In the middle of the nineteenth century, an average American man stood conspicuously tall beside an average Dutch man. Historical estimates put the difference at roughly 5 to 8 centimetres. Today, the relationship runs the other way.
Across about 150 years, Dutch men gained close to 20 centimetres in average height, rising from around 165 centimetres to near 185. American men gained only about six centimetres over the same broad span. A population that began well behind did not merely catch up. It became the modern benchmark for tall stature.
The comparison is real, but it needs one boundary at the outset. This was not a 150-year experiment using one survey and one measurement method. It is a synthesis of military records, historical samples and modern health surveys. The endpoints are approximate, and the people measured at them belonged to different birth cohorts. The reversal is nevertheless much larger than those methodological uncertainties.
The lead America used to hold
The starting point is easy to forget because the modern stereotype runs in the opposite direction. From the eighteenth century into the early twentieth, American men ranked among the tallest recorded populations.
Economic historians John Komlos and Marieluise Baur estimated that the American advantage over Western and Northern Europeans lay between 3 and 9 centimetres in the mid-nineteenth century. A later review by evolutionary anthropologists Gert Stulp and Louise Barrett narrowed the comparison with Dutch men to about 5 to 8 centimetres.
That advantage probably reflected an unusually favorable balance of food and disease. The United States had abundant land, relatively low population density and better access to animal protein than much of crowded Europe. It was not a golden age of universal wellbeing, and the surviving American records are selective. But the broad pattern appears across multiple historical series.
The Netherlands, meanwhile, had passed through centuries in which urban crowding, infectious disease and uneven access to food constrained growth. Hans de Beer’s reconstruction of Dutch physical stature describes an elongated U-shaped history rather than a simple climb. Dutch people in parts of the Middle Ages may have been taller than those living during the country’s early industrial era.
Conditions around the nineteenth-century low point were harsh. Potato blight, rising cereal prices and cholera all fell within the childhoods of men later measured by the military. Estimates for Dutch conscripts around 1860 cluster near 165 centimetres, although some adjustments are needed because many were not yet fully grown.
A rise measured in generations
The Dutch ascent began in the second half of the nineteenth century and accelerated during the twentieth. By around 1930, one reconstruction placed Dutch men at 174.1 centimetres and narrowly ahead of other national populations. The margin widened in the following decades.
The clearest summary of the full reversal appears in Stulp and Barrett’s review of human height variation. It gives about 165 centimetres for Dutch military men in 1860 and near 185 centimetres roughly 150 years later. Over the same interval, it reports a gain of only six centimetres for US men.
A separate worldwide analysis by the NCD Risk Factor Collaboration helps anchor the modern end without pretending the figures are identical. It reanalyzed 1,472 population-based studies containing measured heights for more than 18.6 million people. In its 1996 birth cohort, Dutch men averaged an estimated 182.5 centimetres. The exact modern number shifts with birth year, age, ancestry, sampling and whether height is measured or self-reported. The broader result does not: Dutch men ended much taller than their nineteenth-century predecessors and taller on average than American men.
ScienceBlog recently looked at the related fall in America’s global height ranking. In the NCD collaboration’s century-long series, US men moved from third place to 37th. That was not because Americans became shorter. It happened because their modest gain was overtaken by much larger gains elsewhere.
What height can tell us about a childhood
Height is strongly heritable within populations. That familiar statement can be misleading when carried across time or national borders.
Heritability describes how much of the variation among people in a particular population and environment is statistically associated with genetic variation. It does not say that a population mean cannot move when the environment changes. A garden can contain plants whose different heights are strongly influenced by their genes while every plant still grows taller after receiving more light and water.
Human growth is more complicated, but the distinction is the same. The NCD collaboration identified fetal growth, nutrition and infection during childhood and adolescence as important environmental influences on adult height. Disease consumes energy, suppresses appetite, impairs nutrient absorption and can cause nutrient loss. Food that enters a child’s body is not all available to build bone and tissue.
For historians, adult height therefore works as an imperfect record of what is sometimes called net nutrition: the resources available for growth after the body has met the demands of disease and ordinary metabolism. It can reveal changes that money income alone misses.
