Marx and Engels called them “social scum.” For Alexis de Tocqueville they were “rabble” who, Thomas Jefferson had said, “add just so much to the support of pure government as sores do to the strength of the human body.” They peopled the novels of Balzac and Dickens, and were called “the dangerous classes.” Nowadays we tend to be less judgmental. Hence Ken Auletta’s use of the term “underclass” to describe our current version of this historic stratum.

I confess to mixed feelings about his phrase: it suggests a sociological perspective which may not in fact be there. However, since I lack a better term (“lumpenproletariat” has special connotations) I will go along with underclass, and without quotation marks. Auletta focuses on people he calls “social drop-outs,” by which he means a class below all other classes, and qualitatively different because its members “do not assimilate” to a mainstream way of life. He emphasizes that only a fraction of low-income Americans should be classed as underclass. His concern is with the nine million or so men and women who, unlike others of the poor, “suffer from behavioral as well as income deficiencies.” Among those so afflicted are street criminals and drug addicts, welfare mothers and teenaged truants, plus the “drunks, drifters, homeless shopping-bag ladies and released mental patients who frequently roam or collapse on city streets.” Their shortcomings of character and conduct form the core of his book.

Auletta came to know a cross section of the urban underclass by sitting in on a job-training program conducted in Manhattan. There were twenty-six people in the course, which sought to instill the kinds of skills and habits most adults already have. In some fascinating chapters he relates what he saw and heard over a period of seven months. He also talked with people considered experts, both academic and applied, and cites most of the standard studies on poverty and crime. The Underclass is a valuable book to have, and I do not fault Auletta for failing to arrive at conclusions, let alone solutions. If I suggest there are aspects of the problem he has overlooked, this seems to show that he challenges each reader to pull together pieces of a very vexing puzzle. Since most Americans manage to keep their heads above water, and many do even better, how are we to account for those who don’t?

Most poor people do not descend to crime or end up on the welfare rolls. Hence the difficulty in citing “environment” as the cause. Here are two brothers, brought up in scabrous surroundings. One supports his family by ironing linen in a laundry; the other is in prison for some brutal robberies. Of two sisters from next door, one has seen two children through college, while the other has started her fourth welfare pregnancy. Conservatives grant that slum life can be tough; but if some people lift themselves out, why can’t all the rest? Hence the emphasis on “character” and contempt for those who don’t show greater resolution. Liberals admit that many make it, but add it is too much to ask that all overcome such hurdles. Radicals tell us not to blame the victims; the culprit is the system. But liberals can despair of sloth and violence within the underclass. Still, their remedy remains better “services” to compensate for disadvantages. Conservatives prefer firmness to compassion if we are ever going to suppress destructive traits.

The distinctive feature of any under class is that its members do not conform to the conduct expected of the poor. This country has always had poor people, yet on the whole they have been well behaved and remarkably uncomplaining. They have labored at unappealing tasks, maintained stable families, and disciplined their children. Indeed, for the first half of the current century, the poor met these expectations; which is to say that during this period—including the Depression—the nation did not have an underclass. Auletta generally ignores historical changes, but some understanding of that time may show why we have an underclass now.

By 1900 America’s cities, which earlier had been savage places, were generally safe and surprisingly civil. Even though the slums were oppressive and took a horrid human toll, there was little crime of the kind we know today and in hardly any cases were its victims middle class. The groups that had been violent—most notably the Irish—had by 1900 turned respectable. The next wave of immigrants, largely from Eastern Europe and southern Italy, were more passive to begin with and accepted the conditions they found on their arrival. They may not have liked being poor; but they did not inflict their resentments on the rest of society. Even industrial disputes were remarkably well mannered, with violence coming first from the other side.


