interest in physical reproduction. The channels so created are classed here as "alternatives to reproduction" in the sense that, although none are free from ambiguity in their effects, each appears to result in an indirect reduction of reproduction. Higher Education

That higher education correlates with a decline in fertility among those educated has been an established pattern in the United States, as it has been throughout the world. The American college graduate has tended to marry later and to procreate later; to form more childless families; to form fewer families of four or more children; and to have fewer average children per family. In 1969, for example, the following comparison of couples with the husband under 45 may be made:

The differences are less marked between high school graduates and college graduates, although they still exist. They are most extreme if grade school graduates are compared with a very highly educated group, such as women who have received doctoral degrees. For example, a study of women who received doctoral degrees in 1958 and 1957 showed that, of those in the 40-44 age group in 1965-1966, only 55 percent were married or had been married.222 Of those who had married, one-third had married late by American standards, after they had finished college. Most striking, perhaps, of all, for those who were married, the average number of children was two; 49 percent of these women in their forties had children still pre-school age; and 28 percent of the married women were childless.223 In the case of the woman Ph.D., the postponement of marriage and childbearing, the tendency to childlessness, and the lower number of children in fertile marriages-all tendencies present among college graduates-have been carried very far.

A phenomenon similar to that of the woman Ph.D. has been noticeable among college-educated black per

Unintended Consequences: Laws Indirectly Affecting

sons. In the first comprehensive study of this group in 1936, it was observed that there were striking contrasts between black college graduates and black Americans in general. In the age group 20 to 24, for example, 90 percent of the college graduates were single, compared with 43 percent of blacks reported as single in the 1930 census; of black women college graduates 35 to 39, 30 percent were single in comparison with seven percent of all black women in the same age group who were studied.224 Of the black college graduates who were married, 42 percent were childless; and the average number of children per married family was 1.4. At this time, the black college graduate was an exceptional person-there were under 25,000 graduates since 1914.225 But the phenomenon has continued, though to a less marked degree.2

226

For an extended period, the one exception to the correlation of higher education and lower fertility was the higher fertility of Catholic college graduates in the United States. This anomaly tends to disappear, however, among Catholic college graduates after 1964.2 .227 The new trend among a group which in every other respect-ethnic, economic, social-remained constant, is itself dramatic evidence of the power of education. In the 1960's, Catholic teaching on contraception began to shift from clear and unanimous condemnation to inquiry, doubt, and hesitation among the theologians.228 In a comparatively brief time, a new view of an age-old doctrine emerged; the official position of the papacy, belatedly expressed, was not enough to reverse it. The Catholic college graduates, the most indoctrinated and most interested group, shifted, as theological opinion itself shifted, towards acceptance.

It might be asked, however, if education will continue to play the role it has appeared to play in affecting fertility. By 1965, there had been a marked shift in favor of contraceptive practice by grade school graduates, so that 89 percent of them in the 18 to 24 age group expected to practice contraception.229 If this attitude were translated into consistent action, it would seem likely to bring the family size of grade school graduate down closer to the college graduate level. It could also be hypothesized that when women Ph.D.'s become less of a rarity and when black college graduates become more numerous, the marriage and family patterns of these groups might become more like the American norm. It could be argued that the high rates of celibacy and childlessness and low fertility have been associated with the special characteristics and special effort needed for a discriminated-against group to make an unusual advance, and should not be correlated with the effect of higher education as such.

Plausible though such predictions are, the present correlation of higher education and lower fertility is strong. The foundation for that correlation exists in the value changes effected by advanced education. Forming a perspective in which planning is done on a long time scale, opening up opportunities for creative work besides childbearing and child raising, ordinarily leading to a high estimation of the education of offspring, higher education affects the values of those who participate in it. Family size appears to be related to the valuations engendered. Assuming that the relationship between higher education and lessened procreation suggests a causal nexus, the law's contribution to education may be a significant contribution to the reduction of population growth.

The rise in Federal support to higher education is familiar. Through Title IV of the Housing Act of 1950, the National Defense Education Act of 1958, the Higher Education Facilities Act of 1963, the Higher Education Act of 1965, and a variety of other programs, the Federal government has become a major partner in what was the preserve of private philanthropy and state and local government.4 230 In 1920, 12 million dollars was contributed to the current income of the colleges from Federal sources; in 1950, over one-half billion dollars had a Federal origin; in 1964, over 3.5 billion dollars.2 231 Between 1963 and 1967, Federal obligational authority for college programs, including both construction and student loan funds, increased from 1.8 billion to 3.5 billion dollars.232 The Internal Revenue Code was amended in 1954 to exempt not only college scholarships from taxation but also graduate fellowships involving research; the annual cost of the scholarshipfellowship exclusion is about 50 million dollars.233 In 1971, over half a million students received over 364 million dollars in NDEA loans. Figures of this magnitude underline the Federal involvement.