This is a population-level tool, not a way to judge individuals. A shorter person is not necessarily less healthy, less well nourished or less advantaged than a taller neighbor. Genes, sex, ancestry, age and chance all matter. National averages also hide large differences within countries.
Why prosperity alone does not solve the puzzle
If income were the whole explanation, the United States should not have surrendered its lead to the Netherlands. American income per person was higher for much of the relevant period, and its total calorie supply was hardly meager.
The leading explanations instead concern how resources and health protections were distributed. Clean drinking water, sewerage, vaccination, safer childbirth, improved housing and better treatment of childhood disease all reduce the energy diverted away from growth. Reliable access to protein and micronutrients matters too.
Dutch dairy consumption is frequently singled out. Milk supplies protein and minerals, and feeding research has linked dairy with greater childhood growth in some settings. But “the Dutch are tall because they drink milk” is much neater than the evidence allows. Dairy consumption changed alongside income, sanitation, medical care and family life. It cannot be separated cleanly from that package by looking at national trends.
Social equality is another plausible part of the answer. Extra resources deliver diminishing gains once a child’s needs are already well met. Improvements among poorer families can therefore raise a national average more than further abundance among wealthy families. De Beer argued that the Dutch combination of income growth, health and social insurance, relatively modest inequality and systematic infant monitoring accounted for much of the late twentieth-century performance.
That remains an interpretation of converging historical evidence, not a randomized test. It would be too strong to assign a certain number of centimetres to universal health coverage, milk, clean water or any other single change.
Natural selection is possible, but cannot carry the whole story
A 2015 study added an evolutionary possibility. Using records from 94,516 people in the northern Netherlands, the authors found that taller men born between 1935 and 1967 tended to have more children. If that pattern persisted across generations, it could shift the frequency of height-related variants.
It does not explain a 20-centimetre rise on its own. The same researchers emphasized how implausibly intense selection would need to be for genes alone to move the average so far in roughly six generations. Their rough calculation suggested that the shortest 30 per cent of each generation would have had to leave no children at all. That did not happen.
The measured reproductive pattern also belongs to one region and a limited set of birth cohorts near the end of the long rise. It is evidence that selection may have contributed, not a genetic replacement for the environmental explanation.
Why old military records need careful handling
The nineteenth-century Dutch evidence owes much to conscription. Military systems measured large numbers of young men long before countries ran modern health surveys. They are among the best tools available for reconstructing population stature.
They also create traps. Some armies imposed minimum heights and failed to record men below the cutoff, which can make a population appear taller than it was. The Dutch lottery system was unusually useful because officials had reasons to record undersized men too. A 2020 audit of Dutch conscript records found several selection mechanisms, including death before examination, migration and opportunities for some social groups to evade registration, but concluded that their effects on mean height were limited.
Age remains important. Dutch measurement rules shifted, and many nineteenth-century men continued growing after their examination. De Beer notes that measurement age moved from about 18.75 years to 19.75 years and later 19.5 years. Researchers used repeat measurements and estimated corrections, but a corrected conscript average is not as direct as measuring fully grown adults under one stable protocol.
The records also explain why the famous comparison is framed around men. Historical height data for women are much thinner because they were not subject to mass military examination. The male result should not simply be relabeled as a trend for every Dutch or American person.
The great rise appears to have stopped
The Dutch did not keep adding centimetres indefinitely. Statistics Netherlands reported in 2021 that men born in 1980 averaged 183.9 centimetres, compared with 182.9 centimetres for those born in 2001. Women in the same comparison were 1.4 centimetres shorter in the younger cohort.
Changing population composition explained part of that difference, but not all of it. The agency also found that growth had halted among cohorts whose parents were both born in the Netherlands. Self-reported height and cohort composition make the precise size uncertain, yet the long upward run has at least plateaued.
That ending sharpens the historical lesson. National height is neither a fixed ethnic trait nor a simple prosperity ranking. It is a biological trace left by food, infection, sanitation, inequality, health systems and family conditions during the years when bodies grow.
America began the modern era with a striking height advantage and still gained several centimetres. The Netherlands simply changed much more. Across five or six generations, that difference was enough to reverse one of the clearest national height gaps in the historical record.