Another, less publicized, group of newcomers were from inland rural areas, and brought with them the values of small-town America. People went to work, and would take what they could get. Hardly anyone dropped out, so even the most impoverished became part of the general system. This deportment by the poor continued despite the unemployment of the Thirties and for a decade after the end of World War II. The same held for the postwar arrivals from the South, Puerto Rico, and Mexico. They also came from rural cultures, where they had been sharecroppers or field laborers, and where churches and extended families made for strong community ties. Given this background, the adults in this migration were prepared to join the working poor, as most of them did. It was their children, new to the streets of the cities, who became the first cohort of our current underclass.

The first signs came with the youthful gangs that emerged in the mid-1950s. Here were young people who rejected the traditional restrictions placed on adolescents, along with the authority of their parents to impose them. Just why they did so has never been satisfactorily explained and it is not a question that concerns Auletta. But it can be argued that the sexual revolution started in the slums, as did the spread of drugs. The violence that began to scar the streets preceded the protests of the Sixties by middle-class youths. Controls that had held sway for half a century were rapidly eroding. This was a national phenomenon, but the rebellion took different forms in the slums and the suburbs.

Earlier generations had an effective way to deal with youthful energies: it was to put young people to work as soon as possible. In 1910 adolescents could legally leave school at the end of the eighth grade for full-time employment, which is what most of them did. In New York, which then had a population of 4,766,883, the city’s high schools taken together produced a graduating class of 2,477, of whom 1,514 were girls. Most of their erstwhile classmates were putting in nine- to ten-hour days for $4.00 a week.

Since that time, of course, child labor has been supplanted by compulsory schooling. The trouble is that educational authorities have failed to find forms of instruction that can hold the attention of lower-class teenagers. They are still expected to sit dutifully at desks, reciting assignments that make no sense to them. No wonder, as one expert said to Auletta, “They can’t cope educationally. Don’t ask me why. They can’t hack it.” Well, there are good reasons why they can’t and don’t. The best was expressed by one dropout: “The teacher wasn’t teaching nobody nothing, so I just stopped going.” They are turned off by courses they must take and the way they are taught. Middleclass students, from moderately literate homes, discover how to satisfy requirements with a minimum of effort. Their counterparts in the slums either stop showing up, or at best come unprepared and gaze into the air. Of course some do get through; but if they were the rule we wouldn’t have an underclass.

I am not arguing that required schooling turns teenagers into criminals or encourages early pregnancies. My point simply is that unlike earlier generations we lack effective ways to harness all that energy. Nor is it clear that more job opportunities are a solution for this newer generation. We are often told, for instance, that minimum-wage regulations prevent employers from hiring young people. Yet even supposing such low-wage jobs were to materialize, there remains the question of how many of today’s teenagers would take them and hold them.

Our best models here are our latest group of immigrants. Koreans work long, stolid hours in their fruit and vegetable stores. Vietnamese families save up and buy a shrimp boat in which everyone pitches in. Greeks are at it round the clock, often holding three different jobs, and soon own their own homes, even some others down the block. Whether from Haiti or Hong Kong, these new arrivals mirror the vigor and discipline of their earlier counterparts.

But indigenous Americans see themselves in a different light. “We’re not willing to work for less money,” a highschool student told Auletta. “They’re willing to work for less money.” In this sense, at least, the underclass has assimilated American attitudes. A proper American does not labor a full day for several dollars and change. This is why we need immigrants, legal or illegal, no matter how swollen the ranks of our native unemployed.

It is important to ask how race should figure in any discussion of an underclass. In the course Auletta attended everyone was black or Puerto Rican, and these origins kept recurring in his account of the group. So let it be said at the outset that when race does arise, he is not suggesting that underclass behavior has a genetic basis. The point rather, is that “being black” is in largest measure a condition that white America imposes on a certain segment of our population. People who get this designation tend to end up congregated in settings where handicaps are difficult to overcome. It is this semi-enforced concentration that makes being black and poor more debilitating than being white and poor. I feel obliged to say all this because Auletta never does. He has a chapter describing poverty in Appalachia, where welfare and long-term unemployment have created a white underclass with many characteristics of its Harlem parallel. Even so, Appalachia often seems an isolated island, and an incestuous one at that.


