Americans love nice round numbers. Anticipation of a 200-yard game, the year 2000 or a 12,000 Dow can make us downright giddy. Try this one: 2,000,000.

The folk at the Justice Policy Institute (JPI) have vetted the trends, crunched the numbers and come up with a nice round prediction. On Feb. 15, 2000, America's prison and jail inmate population will top 2 million.

What is involved, though, is a lot more than roundness, says JPI policy analyst Jason Ziedenberg. "What blew me away when I was doing this research was the whole issue of where we stand internationally," he told me. "Next year, America, with under 5 percent of the world's population, will have a quarter of the world's prison inmates."

An astounding portion of the increase will have taken place this decade. We had fewer than 200,000 adults behind bars in 1970, 315,974 in 1980, 739,980 in 1990. By the end of this year, we will have 1,983,084, having added 61 percent more inmates than were added in the 1980s and nearly 30 times the average number added during the five decades before 1970.

Sounds awful, you say, but isn't it working? Isn't America's crime rate going down, arguably as a direct result of what was done during what the JPI calls "the punishing decade"?

Ziedenberg acknowledges the link between incarceration and crime rates -- to an extent -- but he offers this statistical tidbit:

"A number of jurisdictions -- California, Texas and the federal government -- have had huge increases in incarceration rates, but those are not necessarily the jurisdictions that have had the biggest drops in crime. New York and California both increased their prison and jail populations, but California did so at a much, much higher rate that helped drive up the national total. New York, however, experienced a much, much deeper drop in crime, helping to create the so-called New York miracle."

New York's drop in violent crime was sharpest between 1992 and 1997 (38 percent), when it had the second-slowest-growing prison population in the country -- 30 a week -- and when its jail system was downsized, according to Ziedenberg. During the same period, California's inmate population grew by 270 inmates a week, a 30 percent rate, while its violent crime rate fell by 23 percent.

Nor should it be surprising that the link between incarceration rates and violent crime should be so tenuous. The biggest contributor to prison population growth during this decade has been drug offenses. By some estimates, as many as half a million inmates are behind bars for drug offenses, their last crime being either possession or low-level dealing.

No matter, you say; it's their own fault. It is, of course, but it does matter to the rest of us. It matters because we'll be devoting more public resources to keeping people behind bars, diverting some of those resources from things we'd like to do, things such as improving the schools, which might all by itself go a long way toward reducing crime.

It matters as well because almost all the people we lock up will be back on the street one day, not necessarily less dangerous for having spent five or 10 years behind bars.

And it matters a third way, says Ziedenberg. "It is an important indicator of the sort of society we want to be. We're not only out of step with the rest of the civilized world; this doesn't fit with anything in our own history -- or world history."

There's nothing startling in the approaches proposed by the JPI. They include consideration of alternative sentencing, rationalization of drug-sentence schedules and the institution of a continuum of services to address why so many people are in trouble in the first place.

But what may be needed more than new policy alternatives is for us to step back and look at what we're doing -- what we're becoming -- and ask ourselves how much sense it makes to continue along the present path.