The “Cliometricians,” as they call themselves, have married the muse of history to the science of mathematics. These statistical historians are extremely sensitive and defensive about the union for they are aware that traditional devotees of Clio, most historians, regard the marriage as a mésalliance, a forced union of incompatibles. Confronted with equations they cannot read, with techniques they cannot understand, with copious data beyond their comprehension, traditional historians have reacted defensively and belligerently themselves. They see their authority challenged, their humanistic values threatened, their canons of criticism ridiculed, and their cherished classics derided as “soft,” impressionistic, and unscientific. It is not surprising that some of them have overreacted.

Cliometricians Robert Fogel and Stanley Engerman (especially the former) are by now scarred veterans of this guerrilla war in the academy. In this all-out assault on American slavery historians, they have adopted a strategy somewhat more propitiatory than is customary among militant quantifiers. They admit that “history cannot be reduced merely to a science,” that there are no completely “errorless data,” that “not all questions have unambiguous answers,” that even they are at times “forced into speculation,” have not “expunged all ideological influences from this book,” and are “combining the methods of science with the concerns of humanism.” The reader is titillated with the warning that “this will be a disturbing book to read,” that its “revolutionary” findings were “initially discounted, even rejected out of hand” by the authors themselves. They plead for “forbearance on the part of the reader” and promise that “this forbearance will prove worthwhile.”

As a further concession, they present their findings in the first volume “in as nontechnical a manner as possible” in perfectly intelligible if undistinguished prose for the general reader. The second is given over to documentation, defense of method, algebraic equations, tables and graphs, and computer language beyond the comprehension of laymen. A third of this second volume is devoted to a ruthless scourging of traditional historians of slavery, whether of the left or the right, for their slipshod and unscientific ways, and to the merciless clobbering of a few unfortunate victims. Even in the “nontechnical” volume of findings the rattle of electronic equipment is heard off stage, and the reader is coerced by references to “vast research effort involving thousands of man and computer hours” and inconceivable mountains of statistical data. Propitiation and concession do not diminish in the slightest the deadly seriousness of the attack.

The object of the attack is the entire “traditional” interpretation of the slave economy, and the authors define this in “five main propositions” as follows: “1. that slavery was generally an unprofitable investment, or depended on trade in slaves to be profitable, except on new, highly fertile land; 2. that slavery was economically moribund; 3. that slave labor and agricultural production based on slave labor were economically inefficient; 4. that slavery caused the economy of the South to stagnate, or at least retarded its growth, during the antebellum era; and 5. that slavery provided extremely harsh material conditions of life for the typical slave.”

Now no historian holds all these propositions simultaneously, and a few of them have already been discredited by findings of traditional as well as econometric historians. It is therefore difficult to accept them all as “the” traditional interpretations. Some defenders can still be found for each of them, and all of them can be said to have been “traditional” at one time or another. None of the propositions, however, has come under such strong attack as that leveled by our Cliometricians. The last of the propositions, the one about extremely harsh conditions of life, crackles with moral electricity and has had the powerful support of recent neoabolitionist historians, some of whom hold that American slavery in this respect was the worst of all the systems of the New World. We shall turn to that proposition first.

New questions about the comparative plight of the South’s slaves were raised by Philip D. Curtin’s study of the demography of the slave trade, and Fogel and Engerman develop his findings to show that the African slaves flourished biologically in the United States as nowhere else. While slaves in Latin America could not sustain their population by natural increase and had high rates of annual decrease, the South’s slaves multiplied at unprecedented rates. Had the South duplicated the demographic experience of the West Indies, it would have had only 186,000 slaves by 1800 instead of over a million, and in the next sixty years American slaves quadrupled in number. By the crudest test, that of survival and reproduction, the American slaves had no equal or rival anywhere. The authors are careful not to attribute their experience to good treatment but of course it opens a question. On that subject their conclusions mainly concern the last three decades of slavery, and in all these matters it is well to keep in mind that they speak in statistical aggregates and averages—not of individual experiences.


According to the computers of the Cliometricians American slaves were remarkably well fed. Their average daily diet was “quite substantial,” and its energy value “exceeded that of free men in 1879 by more than ten percent.” Far from inadequate, “it actually exceeded modern (1964) recommended daily levels of the chief nutrients”—of protein by 110 percent. Data on slave housing is admittedly more sparse. To say that it “compared well with the housing of free workers in the [late] antebellum era” and that the “typical” cabin “probably contained more sleeping space per person than was available to most of New York City’s workers half a century later” means little without the variable of quality.