Indeed, we have no serious study of the while poor of this nation. At one point Auletta quotes Oscar Lewis to the effect that “by the time slum children are age six or seven they usually have absorbed the basic values of their subculture.” Yet when we read a phrase like “slum children” we immediately see them as black or Hispanic in a “subculture” identified by race. Indeed, we find it hard if not impossible to visualize a white slum. So while we realize that not all whites are middle class or well-paid working class, those who are poor or part of the underclass come across as exceptions; and their condition cannot be linked to a “subculture” save in special cases like Appalachia.

If we cannot point to peculiar characteristics which put certain whites in the underclass, there are some figures that suggest their proportions. For example, 31.4 percent of the persons arrested for the typically underclass crime of robbery are white, as are 46.1 percent of the inmates of local jails. So are 61.2 percent of the young men who drop out before the junior year of high school, and 40.0 percent of all families on the welfare rolls. And of adults aged twenty-two to forty-four with incomes below the official poverty line, 57.0 percent are white. (In all these percentages, I have excluded people of Hispanic origin from the “white” totals.) My concern is that by stressing a peculiarly black “tangle of pathology”—Daniel Patrick Moynihan’s phrase—we stunt our understanding of who ends up on the bottom. One illustration will suffice. We know that while 9.4 percent of all white births are out of wedlock, the proportion for black births is 56.0 percent, a figure often regarded as pathological. But it is also worth noting that in Oregon 12.5 percent of white births are to mothers who are not married, whereas in Maine the proportion of black births out of wedlock is 11.7 percent.


According to Auletta and many other commentators, a major factor in the growth of an underclass has been the “feminization of poverty.” By this they have in mind the increase in households headed by women, whose incomes tend to cluster at the low end of the income scale. In the past we were less apt to think of women as being poorer because most lived with their husbands and shared in their incomes. Today we have more women on their own not only because divorce and separation rates are higher, but because the men are more likely to remarry.1

In economic terms, the median income of divorced women is 70.2 percent of that for divorced men; and among separated persons, women’s incomes are 57.6 percent of men’s. In fact the phrase “feminization of poverty” reveals a rather obvious bias. After all, it is usually the man’s departure that decreases the woman’s income, whether he leaves right after assisting in a conception or many years later.

It has become commonplace to comment on the fecundity of the poor. Auletta cites Moynihan again, this time on “the freedom now by and large enjoyed by low-income groups to produce children they cannot support.” Part of the complaint is that society will end up paying the costs, which are not only monetary. But the indictment also implies that the right to sexual enjoyment is one that should be earned, which the poor have not done. In fact, we have no reliable data on whether the poor are more sexually active than other social classes. Whether they are or not, it is clearly more difficult to bring them to use reliable methods of birth control, or abortion afterward. As one social worker told Auletta, “Lots of girls feel that if they get to be eighteen and they don’t have a baby, they’re not a woman.” She might have added that a lot of men do not feel they have proved their masculinity unless they have sired children with several different women. More than that, a further mark of manhood lies in being free to leave these offspring with their mothers. Better-off men do this also, but with the difference that more of them foot at least some of the bills.

Mothers on their own may choose welfare. Many people have concluded that the availability of welfare encourages husbands to leave home, knowing that public stipends will provide for their wives. Others go further, suggesting that women deliberately have more children to get larger checks. While there evidently are such cases, it is not clear they play a significant role in the growth of welfare rolls.2 At the same time, however, it does seem true that some pregnant women decide to have and keep their babies because they know support will be forthcoming. Between 1970 and 1980 the proportion of women-headed households in which the mother had never been married almost doubled, from 8.6 percent to 16.4 percent. Auletta worries a lot about welfare because he sees it as a breeding ground for the underclass. We certainly know that among Americans who drop out, a disproportionate number come from broken homes. But the real question is whether welfare-supported homes differ from one-parent households whose incomes comes come from other sources.