Aid to the Poor. S. 659, passed by 51 to 0 by the Senate on August 6, 1971, aimed to increase the Federal commitment to education. Two new programs under the bill were of particular interest: "Basic Educational Opportunity Grants" to which any student accepted at an institution of higher learning would be "entitled" and Cost-of-Institution Allowances to the institutions matching the student grants.234 The new programs would have involved Federal expenditures of over five billion dollars in three years.235 They were designed to give the greatest aid to students from low-income families and to the institutions enrolling such students. They were a direct response to the fact that today a student from a family with an income of $15,000 is almost five times more likely to attend college than a student from a family with a $3,000 income.236 The proposed Act

sought to enlarge the possibilities of education for the poor.

The grant schedule of the Act was geared not only to family income, but to family size. The more dependents there were in the family, the more the student would receive from the government.237 If these subsidies were to be analyzed as benefits to the parents, they would appear to be benefits to reproduction. But they were better analyzed as benefits to the students themselves. The extra grants to children of large, poor families would have made it possible for substantial numbers to be initiated into higher education. Sidney P. Marland, Jr., Commissioner of Higher Education, testifying in favor of such an approach, emphasized what the availability of college money would do for the hopes of the poor family.238 Even more important was what it could have done for the values of the students benefited.

The Senate bill met with opposition in the House. Partisan opposition and objection to the favor shown the poor led to a different measure being passed by the House; the population aspect of the bill was not emphasized; and S. 659 is in conference at this writing. Its provisions stand as an example of what could be done if the Federal government were to concentrate on educating at a college level those whose reproductive behavior might be most powerfully affected by higher education. Aid to the low-income large family would be of particular importance to black children. In 1969, only 343,000 blacks between the ages of 18-21 were enrolled in college.239 A substantial increase in economic aid would assist in achieving a result that the dismantling of legal barriers to educational opportunity has not yet accomplished.

Aid to Women. The parallelism insisted upon by some advocates of women's rights between the position of blacks and the position of women does not extend to the likelihood that women will be more apt to come from large poor families.240 In four specific ways, however, there have been parallels in educational experience, and the barriers based on sex have been questioned only after invalidation of those based on race. The parallels are these: (1) There is discrimination in admission of women to college and graduate school. A recent study of 240 colleges showed that, at the lower end of the admission group, women were discriminated against.241 As to graduate schools, some law schools, such as Harvard's, did not admit women at all before the 1950's. Medical schools and architecture schools still use

quotas.242 (2) Women have been denied admission to the most prestigious colleges in their area, as was the case, for example, in Virginia until, in 1970, litigation. led the University of Virginia at Charlottesville to open its doors to women.243 (3) Girls have often been put on

educational tracks that led to lower paying forms of work which stereotype suggested were appropriate. This kind of teaching in high school curricula has been continued in counseling at college.244 (4) De facto, women have often been segregated in all-girl colleges, where the feminine stereotype might be expected to be reinforcing.245

It could be argued that total educational equality for women, including, for example, the abolition of separate dormitories, might result in an increase of sexual activity and that single-sex colleges actually serve a population-related purpose by reducing the propinquity of sexes which is the necessary basis for marriage.246 Against this argument must stand the loss of educational opportunity which segregation by sex imposes. Most probably, if higher education is to be a path to lower population, it will be by better coeducation.2 247

Rewarding and Responsible Employment

Although correlations have been noted between the employment of women and reduced childbearing, what seems particularly significant in presenting an alternative to childbearing is employment which can satisfy psychic needs for a sense of status, function, and purpose.248 Childbearing tends to be overwhelmingly important for women whose lives afford no other substantial purposes. While employment of any kind may be an alternative to a life centered on continued childbearing, and while employment of any kind may be invested with dignity, employment will be a more powerful alternative if its character, compensation, and public image are high rather than low.

The substantial gap that still exists between the wages of men and women indicates that the employment opportunities of the sexes are not equal. As of 1970, women working full-time received a median wage only 60 percent that of men working full-time.249 Black women have been at the bottom of the wage scale earning, in 1970, less than half of the median wage of white men.250 Women seeking management positions in industry have had little success.251 At the other end of the scale, domestic household help has continued to be among the poorest paid of jobs. This work traditionally has been performed by women and, to a substantial extent in the United States, by black women. The median annual income of this group in 1967 was $1,398.252 At the higher end of the income scale, women accounted for only 10 percent of holders of doctorates and three percent of the lawyers in the United States.253 In light of these patterns, laws which affect the employment of women are significant, and

Unintended Consequences: Laws Indirectly Affecting

laws affecting the employment of black women are doubly significant.