Evidence on health is more solid, though the contribution of medicine, however lavishly used, is doubtful. Limited samples indicate remarkably few days lost from illness of workers. Planter concern for the welfare of pregnant slave women paid off with a lower mortality rate in child bearing than that of southern white women. The infant death rate of slaves was “virtually the same” as that of southern whites. No doubt the ghosts of many hungry, cold, and ailing slaves protest the contrary, but Cliometricians use averages. Life expectancy at birth for slaves was four years shorter than for American whites as a whole, but it was the same as that for the Dutch and the French, higher than for Italians and Austrians, and “slaves had much longer life expectations than free urban industrial workers in both the United States and Europe.”

Whipping fell from favor as an acceptable punishment by the nineteenth century in the free states, but “persisted in the South because the cost of substituting hunger and incarceration for the lash was greater for the slaveowner than for the northern employer of free labor.” The slender basis used for the quantification of whippings is unimpressive, but the authors found brutal and sadistic masters relatively rare. Whipping was an integral part of the system of discipline and motivation, but the work attitude desired by the master “could not be beaten into slaves. It had to be elicited.” For that “they developed a wide-ranging system of rewards,” rewards for short-run performance such as prizes in goods or cash or unscheduled holidays; for the longer run, year-end bonuses, and longer still, promotions in economic and social status—from field hand to artisan, from that to driver and on to overseer, the right to hire out and the right to self-purchase—all with tangible payoff. Slaves could double and triple their “basic income” of upkeep and frequently did. The uniform level of income for slaves is pronounced a myth, along with exclusive reliance on the lash for incentive to work.

Another demonstrable myth was that all slaves were menial laborers. The fact was that about 7 percent of the men held managerial posts, about 12 percent were skilled craftsmen or artisans, and another 7.4 percent held domestic posts such as coachmen or house servants. Slave society thus “produced a complex social hierarchy which was closely related to the occupational pyramid,” with one out of every five adult slaves holding “preferred occupational positions.” One of the most significant discoveries of Fogel and Engerman is that among moderate-sized plantations (sixteen to fifty slaves) fewer than one out of six employed a white overseer, and even on the larger ones (more than fifty slaves) only one out of four. The consequence was that much of the responsible plantation management and agricultural supervision was in the hands of black slave overseers, drivers, and gang foremen. Black overseers were regularly in charge of the whole production side of plantation operation, including entrepreneurial decisions, field operations, and supply purchasing.

How can this picture of economic order and social stability possibly be reconciled with the slave market and all its horrors—broken marriages, divided families, child sales, systematic slave breeding—“like oxen for the shambles”—the “stud farms,” the “black harems” kept by dissolute masters and their mixed-blood brood, the famous “submissiveness and promiscuity” of slave women? The answer of this study is that clear economic advantage combined with Victorian morality to promote a strong, stable, nuclear slave family. It was not a “matriarchal” family with an emasculated, transient father. “For better or worse, the dominant role in slave society was played by men, not women.” Planters recognized husbands as head of the family and the family as an institution of “central importance” on the plantation. It was the housing unit, the administrative unit, an important instrument for labor discipline, and the main instrument for procreation and child-rearing. Elaborate systems of rewards, sanctions, codes, and ceremony were employed to promote family stability, and the system by and large was a success.


Infractions there were, as in all codes, and slavery encouraged some of them. Many masters did take slave mistresses and concubines, and promiscuity did occur in the quarters. These investigators, however, find no evidence “which sustains the charge that promiscuity among slaves was greater than that found among whites.” The pattern of child spacing among slave mothers between the ages of eighteen and thirty was “somewhat over two years,” which would be normal for breast-feeding mothers not using contraception. More important, although motherhood in the early teens was not unknown, the average age of a woman at the birth of her first child was 22.5. The conclusions are that “the great majority of slave children were borne by women who were not only quite mature, but who were already married,” that “only abstinence would explain the relative shortage of births in the late-teen ages,” that “prevailing sexual mores of slaves were not promiscuous but prudish,” and that “slave life pivoted around stable, nuclear families.”

The quantification of sexual exploitation and miscegenation presents more problems. At least it appeared that in 1860 mulattoes were more prevalent among free Negroes and in cities, where they were about 25 percent of the Negroes. But among rural slaves, who made up 95 percent of the slave population, only 9.9 percent were mulatto. The trouble with calculating white paternity is that unions between mulattoes and blacks also produced mulattoes. By a genetic equation beyond my grasp it is calculated “that the share of Negro children fathered by whites on slave plantations probably averaged between 1 and 2 percent.”