Between 1950 and 1980, the number of families receiving funds from the Aid to Families with Dependent Children (AFDC) program grew from 651,000 to 3,843,000, a sixfold increase. During this period the number of mothers on their own rose by a factor of four, from 1,256,000 to 5,030,000. Put another way, in 1950 there were 518 welfare households for every 1,000 solo mothers; by 1980 that ratio had risen to 764 per 1,000. The justification for the welfare option is that mothers with children at home should not be forced to work. They are, as one government report asserted, “needed in home fulltime as homemakers.” Women who have wage-earning husbands are not expected to take jobs; even now not many with young children have full-time employment. A welfare mother is in an analogous situation, with her allowance coming from the government rather than a resident husband. Is there anything wrong with that?

Before we pass judgment it would be well to look at what we know about the welfare population. As of December 1980, it consisted of 3,842,535 families with 7,599,376 children, representing 12.4 percent of all households with children under eighteen and 11.1 percent of all children of that age. The most recently published full-scale study of “recipient characteristics” was conducted by the Social Security Administration in 1977. Its findings are revealing, and on the whole we can assume they still hold true. To begin with, it turned out that only 69.7 percent of AFDC families had a mother as the head. In 13.1 percent the father was in fact present and received the checks; and in another 12.9 percent the householder was one or another grandparent. The remaining 4.3 percent were other relatives or people not related to the children.

Altogether, 40.3 percent of the families had only one child and 27.3 percent had two, which means that two-thirds were of fairly modest size. Only 16.3 percent had four or more children. Moreover, 85.0 percent of all the households had at least one child still under twelve, a figure to be noted when the work option is raised. As it happens, in 13.8 percent of the families the householders were employed on a full-time or part-time basis, but received welfare because their wages were very low. In addition, 48.3 percent had been on welfare for less than two years, and 20.1 percent for three or four. As for longer-term recipients, 18.5 percent had been on for five to ten years, and 5.4 percent for over ten.

Among the mothers who headed the household, 56.3 percent were divorced or separated, and only 3.1 percent were widowed. A depressing 40.6 percent were women who had never been married. So far as the children were concerned, 14.9 percent still had their father in the home, although in the majority of these cases he was disabled or unemployed. For another 32.4 percent, the father had been present within the last five years. However, for 18.1 percent he had not been in the home for five years or more; and with 34.6 percent he never had been.

As many as 33.8 percent of the householders were high-school graduates, and 5.6 percent had attended college, which puts 60.6 percent in the dropout category. In ethnic terms, 43.0 percent of the households were black and 12.6 percent were of Hispanic origin, while the remaining 40.0 percent were white and 4.4 percent had other or unstated backgrounds.

These figures show that while some families fit the underclass pattern of long-term welfare recipients, with large numbers of children and a black unmarried mother, many if not most do not fit that mold. Plainly, a lot of people take themselves off welfare after a year or so on the rolls. Indeed, at any given time 46.8 percent of women who are legally eligible are not receiving such assistance. Who goes on welfare and who does not depends largely on class, or more precisely on the kind of job she is able to obtain. Our most recent figures here, for 1979, show that of women on their own with children in the home, 20.5 percent had incomes between $10,000 and $15,000, and 21.5 percent were over $15,000. In view of welfare ceilings, sums like these have to come from earnings.

If a woman cannot get that kind of job, she is better off on welfare, which is not only tax-free but includes food stamps and Medicaid benefits. As it happens, another, less well-known, option is available. It will be recalled that 12.9 percent of welfare checks go to grandparents. In fact, in Georgia that figure is 19.6 percent, with 18.2 percent in Florida, and 17.8 percent in Texas. This happens because many working mothers send their children to live with grandparents, not an easy step to take but indicative of their wish to escape the underclass.

