In her classic study of welfare and welfare policies in Victorian England, Gertrude Himmelfarb reminded us of "the mischievous ambiguity of the word 'poor.' " "Poverty" does not refer to a single condition; instead, it can be used to describe a varied range - with starkly different causes, consequences, and defining characteristics.

It is well to consider this "mischievous ambiguity" as we reflect upon the travails of modern "anti-poverty" policies. America's longstanding attention to the national "poverty problem," and its continuing official efforts to redress it, may at first suggest that we have been grappling for decades with a well-defined, but intractable, social affliction. But this is not the case. Over the past several decades, the nature of "poverty" in the United States has been utterly and completely transformed. Indeed, that transformation has been so radical that earlier generations of anti-poverty crusaders might find our modern variant of "poverty" unrecognizable - or even unimaginable.

A simple "thought experiment" makes the point. Suppose we were transported back to the Great Depression, and managed to buttonhole some of the men or women then in Washington busily formulating President Roosevelt's New Deal. Suppose further that we were allowed to tell them a few things about America 60 years hence -about a world he or she may not live to see, but is striving to influence through Depression-era reforms.

We could offer a glimpse of the future by revealing these six key facts about the American future:

First: The United States would enjoy a tremendous economic ascent in the decades ahead. Between 1935 and 1993, by the estimates of the Commerce Department's Bureau of Economic Analysis, America's gross domestic product (GDP) grew by about 650 percent in real terms - after adjustments for inflation. Over those same years, estimated per capita GDP rose by over 270 percent - nearly quadrupling, in other words. Even measured against the boom year of 1929, America's 1993 GDP would be over six times as large, and per capita GDP nearly three times as great. The economy's long-term performance would significantly exceed the expectations of the best economists of our New Dealer's time. Joseph Schumpeter, for example, proposed that the long-term growth rate of the American economy might be as high as 2 percent a year, and its per capita growth rate as great as 1.4 percent a year, between 1928 and 1978 - and these projections were intended to startle his audience for their optimism. In the event, the American economy grew at an average estimated pace of nearly 3 percent a year between 1929 and 1993; per capita output, at 1.7 percent per annum.

Second: Future generations would never again face anything like the unemployment crisis of the 1930s. In 1992 - a relatively unfavorable "recession" year - the civilian unemployment rate would be only a third of its 1934 level; for the non-farm labor force, the unemployment rate in 1992 would be less than one fourth the 1934 level. By the 1990s, moreover, even long spells of joblessness would not typically raise the prospect of losing one's house, one's car or the opportunity to send one's children to college.

Third: The mechanization of agriculture between the 1930s and the 1990s would, for all intents and purposes, bring a virtual end in America to the age-old burdens of field labor. In 1930, over 21 percent of America's workers toiled in the agricultural sector; by 1993, less than 1 percent of the U.S. workforce labored as farmhands. A New Dealer would deem this great shift auspicious for at least two reasons: 1) because earlier Americans knew farm labor as poorly paying, physically demanding work, and 2) because agricultural mechanization would permit a great movement of the people from the countryside into America's cities (the traditional centers of opportunity and learning).

Fourth: Tremendous increases in life expectancy - a veritable explosion in health - together with changes in U.S. fertility patterns would very nearly bring an end to the tragedy of orphanhood in America. In 1920, by the estimate of researchers in the Social Security Administration, something like 8.5 percent of America's children under the age of 18 had lost their fathers; another 2 percent had lost both parents. By 1965, the estimated share of paternal orphans among the nation's children had fallen by two-thirds, and only 0.1 percent - a tenth of 1 percent -were full orphans. The numbers are lower still, we can be fairly sure, for the 1990s.

Fifth: Court decisions and legislative initiatives would put an end to legalized segregation and legalized discrimination on the basis of race. By the early 1990s, in fact, this grave stain on the fabric of American society had been removed for more than a generation; equality of opportunity - irrespective of race, creed, or color - was at last the law of the land.

Finally: We would have to mention the vast increases in government expenditures for persons in financial need between the Depression and the early 1990s. By the early 1990s, state, local and federal governments would be spending over a quarter of a trillion dollars a year explicitly and expressly on persons in financial need. In 1992, a recession year, the total of cash and non-cash public benefits for persons with low incomes reached $290 billion. That total works out to over $5,600 for every man, woman and child in the lowest fifth of the nation's income ladder - over $22,000 for a hypothetical family of four. Between the Depression and the 1990s, of course, the dollar lost a great deal of its value to inflation: $22,000 from 1992 were roughly equivalent in purchasing power to 2,300 1929 dollars. But in 1929, a family spending $2,300 a year was considered quite well-off. In that boom year, in fact, only about a fourth of all American families managed to consume at that level.