By far the most damaging moral indictment of slavery used by the abolitionists was the charge of systematic breeding of slaves for the market, the “stud” farms that evoked the barnyard metaphor equating human beings with livestock and their calculated manipulation to maximize fertility for profit. It was the ultimate shame of slave society. That the practice existed is still an article of faith among some historians, and to complicate matters among some Cliometricians as well. Serious economic historians have used the “slave breeding” theory to explain how planters of the “soil-depleted” slave-exporting states kept in business by profits on slave sales.

In spite of all this, Fogel and Engerman dismiss the theory as a “myth.” They point out that not a single authentic case of a “stud” farm has been turned up, that the total net income of the slave-exporting states from slave sales “accounted for less than one percent of farm receipts” in 1860, that even if interference with normal sex life of slaves could have increased fertility to the biological maximum, the increase in net income per working slave would have been “less than a dollar a year per slave.” And this trivial gain could have been more than offset by loss in worker morale, by runaways and slowdowns. Besides, the crude equation with livestock fails to take into account human emotional factors in conception, factors that also produced distraught and disgruntled field hands. Of course planters did seek to influence fertility patterns, but they did this through economic and honorific incentives akin to those used by various governments today.

It remains to fit the slave trade into this revision. The great movement of slave population was between the upper and seaboard states and the Gulf states; the exporting and importing states, or as the abolitionists had it, the “breeding states” and the “buying states.” Between 1790 and 1860 a total of 835,000 slaves made this western move. The Cliometricians find that 84 percent of the slaves in the westward movement migrated with their owners and only 16 percent were moved by slavetraders for the market, or in the old phrase “sold down the river.”

What was the cost to slave family integrity in broken marriages and children torn from parents? Using the vast slave records of New Orleans between 1804 and 1862 (never previously explored by historians), Fogel and Engerman find that 84 percent of all sales of slaves over fourteen involved unmarried persons. Allowing for the percentage sold with their mates and those voluntarily separated or widowed, they conclude that “it is probable that less than 2 percent of the marriages involved in the westward trek were destroyed in the process of migration”—perhaps (they speculate here) not “significantly greater among blacks than it was among whites” caught up in the westward movement.

As for the legend that children under thirteen were “hardly less than a staple in the trade,” it seems that only 9.3 percent of the New Orleans sales were children of this age. Since 15.9 percent of slave children under thirteen were orphans because of the high death rate of that time, the full extent of the interregional slave trade in children could be accounted by “just one out of every 810 orphans.”

Of course the slaves “sold down the river” were only a small part of the total domestic slave trade. Projecting figures on slave trading in Maryland (local and interstate) between 1830 and 1840 to the national level, the authors calculate that total slave sales in the country averaged about 50,000 a year over the period 1820 to 1860. This meant that “on the average only one slaveholder out of twenty-two sold a slave in any given year, and that roughly one third were estates of deceased persons.” Typically, slaves were purchased for use rather than for speculation, as indicated by very low sales rates of ongoing plantations. A sample of nineteen of the latter with a total slave population of 3,900 reveals that over a period of ninety years only seven slaves were sold.

Turning from the effect of the slave economy on the slaves to its effects on the slaveholders, the nonslaveholders, and the South as a whole, the Cliometricians martial their statistical forces to answer a host of old questions. Did planters, as many historians have held, find slavery essentially unprofitable and maintain it from irrational, aristocratic, noneconomic motives? Was slavery moribund on the eve of the Civil War, doomed already to extinction, and were planters pessimistic about their future? Were their agriculture and labor force inefficient and unproductive compared with free agriculture? Was slavery incompatible with industry and urban life? Were nonslaveholders impoverished and despondent? Was the South’s economy stagnating and its growth arrested?

The response to each of these questions from Fogel and Engerman is a ringing negative. The planters were doing very well indeed. Enjoying average earnings of about 10 percent on the market price of their bondsmen, they received “rates of return equal to, or in excess of, the averages which obtained in a variety of nonagricultural enterprises,” New England textiles and southern railroads, for example. Patriarchs and paternalists they were, but “paternalism is not intrinsically antagonistic to capitalist enterprise.” Far from being “precapitalist” or “uncommercial,” they exhibited “as much shrewdness as could be expected of any northern capitalist.”

The theory that irrational and uncommercial planters were seized by an irresistible impulse to an overproduction of cotton that doubled output in the 1850s and threatened to sink the cotton kingdom is dismissed as nonsense. It was a fabulous boom in profits for planters brought on by expanding world demand for cotton. They should have produced even more cotton. Two more famous theories hold that slavery was doomed to self-destruction within a decade or two without a Civil War. One is the theory that slavery had already reached its “natural limits” of geographical expansion, and the other that wherever slavery “touched urban conditions it was in deep trouble.”