Differential legal treatment for women was first accepted by the Supreme Court in a spirit of benevolent paternalism. From the Muller case in 1908, sustaining the constitutionality of Oregon's maximum hour laws for women, to the Breedlove case in 1937, finding no unfairness in the exemption of women from the Georgia poll tax, the Court exhibited the chivalrous side of male chauvinism.254 The two sides of this spirit are captured in the observations of Justice Butler in Breedlove:

The tax being upon persons, women may be exempted on the basis of special considerations to which they are naturally entitled. In view of burdens necessarily borne by them for the preservation of the race, the State may reasonably exempt them from poll taxes... The laws of Georgia declare the husband to be the head of the family and the wife to be subject to him. Sec. 53-501. To subject her to the levy would be to add to his burden.255 Naturally entitled to special consideration, but legally subject, women were protected from themselves. In 1948, the Supreme Court upheld a Michigan statute denying women the opportunity to act as bartenders, 256

A new Federal history began with a Presidential order in 1963 against discrimination in Federal employment.257 In the same year, the Fair Labor Standards Act of 1938 was amended by the Equal Pay Act requiring employers to pay equal wages to both sexes. Discriminatory seniority systems, however, were left untouched by this amendment. Title VII of the Civil Rights Act, enacted in 1964, made unlawful any action by an employer or by a labor union to discriminate against any individual because of his "race, color, religion, sex, or national origin."258 The Equal Employment Opportunities Commission (EEOC) was created by the same law with power to hear complaints, seek informally to eliminate discrimination, and recommend institution of civil action by the Attorney General to enjoin discrimination.259 The powers of the EEOC are not enormous-they are, for example, a good deal less than those of the National Labor Relations Board-but the Commission has provided Federal standards for identifying discrimination.

These standards forbid refusal to hire a woman because of assumptions as to the employment characteristics of women, such as an assumption that the turnover among women is higher than among men. They forbid discrimination based on stereotyped characterization of the sexes, such as that women are less capable of

aggressive salesmanship. They forbid using the preferences of coworkers or clients as justification for discrimination. They reject the need to provide separate toilet facilities as an excuse "unless the expense would be clearly unreasonable."260 In the airline industry, discrimination against women on the basis of marital status and age was condemned by the EEOC. The condemnation was not effective in itself; but following the EEOC action, collective bargaining ended the airline's traditional requirements for stewardesses to be young and single.261 This indirect effect of the Civil Rights Act suggests the force of legal standards even without legal

sanctions.

In 1971, the Supreme Court had its first occasion to interpret the statute.262 Ida Phillips was denied a job by Martin Marietta Corporation as an "assembly trainee" because she had children of preschool age. Men with children of preschool age were being hired. The EEOC failed to conciliate Mrs. Philipps' grievance and found reasonable cause to believe Martin Marietta had discriminated on the basis of sex. Mrs. Phillipps sued the company on behalf of herself and all women similarly situated. The District Court summarily dismissed the suit; the Court of Appeals affirmed the dismissal; but the Supreme Court ordered a trial on the issue whether the requirement of no preschool-age children for women was "a bona fide occupational qualification reasonably necessary to the normal operation" of the business. If so, under the statute, the discrimination was permissable; otherwise it was unlawful. Justice Marshall, in a special concurrence, doubted that it was possible for the company to show that its requirement was related to the job. In opening up this possibility, the Court, he feared, had "fallen into the trap of assuming that the Act permits ancient canards about the proper role of women to be a basis for discrimination." Congress, he added, had "sought just the opposite result." The only exception he would permit to equal treatment was one where a job required "physical characteristics necessarily possessed by only one sex."263 The equal employment rights of 14 million women with children under six were at stake.264 The success of the statute in deterring discrimination based on sex may depend on whether this approach ultimately prevails.

While the Federal government has moved generally against sex discrimination, it has retained discriminatory practices in the Armed Forces. By regulation of the Secretary of the Army, for example, only two percent of volunteers may be women.? 265 Women, of course, are exempted from the draft; and in 1968, a Federal district court sustained this exemption, upon challenge by a male, on the basis of a domestic stereotype of women, who, the court observed, are "to keep the home fires

burning."266 Prior to 1967, women could not be appointed flag or general officers in the Armed Forces; there were restrictions on the number of percentage of women officers at lower levels; and there were earlier mandatory retirement provisions for women officers.267 These inequalities were eliminated by statute.268 Nonetheless, there were only 2,300 line officers and 10,000 medical officers who were women at this time." 269 West Point, Annapolis, the Air Force Academy, and 12 other public military schools remained closed to women.270 The likelihood of women becoming part of the military elite was not very much enlarged.

The levity which punctuated argument of the Philipps case in the Supreme Court this year has already been noted as a sign that the Supreme Court was not likely to take sex discrimination as seriously as race discrimination.2 271 Twenty years earlier in the woman bartender case, Justice Frankfurter had characterized the topic as "beguiling" before adding severely "it need not detain us long.”272 These masculine responses are no doubt to be read as defenses which, until men's attitudes change, will affect the law on equal opportunity for

women.