Is welfare bad for its recipients? Generally speaking, we have a shifting standard. As was noted, if a woman is able to keep a wage-earning husband, we do not ask that she work, and some even add that her proper place is in the home. But if she loses—or leaves—such a husband, or never had one, the standard changes and she is expected to work, even if she retains custody of the children. While we do not worry about the “character” of the housewife with a husband, the “dependence” of the welfare mother unsettles many people, and not only conservatives. Hence the view that she ought to go to work—as middle-class mothers do—not simply to save the taxpayers’ money but for her own sake as well. (Liberals temper their position by calling for high-quality day care.)

Work is undoubtedly what Thomas Sowell had in mind when he told Auletta, “I imagine that we could absorb every unemployed person as domestics or babysitters.” That was what happened in the past, when welfare did not exist or was difficult to obtain. As recently as 1940 there were 2,277,000 women employed as household workers, enough so that even during the Depression every middle-class family had domestic help.3 If that proportion were restored today, we would have almost four million domestic servants, enough to empty the welfare rolls. (The current figure is about one million.) An additional assumption is that were mothers made to work, that regimen would serve to discipline their children. One person from a welfare background remarked to Auletta that nowadays “young males rarely fear their mothers.” Were she to work, this reasoning goes, she would attain their respect.

Finally, there is the worry that among all too many recipients, taking welfare is no longer seen as a stigma. In the past going on relief was considered shameful, a feeling the authorities did everything to encourage. In recent years, the underclass has evolved its own value system, which regards assistance as a right, akin to public entitlements claimed by other classes.


Auletta makes “street criminals” a special category within the underclass. He says nothing about members of organized syndicates, wholesale drug dealers, or such specialists as art and securities thieves, presumably on the ground that they are really middle class. Most street criminals are poor and stay so: note how few can raise the cash for bail once they are arrested. At the same time, it is worth asking whether street crime is a principal career for more than a small group of people.4

Satisfactory statistics are not easily come by, because it is hard to know which people go in the “criminal” category. For example, a total of 548,809 robberies were reported in 1980, and an equal number probably went unreported. But we have no idea how many people committed them. Similarly, in 1980 there were 139,476 robbery arrests; but we don’t know how many separate crimes those charges represented. At the end of 1981 there were 369,000 persons in state or federal prisons, serving at least one-year sentences for serious offenses. While not all their crimes were violent, these are people who will go through life with a prison record. Moreover, we can generally assume that at one time or another most people who engage in violent crimes will serve some prison sentence, particularly if they are poor.

Even so, Auletta cites figures to suggest that a rather small contingent makes prowling the streets their full-time profession. One Philadelphia study found that about 1,800 persons committed 95 percent of all the city’s robberies. For New York it has been estimated that 70 percent of the robberies were carried out by approximately 2,000 people. Still, it doesn’t take many people, as one policeman put it, to “turn streets into nightmares.”

For most Americans, “crime” now means street crime, which in turn means face-to-face encounters with someone prepared to kill you, perhaps even on a whim. Hence the wish that they could be somehow rounded up and sent somewhere for good. All this hardly bears reiteration, and what we need is further insight into why we have so much violent crime. While Auletta has a chapter summarizing recent writing, I think this is one field where too many studies obscure some simple truths.

Just as all women left on their own do not go on welfare, so all men in difficult circumstances do not turn to robbery. If we want to know which ones do, it is best to begin with a basic statement about the nation as a whole. Within the American population there is a group of people who might be called persons with larcenous proclivities. These individuals can be found in all classes of society, including the very rich. This proclivity of theirs leads them to seek out money through unaccepted channels, generally by ways other than honest work. We do not know why some people pick up this proclivity and others don’t, although being exposed to it as a widespread practice certainly helps. Except in extreme conditions, people do not steal out of dire need. They do it because you get more with less effort than you do by working.