These six revelations about their future would indeed inform our New Dealers about profound social and economic changes in store for America. But would they prepare our reformers for what awaited them in the early 1990s?

Hardly. Just consider what else we would have to tell them to give them an accurate impression of the country's true social and economic situation:

First, we would surely have to mention the failure, for a full generation, to achieve any appreciable reduction in the official index of poverty in America. In 1993, in fact, the officially measured poverty rate for the population as a whole was slightly higher than it had been in 1966, 27 years earlier. Most troubling of all, the official poverty rate for children in the early 1990s was reported to be nearly 1 1/2 times as high as in the late 1960s, and distinctly higher than it had been in the mid-1960s, when the "War On Poverty" commenced.

Second: The progressive rise in the proportion of American children living in fatherless homes. In 1993, only 74 percent of America's families with children had both a mother and a father in the home -a lower proportion than in 1946, despite the disruption imposed, and the casualties inflicted, by the World War II. By 1993 nearly a quarter of the nation's children under 18 years of age were living in a female-headed household; over 40 percent of the country's children were in families that did not include their biological father.

Third: That over a million persons, at any given time, were incarcerated in correctional facilities in America in the early 1990s - a figure that speaks to the huge increase in criminality since the Great Depression. In 1992, 1.3 million prisoners were being held at any given moment in local U.S. jails and state or federal prisons. Over four times as many persons were serving time in prison in 1992 as in the early 1960s. In 1992 nearly 12 million arrests were processed for criminal offenses - three times as many as thirty years before. In 1990, roughly 1 adult American in 40 was under "correctional supervision": in jail, in prison, on probation, or on parole. For men over 18 years of age, that ratio was 1 in 24; for black adults, 1 in 13; for black males over 18, roughly 1 in 7.

Fourth, and by no means least: That a higher fraction of Americans were on "relief" - or what we now call "means-tested public assistance" - in the early 1990s than had been in the depths of the Great Depression. The U.S. Census Bureau publishes an annual report on poverty in America; its estimates are instructive. For the nation as a whole in 1992, 24 percent of the population in households that received some form of means-tested assistance. Among African-Americans, the proportion of the population in households receiving some form of means-tested assistance was nearly 53 percent; over 28 percent were obtaining means-tested cash assistance. By contrast, in January 1935, according to the National Resources Planning Board, an estimated 26 percent of the country's black Americans were on relief. Lest one presume that the pattern is particular to American blacks, we can look at the trends for the white population. That same Depression-era report estimated that 16 percent of American whites were on relief in January 1935; by 1992, according to the Census Bureau, nearly 20 percent of the country's whites were seeking and obtaining some kind of means-tested assistance. In 1992 over 30 percent of America's white children under six years of age resided in a household receiving at least one type of means-tested assistance.

Now, what would New Deal reformers make of these facts? How would they reconcile what we had told them about the future with these unpleasant, but unavoidable, realities? We may guess that, from the perspective of a New Deal reformer, there would be something very, very wrong with the picture of modern America.

Despite tremendous material advances, revolutionary improvements in knowledge and technology, and a vast augmentation of national wealth, the country's domestic social problems have by no means been eliminated. Paradoxically, in the early 1990s such problems as crime, dependency and family breakdown were far more acute than they had been during the Great Depression - when general income levels, and general levels of schooling, were so much lower. For a reformer or an idealist from an earlier time, our contemporary social problems would be especially troubling because they would be so completely unexpected.

Ours is now a nation whose social and economic problems are no longer familiar, nor in any real sense traditional. Modern America is a country inhabited by large numbers of prosperous paupers and affluent savages. As mass phenomena, these are quite new. The patterns of mass behavior attendant upon these syndromes constitute major problems for our society today. It is precisely these patterns, in fact, that account for much of what would surprise or dismay earlier eyes on surveying our contemporary landscape.

America's new "poverty," moreover, is highly resistant to the instruments that have been traditionally employed to alleviate it. The new poverty does not exist for lack of money, or education attainment - quantities government can provide. There is a great need underlying the new poverty in America - but it may be for something that our government cannot provide. Until we face these facts, little progress can be expected in our ongoing national struggle against "poverty."

Nicholas Eberstadt is a researcher with the American Enterprise Institute and Harvard University. This article is adapted from an essay in the current issue of Society magazine.