UNLIKE THE problems of children in much of the world -- age-old problems of disease, new problems of ecological disaster -- the problems of children in the United States are overwhelmingly associated with the strength and stability of their families. Our problems do not reside in nature, nor are they fundamentally economic. Our problems derive from behavior.

It has taken a long time for public policy to take notice of this fact. But last month, the Family Support Act of 1988 quietly began full-scale operation, changing the terms upon which the government deals with families that depend on it for sustenance with the hope of encouraging the sort of behavior that can best protect their children.

In the act, Congress laid down a set of mutual obligations. Society owed single mothers support while they acquired the means of self-sufficiency; mothers owed society the effort to become self-sufficient. Absent fathers owed child support to both. Mandatory child-support guidelines are required of every state, something unheard of until now. Wage withholding for child support payments went into effect nationwide.

The act was described as "welfare reform." Yet something much larger was involved. Reform means "to restore to an earlier good state." There was no such earlier good state. Aid to Families with DependentDaniel Moynihan (D) is the senior senator from New York. Children (AFDC) -- welfare -- began as a widow's pension, as a little noticed provision of the Social Security Act of 1935, designed to bridge the period until survivors insurance became generally available under Social Security, as it is today. In the meantime, we experienced a vast, still little understood social change involving a huge increase in the number and proportion of children born out of wedlock. Thus in 1988, as recently reported by the National Center for Health Statistics, the proportion of such births for the first time in our history crossed the one-quarter mark. With some 4 million births per year, this means more than a million such children are now born each year. The AFDC program has cared for these children and their typically youthful mothers.

As this transformation took place, the import of the term "welfare" changed as well. A once honorable standard, a stated public goal, proclaimed in the preamble of the Constitution, welfare became a term of opprobrium -- a contentious, often vindictive area of political conflict in which "liberals" and "conservatives" clashed and children were lost sight of.

The term even disappeared from the ranks of Cabinet departments. Although it was by far the most important function of the Department of Health, Education and Welfare, the term "welfare" vanished when the department was split up in the late 1970s. None wished to be associated with the subject.

The Family Support Act sought to restore respectability to the subject by establishing a new regime to deal with a new reality: Poverty in the United States is now concentrated in single-parent families. Last year, 64.7 percent of all poor families with children were headed by a single parents. Moreover, starting in 1974, for the first time in our country's history and possibly for the first time in any advanced society, children became the poorest group in our population.

Such families, moreover, appear to be relatively unreachable by standard economic policy prescriptions. The annual poverty status report recently issued by the Bureau of Census, for example, provoked concern that the poverty rate seems stuck, that the economic growth and high employment levels of the last decade made no impact on the poverty rate. The number of single-parent families grew. With the advent of AIDS and "crack" the no-parent family appeared.

Newly available data has given us a far better look at the true extent of welfare dependency among children over the course of their young lives. Back in the '60s the Office of Economic Opportunity funded a longitudinal study of family incomes by the Institute for Social Research at the University of Michigan. The researchers have now computed the incidence of welfare dependency among children born in the years 1967-69, the first cohort they tracked.

The findings are dismaying: Almost one quarter (22.1 percent) of children born in the late 1960s were dependent on AFDC for at least one year of their life before reaching their 18th birthday. Welfare dependency is endemic in the United States today. It is a common experience of children. By race, 72.3 percent of black children and 15.7 percent of nonblack children were supported by AFDC at one point or another during childhood.

Moreover, data for children born in the early '70s suggests a worsening trend. In the 1970s, the incidence of dependency jumped sharply among the youngest children, with, for example, about 60 percent of black children receiving AFDC before age seven, and a corresponding increase for white children. Welfare participation probably rose again in the 1980s. No one knows for sure.

But the new fact is that we will know. For Congress wrote into the new welfare law extensive provisions for the evaluation of the impact of the programs, especially the Job Opportunities and Basic Skills training program (JOBS). The object was to establish a set of social indicators that will tell us over time the extent to which child dependency is increasing or decreasing -- and what, if anything, government programs are doing to affect that dependency. Perhaps now we can wring the rhetoric out of the subject and get down to facts.