Both theories fail the test of statistical analysis. Actually, land devoted to cotton nearly doubled between 1860 and 1890, and more than doubled again by 1925. The world demand for cotton grew more rapidly than the supply. Had there been no Civil War the price of slaves would have risen an estimated 52 percent by 1890. Slavery was not about to “self-destruct.” Nor was slavery incompatible with cities. Slave population in cities declined during periods of agricultural prosperity because cities were unable to compete with planters for their services. Whatever political rhetoric they used to the contrary, the behavior of slaveholders proved that they were no pessimists, for they “not only expected their social order to endure but foresaw an era of prosperity.”

Fogel and Engerman place heavy blame on the abolitionists, especially Cassius Marcellus Clay and Hinton Rowan Helper (both Southerners), and particularly Frederick Law Olmsted, for misleading subsequent historians about the character and strength of the slave economy and the culture and capability of the slaves. Because they hated slavery so much, they sought to prove it not only bad morally but disastrous economically, not only bad for blacks but whites as well, nonslaveholders as well as slaveholders. They contended that it was inefficient and wasteful, inferior in all respects to the North’s free labor system, that it impoverished the whole region, degraded all labor, misallocated investment, inhibited industrialization, thwarted industrialization, and stifled progress generally. It had turned nonslaveholding whites, 70 percent of the white population, into a “promiscuous horde” who were “little removed from savage life, eking out a wretched subsistence by hunting, by fishing, by…occasional jobs, by plunder.”

The keystone of the economic indictment, however, was the inefficiency of slave labor. And because of their unconscious racism the abolitionists fixed upon the African slaves their enduring stereotype of laziness, incompetence, stupidity, and sloth, of being habitual shirkers, malingerers, pilferers, and liars. Negro domestics, said Olmsted, could not do half the work of “the commonest, stupidest Irish domestic drudges at the North.” Those modern historians and writers who shared a hatred of slavery, and even some like Ulrich B. Phillips who did not, perpetuated the abolitionist views of the slave economy and the slave.

With all these indictments of the economy and the slave our Cliometricians take strong issue. By use of elaborate data, techniques, and equations, they arrive at the conclusion that “southern agriculture as a whole was about 35 percent more efficient than northern agriculture in 1860.” That is, with a given amount of labor, land, and capital, a southern farm could produce 35 percent more than a northern farm. Both southern free and slave farms were more efficient than northern farms, but while the South’s free farms were only 9 percent more efficient, slave farms were 40 percent more efficient. Efficiency was correlated with scale, and all free farms were small while all large ones used slave labor. There were variations in area and land, but slave plantations in the newer southern states outstripped northern free farm efficiency by 53 percent. There were no clear signs of land depletion or declining yields in the older states. In fact, the average value of land in Virginia and the Carolinas increased 60 percent in the 1850s.

These triumphs of agricultural efficiency are attributed to planter ingenuity and the excellent quality of slave labor. No cavalier fops or lazy triflers, the leading planters were alert to every advantage, atune to scientific advance, and alive to every detail and to nothing more than details of labor management. Their main secret was superb labor management. Specialized and interdependent, each with an assigned set of tasks throughout the year, drivers, foremen, plowmen, hoe hands and so on through nurses and cooks “were as rigidly organized as in a factory.” The crucial element was the gang system—plow gangs, harrow gangs, hoe gangs—which kept each other under pressure and maintained a steady, intense rhythm of work. The plantation regimen was “more like a modern assembly line than…the routine in many of the factories of the antebellum era.” In fact, the authors claim that “the great plantations were the first large, scientifically managed business enterprises” and the black slaves were “the first group of workers to be trained in the work rhythms which later became characteristic of industrial society.” Olmsted was the “preindustrial man,” not the big planter.

It could not have been done without slave labor. Free labor would not submit to the gang regimen, nor would blacks after freedom. But it could not have been done either by the type of stupid, malingering, slothful, incompetent slaves described by the abolitionists and neoabolitionists. On most large plantations the top nonowner management was black, and under its black drivers and foremen secured from fellow slaves the most efficient and productive labor performance in American agriculture and anticipated some aspects of labor discipline in future industrial America.