In the middle class and up, these people can perform their thefts without inflicting physical harm, since their employment gives them opportunities to manipulate paper. Auletta quotes Charles Silberman’s reminder that “well-bred people steal far larger sums than those lost through street crime.” While that is fairly common knowledge, Americans still want a man who has robbed ten people of $50 each put away for a considerably longer period than a bank officer who slips $5,000 out of ten different accounts. The reason, needless to say, is that the violence of the former crime, even if only as a threat, multiplies its evil in our eyes. Hence the proposal that if our prisons are overcrowded, we should reserve the spaces for violent offenders. The banker, after all, is really one of us and can make restitution in other ways.

However, those among the poor who have a proclivity for larceny have quite limited choices. People with good working-class jobs can steal tools and materials from factories or building sites. But there really isn’t much a poor person can do except lurk about the streets or hold up a liquor store. I might add that most poor people who choose to commit crimes do not enjoy outdoor work. The risks of getting caught can be high and the proceeds are usually small. They would much rather be doing embezzlement in an air-conditioned office. One reason why we have so much street crime, then, is that we do not have enough indoor jobs to accommodate everyone who wants to make something on the side. We might have safe streets if we could give the poor the same opportunities for pilferage as the middle class has. As matters now stand, we direct most of our deploring to the kinds of crimes the poor commit, and are most emphatic in asking them to exercise self-restraint. Or at least until they have raised themselves to where they can fiddle a computer.

Is there something about America’s economic system—its capitalist foundation or its commitment to technology—that fosters the creation of an underclass? Auletta rejects the Marxian explanation that underclass people are a pool of “surplus labor,” only offered employment when extra hands are needed. I would agree that underclass people tend to be nonemployed rather than unemployed, being outside the labor force rather than out of work. At times when more employees are needed, it is not the members of the underclass who are asked to fill the spots. We are more likely to use illegal or semi-legal aliens, or to ship the work to cheaper labor abroad.

Of course it can still be in the system’s interest to have an underclass around, to show what can happen to any one of us if we fail to maintain a productive position in the economy and society. Moreover, in so far as the poor are seen as a major source of problems, their presence shifts attention from arrangements that cause even greater dislocations. A public preoccupied with crime and welfare cheating will spend less time scrutinizing the practices of corporate capitalism.

While Auletta alludes to “structural causes of poverty” all he really does is quote various writers who call for guaranteed jobs and incomes and a redistribution of wealth. In fact, the “structural” problem has chiefly to do with the kinds of employment our economy has chosen to create. Too much can be made of the argument that we no longer “need” as many unskilled workers as in the past. As has been noted, the jobs taken by immigrants and aliens show that there are still many openings for low-wage labor. (The big decline has been for well-paid blue-collar employment.) More to the point is how the economy has organized itself to reach its conception of optimum performance. Its decisions here have less to do with the demands of technology than with other conventions concerning personnel needs.

Before exploring this phenomenon, we should note that unemployment is not the major variable. In the postwar period, the unemployment rate has varied from 3.5 percent (in 1969) to 9.5 percent at this writing. If we consider the former as tolerably close to “full employment,” then there are about 6.5 million currently jobless people who would not have been so in a better year. It is these people, at the edges of the labor force, who comprise the real pool of “surplus labor” in Marxian analysis. They are not part of the underclass, but employable persons at the bottom of the hiring list. While it is true that even well-paid people can find themselves out of a job, unemployment always takes a greater toll at the lower levels. In April of this year, with the rate for factory workers at 16.9 percent and laborers at 19.2 percent, for professional and technical personnel the figure was 3.2 percent and among managers and administrators it was 3.3 percent. I do not wish to minimize the impact of unemployment. But it is not—thus far—a crucial element in the creation of an underclass.

“It is true,” Auletta remarks early in The Underclass, “that the American economy generates too few jobs.” Unfortunately he does not go on to ask why this is so. Our most recent figures are for 1980, during which, on an average day, the economy had available 97,270,000 paid positions. Throughout 1980, a total of 116,178,000 people were employed at one time or another, which means that in many cases several people in succession might fill a single job. (Note the turnover behind your local McDonald’s counter.) More important, among those who worked, only 64,740,000 people—55.7 percent of the total—had full-time jobs for the entire year. In short, in 1980 the economy created fewer than 65 million full-time jobs, of which just about half paid over $15,000 and half paid less.