Celibate Service

Apart from the voluntary renunciation of marriage by members of Catholic religious orders and by Mormon missionaries during their period of mission, Federal law has created organizations whose form has discouraged marriage for individuals enrolled in them. The Peace Corps, as it existed in its first decade, is the outstanding example of a voluntary association of this kind, formed by law inviting volunteers to serve "under conditions of hardship."273 The law as written made no provision for living allowances, housing, transportation, clothing, or subsistence of either spouses or children of the volunteers. Only the spouses and children of leaders were entitled to living and other allowances, and then only by Presidential grace in exceptional circumstances.274 Volunteers occasionally married each other. But in general, the result was a corps of young persons who, as part of their dedication to service under conditions of hardship, postponed marriage and procreation.

In 1970, the law was amended to suggest different expectations by providing for living, travel, education, and leave allowances and health care for minor children of the volunteers.275 The Peace Corps has never been numerically large enough to have any demographic effects; but, as a Federal model for channeling the young into activity alternative to reproduction, it is not without exemplary interest. The Job Corps, whose allowance and support provisions were also geared to the

single person, is a second, less impressive example of the displacement of marital activity by Federal service.276

The Selective Service law could be viewed as a compulsory system of discouraging marriage for those actually inducted into the armed forces. Reduction of the normal opportunities to meet likely mates and the reduction of opportunity to live with one's spouse if married could be seen as pressures to defer marriage. On the other hand, the pay and allowance provisions for servicemen do not assume that they will be celibates.277 Moreover, the granting of draft deferments to persons who have conceived children actually affords an incentive for early procreation.278 Another statutory ground for deferment, the status of sole surviving son, reveals a special Congressional tenderness for perpetuation of the family name by procreation.279

In sum, only the faintest adumbrations of a Federal system of celibate service, in which procreation is postponed, exist.

Adoption

Not only an alternative but a substitute for reproduction is adoption-an approach as old as Thomas More's Utopia and a reality in the United States since World War II.280 Between 1952 and 1966, over 1.5 million children have been adopted, at least 50 percent of them by nonrelatives.281 Probably a majority of the adopters were fertile; for an undertermined percentage of this goup, adoption deferred or supplanted reproduction.

The children least likely to be adopted have been black children.282 It has been sometimes supposed that black adults have been ungenerous in their response to this situation. The chief variable affecting willingness to adopt, however, appears to be economic. If economic differences are controlled, blacks adopt at a higher rate than whites.283

This fact suggests the importance of experiments such as that begun by California in 1969, whereby the state actively promotes the adoption of children, who "because of their ethnic background, race, color, language, physical or mental, or emotional or medical handicaps, or age have become difficult to place in adoptive homes."284 To encourage their placement, the state provides information as to the children's availability and gives legal assistance in completing the formalities of adoption. Above all, it provides, for three years, the kind of financial assistance which would be paid to foster parents. A realistic step has been taken to making adoption practicable for those the statute describes as "families of lower income levels" and "those belonging to disadvantaged groups. "285

Unintended Consequences: Laws Indirectly Affecting

Adoption of unadopted children might also be assisted by other legal changes. Barriers to interracial adoption in a few states statutory, in more states informal judicial policy-could be eliminated by judicial interpretation of the Constitution. The use of race to deprive anyone of a form of parenthood could be found impermissible under the Fourteenth Amendment.287 That age, marital status, or religion of the prospective adopters should be treated as absolute disqualifications could also be questioned on constitutional grounds if the courts would recognize that parenthood by adoption was entitled to protection not inferior to that accorded biological parenthood.288 Such a judicial development, giving the fullest recognition to parenthood by choice and parenthood severed from procreation, would have a symbolic significance beyond the adoptions actually affected.

MAKING THE UNINTENDED INTENDED

Suppose, in order to lower the birth rate, a system in which the poor with large families-say, every couple with an income of less than $3,500 and with more than three children-were subjected to a combination of taxes which, in part, increased in direct proportion to the number of children they had to feed, and which, in total effect, left the couple a smaller percentage of income disposable for their own needs than a couple with fewer children. To heighten the lesson, suppose the homes of such large families were randomly selected for systematic destruction by public authority, and that the government made it more difficult for such families to find new housing commensurate with their size. To accentuate the contrast, suppose that Federal grants were given to childless couples who purchased homes on credit and to builders who would construct housing designed for couples with no more than two children. This supposed system would differ from the present predominant pattern in only two respects-the absence of the ameliorative effects of health and welfare measures, and the presence of a new symbolic element. That symbolic element would consist of the governmentally determined message that each deprivation inflicted on the large family and each benefit bestowed on the small family was intended to deter population growth. Undoubtedly, that symbolic value attached to the benefits and burdens would make the new law a far