The results measured in per capita income and rate of economic growth for the South as a whole are striking refutation of the abolitionists’ economic indictment of slavery. The per capita income of the Northeast was extraordinary and exceeded that of all other regions, but that of the South was also very high and exceeded the per capita income of the other free states, the present Middle West, by 14 percent. In fact, the South was richer in 1860 than any country of Europe save England, and Italy did not attain the southern per capita income of 1860 until the eve of the Second World War. Instead of stagnating, per capita income was growing 30 percent more rapidly in the South than in the North between 1840 and 1860. In that period the South’s rate of growth was 1.7 percent a year, a rate that only Sweden and Japan have been able to exceed substantially over a long term.

Granting the amount of wealth, what of inequalities of distribution, for example the belief that 70 percent of southern whites lived in poverty or on the border of hunger? While wealth was more unequally distributed among farmers of the South than those of the North, it was less unequal than wealth distribution in urban areas. Since much the larger urban population was in the North, the distribution of wealth, so we are told, “was roughly the same above and below the Mason-Dixon line for free populations.” Yes, but that leaves the unfree to account for.

Here our Cliometricians enter a line of reasoning and a series of sophisticated computations that baffle the reviewer’s comprehension and leave him more skeptical than does any other finding (and the promptings of skepticism are not rare). But to report their conclusions, they find that on the average only “12 percent of the value of income produced by slaves was expropriated by their masters,” and that this falls well within modern rates of expropriation. They go further to say that in “the course of his lifetime the typical slave field hand received about 90 percent of the income he produced.” That is built on the calculation that the slaveholder did not break even on his investment until the slave was twenty-seven, with a life expectancy of thirty-six.

Barrington Moore, Jr., once remarked that the reason leftist historians shun quantified history is that it usually proves that things were not so bad before the revolution. So it has proved in this case. Of course Messrs. Fogel and Engerman are perfectly aware that they will be charged with concocting the best defense of the Jolly Old Institution since the days of John C. Calhoun and George Fitzhugh. In fact, many of their findings would serve as documentation of the proslavery argument. They report preliminary reactions: Are “we merely trying to shock fuzzy liberals?” And again, “What are you guys trying to do? Sell slavery?”

They deny this heatedly. But they admit they are distressed over the apparent clash between their findings and their personal values. This is reflected in the emotional and moralistic rhetoric of their concluding chapter and in the incongruously neoabolitionist title they have chosen for their book. They protest their abhorrence of slavery and its moral underpinnings, their awareness of its (unquantifiable) psychological and cultural deprivations, their appreciation of the contrast between concrete existential experience and abstract statistical aggregates. Their protests deserve charitable credence, and their critics should observe traditional civilities.

One of the authors’ justifications for defying received opinion deserves special consideration. This is their desire to set right the slavery heritage with which historians have saddled modern American blacks. The conventional picture is that their plight is due not to “biological inferiority, but sociological circumstances. Blacks were the pitiful victims of a system of slavery so repressive that it undermined their sense of family, their desire for achievement, their propensity for industry, their independence of judgment, and their capacity for self-reliance.” This is the reason that our best young black historians have begun to reject one feature of the traditional abolitionist picture of the slave and to seek in the slave experience the foundations of black cultural, psychological, family, and moral resources. It is why they have rejected the “sambo” concept of the slave personality. For if this simpering, cringing, emasculated, childish creature were really the end product of the slave experience, what foundations are there for racial pride and hope for the future?

Had they thought of it, the authors might have pointed out also that the traditional picture of the pitiful, emasculated slave is the standard rationalization of policy makers for modern problems of the black minority. It runs like a litany through “background” papers on welfare, employment, the “matriarchal” family, the public schools. It turns up in Supreme Court opinions and presidential addresses. It has become the conventional way of shifting responsibility for the faults and failings of the free enterprise system to the shoulders of a forgotten and long discredited class of a remote period.

The Cliometricians have been forewarned by preliminary skirmishes of the ordeal that faces them from their critics. They report of confrontations with outraged colleagues the “mutual recriminations,” “fiery stares,” “thinly disguised insinuations of racism,” “caustic charges of naïve romanticism,” and a “torrent of passionate speeches.” This is to be expected. For they are not only taking on embattled colleagues over professional issues but also a deeply entrenched national credo.

Their findings, their data, and their methods should have the most thorough and unsparing criticism by the best qualified experts, traditional as well as scientific, especially those who speak their own language. Some of the more damaging criticism will probably come from the latter. The Cliometricians are vulnerable at many points and their conclusions provoke valid doubts. But it would be a great pity if the passions aroused by the controversy were wholly to discredit their approach and obscure the genuine merits of their contribution. They mark the start of a new period of slavery scholarship and some searching revisions of a national tradition.

This Issue

May 2, 1974