Is that the best we can do? One way to tackle this question is to draw on a notion that goes back to classical economics. David Ricardo and his contemporaries posed the existence of a “wage fund,” the pool of cash a country makes available for wages and salaries. In 1980, the sum Americans received in such earnings totaled over $1.4 trillion. (They also had nearly $322 billion in other forms of income.) Just how an economic system goes about deciding how many people will partake of this fund, and at what levels of compensation, is a conundrum economists have never really answered.

For our purposes we may note that the 64,740,000 people with full-time employment received 80.2 percent of the total wage fund, leaving 19.8 percent for the other 51,438,000 who worked for shorter spells. (The former averaged $17,570; the latter, $5,440.) One fact which determines the number of participants in the pool is how many men and women receive higher than average wages. The fact that 7,552,000 people each made over $30,000 in 1980, for example, meant that they used up money that could have been devoted to putting more people on the payroll. (This becomes even more egregious when both partners in a household have $30,000-plus salaries.) In short, an economic system can decide how many jobs—or how few—it wishes to create from its $1.4 trillion fund. On the whole, America’s economy has chosen to create comparatively few.

I have no wish here to open the question of whether we “need” to pay some people more than, say, $30,000 to ensure that certain things get done. Nor do I want to consider whether “natural” market forces bid up certain salaries.@f5@cHowever, I am prepared to argue that we should not expect any real expansion of that $1.4 trillion fund or its current dollar equivalent. At best, it might be maintained that with so much of our industry operating at less than capacity, were more people put to work more goods and services would roll off the lines. However, the problem is how much of that production would be net growth and how much would be eaten by inflation as happened in the 1970s. It is safer to assume that $1.4 trillion is a fixed sum, so that if we want work for the underclass and others not currently employed, then those now employed would have to share their wages.

Suppose, then, that we really want to provide employment for all liberated housewives, bored high-school students, energetic senior citizens, and persons with criminal or drug addiction records, plus a fair proportion of the physically handicapped and mentally impaired. What work might these new entrants to the labor force do? It is easy enough to reel off tasks like cleaning up graffiti or tutoring slum children. The difficulty is that while it might be nice for such things to be done, at this point we do not want to pay out real cash—i.e., diminish our own incomes—to have them accomplished. Indeed, putting more people on payrolls would require some redefinition of what constitutes a job. On the other hand, it bears mentioning that even now we do not require that all positions be productive. Many presumed professional posts are largely decorative and not a few vice presidencies are chiefly ceremonial.

Still, many industries would argue that they do not need more workers than those on their rolls. Even were Procter & Gamble to call back its laidoff people, it would not require more than the 60,000 force it normally has. Were it asked to take on another 10,000 (say, with public subsidies) its organization and technology are such that the extra employees would be more trouble than they are worth. We all have been told how improvements in technology can cut back on labor needs. Indeed, a few firms have found they reach optimum earnings with fewer employees. Between 1967 and 1981, Shell Oil’s payroll actually declined while its sales more than doubled in constant-value dollars.

The 1980 Census counted approximately 154 million people between the ages of sixteen and seventy. That as many as 116 million of them did some kind of work suggests that the problem may be less one of making more jobs than of extending their duration. (I would guess that even Auletta’s most dismal dropouts put in some weeks here and there, washing dishes or mopping floors.) But let us imagine that we wanted to create full-time employment for 125 million people, without for the moment asking what they would in fact do. If we did, the average annual wage from the $1.4 trillion fund would fall to $11,200, which would be 36.3 percent less than the $17,570 averaged by fulltime workers in 1980. Of course some families might come out ahead, as additional members entered full-time employment. At least they would if husbands agreed to work for less, so as to open up positions for their wives and children.


