merit scholarship would be a legitimate B, plus a 1000 SAT. Georgia's average SAT is 984. The reality is, though, that 39 percent of HOPE scholars score below 1000. That number includes 67.4 percent of blacks and 32.4 percent of whites.

That's not to say two-thirds of blacks and one-third of whites would lose the means to pay for higher education. The federal Pell grant for low-income students provides an average of $2,457 per year. Georgia supplements that with a new program called LEAP (Leveraged Educational Assistance Partnership), which provides Pell grant students up to $2,000 per year.

Tuition and fees at the most expensive public colleges in Georgia average $3,515. The state's most affordable junior college, East Georgia College in Swainsboro, has tuition and fees of $1,428.

The problem driving the HOPE upgrade is that the state's gambling enterprise projects flat to slow growth, while scholars continue to bloom. Within four years, the gap is projected to be $250 million. Filling the gap with tax revenues would be foolish, merely delaying needed decisions about eligibility standards.

The question becomes how to deal with the racial disparity. One possibility is to simply impose reasonable standards and hold students to it. That's ideal. It would take away many opportunities to game the system on grades. It would, too, encourage discipline and study, returning merit to HOPE.

Another option would be to set the standard and phase it in over four years. Students need to know by the eighth grade what they need to achieve for college admission. Starting at 850 -- a ridiculously low SAT standard (only 18.2 percent of HOPE students who score below 900 keep it after one year) -- the HOPE threshold could be raised over four years to 1000, lessening the racial disparity.

A third option would be to raise the standards for HOPE, but to add a new service-based stipend to supplement LEAP and Pell.

The state's very serious problem is that black families are making choices that leave students unprepared for college. Only about 1 percent of Georgia's high school graduates are blacks who qualify for admission to the University of Georgia, for example. Blacks are 28.7 percent of the population and 1 percent of the UGA eligibles. What happens?

The problem starts before birth. Seventy percent of black children are born into unmarried households. From there, life is a desperate upward struggle against heavy odds. For the underclass, the odds may be insurmountable.

Schools, as now structured, cannot possibly salvage children who have been neglected and abused from birth, who exist in worlds where TV is reality and the normal two-parent family is not. Meanwhile, government workers of some stripe -- social workers, teachers, counselors, police officers, probation officers -- drop in, trying to be the family that never formed.

High school and college students could help. As tutors. As big brother counselors. As wiser heads available to children who need friends. In return for that service, students could earn vouchers of $1,500 or $2,000 for college. Work for aid.

It's not ideal. But it is important to get underclass blacks to value education and to think of college. HOPE standards that exclude two-thirds won't fly politically.

Our challenge, then, becomes to define the behaviors we want in students -- academic seriousness, individual responsibility, personal achievement, upward mobility -- and to design programs to encourage them. Leaving a phony scholarship program unchanged is not the way.