We are also beginning to collect information on the effects, or the correlates, of welfare dependency. Thus, Nicholas Zill of Child Trends reports that welfare children are twice as likely to fail in school as other children. By the teen years (ages 12-17), 36 percent of AFDC teens have repeated a grade, more than twice the rate for non-poor teens. What all this tells us is that it is not going to be easy to make the JOBS program work, to break the cycle of dependency, to put an end to child poverty. We knew that when the legislation was crafted. We also knew we faced a complex problem concerning teenage pregnancy, the start of the welfare cycle.

One fact is plain: many more unmarried teenagers are giving birth. (Interestingly, this has occurred despite the fact that total birth rates for teenagers have declined dramatically in the last quarter century.) Causes for this increased birth rate for unmarried teenagers include higher rates of sexual activity at younger ages, declining marriage rates for young persons and the mysterious development that our species now becomes fertile much earlier -- four to five years earlier -- today than, say, a century and one-half ago.

All of these factors place a great pressure on conduct among, well, children. Recently, Joseph Berger in The New York Times presented a powerful account of the new proposal by the chancellor of the New York City school system to distribute condoms in the schools. Berger's report contained this passage:

"Sean, a sturdily-built youngster who said he had been sexually active since he was 10, did not think condom distribution would have much impact, since he and the boys he knows do not like to use them.

" 'It's not really a good sensation,' he explained. He might use condoms, he said, only with a girl who is poorly dressed, does not appear to keep herself clean and is sleeping with others. But he said he would not use condoms with his steady girlfriend -- whom he calls his wife -- even if she might get pregnant. 'If it happens, it happens,' he said. 'There's nothing I can do about it.' "

Nor yet his partner or partners. In 1970 a live birth to a mother aged 10 was recorded in New York City.

At minimum these sociological changes mean that high rates of births to young, single women may be with us for a long time. We need to address this problem openly and honestly. To ask the right questions and collect the right data. This is the basic strategy of the Family Support Act. We will not even begin to know whether it is having any effect until the year 2000 at the earliest, perhaps the year 2010. (To those who may wish to protest that this is too long, I would answer that they should have thought of that a quarter century ago when we first spotted this social change.)

But we already see considerable convergence between previously divided schools of social thought. In a decade notable for its divisions, the Family Support Act was notable for the breadth of support it received when enacted in 1988. It passed both Houses with bipartisan support, indeed near unanimity in the Senate, and was welcomed when it arrived on the president's desk. Now, two years later, American federalism, for all its bumbling antique idiosyncrasy, has done it again. Everyone from the Virgin Islands to Guam is in compliance; the act is underway.

Even the social scientists of left and right are beginning to agree. Recently the Progressive Policy Institute issued a fine report, "Putting Children First: A Progressive Family Policy for the 1990s." The report began with a quotation from Karl Zinsmeister taken from the conservative journal The American Enterprise:

"There is a mountain of scientific evidence showing that when families disintegrate, children often end up with intellectual, physical, and emotional scars that persist for life . . . . We talk about the drug crisis, the education crisis, and the problem of teen pregnancy and juvenile crime. But all these ills trace back predominantly to one source: broken families." At the United Nations World Summit for Children in September, the Americans passed out a 34-page document "Goals for the Year 2000: A National Program of Action for Children." There were two subjects: education and health. The education goals were those set forth by President Bush and the National Governors Association a year ago September. To be blunt, they are illusory. Indeed, they are essentially the same education goals set forth by President Reagan in 1984 -- for the year 1990. By contrast, the health goals seem doable -- "Reduce the death rate for children by 15 percent to no more than 28 per 100,000 children aged 1 through 14."

What was missing in our goals presented at the U.N. was any measure of family stability and its various correlates. We have got to start getting specific, asking hard questions and accepting honest -- not ideological, not political -- answers. The French theologian Georges Bernanos put it as well as can be done. "The worst, most corrupting lies," he wrote in 1937, "are problems poorly stated." For a generation now we have avoided this central subject of American life with talk of welfare queens and welfare rights. Can we not get back to the children?

Daniel Moynihan is the senior senator from New York.