Every society that has poor people must devise methods of control. In America, the presence of police and prisons has always been important. Also, a minimal set of services can damp down discontents. But the most effective means is to persuade the poor to impose controls on themselves. They should, of their own volition, regard hard and regular work as a badge of self-respect and a promise of escape. To some observers, the growth of an underclass suggests that many poor people are not showing the character they did in the past. Conservatives believe that the loss of this trait is due to coddling, by courts and welfare officers and liberal policy makers. If assistance were withdrawn, people would show their mettle. If punishment were harsh and sure, wouldbe criminals would straighten out. In this respect conservatives are egalitarian: they believe everyone is capable of inner strength, and that tough conditions are the way to encourage resolution. Liberals tend to look on force as an admission of defeat. And talk of character makes them uneasy: it smacks more of McGuffey readers than current social science. Hence their preference for therapy and education as vehicles of change.

Whether some people are inherently stronger than others will always be a riddle, as will the roles of heredity and environment and chance circumstance. What we do know, though, is that most of those born into a class that provides a measure of security will never have to worry about whether they are strong or weak. A moderate amount of effort will keep a college graduate at a respectable plateau. Still, American society is a competition, and it is possible for its losers to drop to the very bottom. The underclass is evidence enough that there is no safety net, unless by it we merely mean forestalling starvation. However, weakness of character is not a sufficient explanation for the group of underclass people Auletta calls “the traumatized.” Ann Marie Rousseau’s Shopping Bag Ladies makes that very clear.

Her book mainly lets the women speak for themselves, eighteen of them in Boston, San Francisco, and New York. Some literally carry all their possessions around; others stash them in coin lockers. They sleep on station benches when they can, in toilet stalls when they cannot. Shelters are not always welcomed: even their shoes may be stolen. Rousseau clearly gained their trust, for they talked to her freely. Most were in their forties and fifties, and come from good working-class or middle-class families. One woman had a high-pressure job with the American Broadcasting Corporation. Another was married to a Pentagon civil servant. A third’s husband was an American Airlines pilot. Their most common experience was a mental breakdown, a period of hospitalization, and then release. But not to their families. Their marriages were over; parents or children, if they had them, felt they could not take them in. So they roam the streets and sleep where they can. Those who manage to get public assistance often lose or don’t receive the checks, so they are unable even to maintain a single room in a derelict hotel.

The chief reason these women are now seen so much on the streets derives from a bold liberal step taken twenty years ago. The idea was to use newly developed drugs so that mental patients could learn to function on their own. President John F. Kennedy called for the creation of local centers that would assist with the adjustment. “When carried out,” he said, “reliance on the cold mercy of custodial isolation will be supplanted by the open warmth of community concern and capability.” Between 1960 and 1979, the patient load in psychiatric hospitals dropped from 672,000 to 186,000.

Did Kennedy or his advisers really believe that there were welcoming communities to which patients could return? If so, it would be nice to know what warm and caring havens they had in mind. In the name of “deinstitutionalization,” thousands of men and women were dumped on city streets (“from the back wards to the back alleys” is Rousseau’s phrase) at which point those who had freed them lost interest in mental health and turned to other projects. Of course, the back wards weren’t any better. The point is that even now we do not know what kind of services, if any, these wandering women need if they are to have a humane life. A reasonably decent room to go to comes to mind for a start. Rousseau’s book calls attention to them, as the saying goes. Yet I feel constrained to add that its glossy photographs and handsome format make it another instance where the medium overwhelms the message.

Shopping-bag ladies belong to the underclass: if anyone has fallen out of society, they certainly have. Moreover they make their presence felt simply by roaming in our midst, making the rest of us wish they would simply go away. But I’m not so sure they should. Better than slum criminals or welfare mothers, they serve as a warning that even people like ourselves can fall to the bottom and no one will care. The underclass has openings for all Americans. In that sense, if no other, it is eminently democratic.

This Issue

August 12, 1982