Paleodemography is the study of human demography in antiquity and prehistory. More specifically, paleodemography looks at the changes in pre-modern populations in order to determine something about the influences on the lifespan and health of earlier peoples.
Reconstruction of ancient population sizes and dynamics are based on bioarchaeology, ancient DNA as well as inference from modern population genetics.
World population curve. Log scales are used for the population axis, to allow better reading of the data, and for the time axis, to compensate for the higher availability of census data for recent years.
Skeletal analysis can also yield information such as an estimation of age at time of death. There are numerous methods that can be used, and it is best to field questions of further interest to an osteologist or bioarchaeologist. In addition to age estimation and sex estimation, someone versed in basic osteology can ascertain a minimum number of individuals (or MNI) in cluttered contexts—such as in mass graves or an ossuary. This is important, as it is not always obvious how many bodies compose the bones sitting in a heap as they are excavated.
Occasionally, historical disease prevalnce for illnesses such as leprosy can also be determined from bone restructuring and deterioration. Paleopathology, as these investigations are called, can be useful in accurate estimation of mortality rates.
The increasing availability of DNA sequencing since the late 1990s has allowed estimates on Paleolithic effective population sizes Such models suggest a human effective population size of the order of 10,000 individuals for the Late Pleistocene. This includes only the breeding population that produced descendants over the long term, and the actual population may have been substantially larger (in the six digits). Sherry et al. (1997) based on Alu elements estimated a roughly constant effective population size of the order of 18,000 individuals for the population of Homo ancestral to modern humans over the past one to two million years. For the time of speciation of Homo sapiens, ca. 130,000 years ago, Sjödin et al. (2012) estimate an effective population size of the order of 10,000 to 30,000 individuals, and infer an actual "census population" of early Homo sapiens of roughly 100,000 to 300,000 individuals. The authors also note that their model disfavours the assumption of an early (pre-Out-of-Africa) population bottleneck affecting all of Homo sapiens.
Estimates of the total number of humans who have ever lived range in the order of 100 billion. It is difficult for estimates to be better than rough approximations, as even modern population estimates are fraught with uncertainties on the order of 3% to 5%.Kapitza (1996) cites estimates ranging between 80 and 150 billion. Haub (1995) prepared another figure, updated in 2002 and 2011; the 2011 figure was approximately 107 billion. Haub characterized this figure as an estimate that required "selecting population sizes for different points from antiquity to the present and applying assumed birth rates to each period".
Robust population data only exists for the last two or three centuries. Until the late 18th century, few governments had ever performed an accurate census. In many early attempts, such as in Ancient Egypt and the Persian Empire, the focus was on counting merely a subset of the population for purposes of taxation or military service. Thus, there is a significant margin of error when estimating ancient global populations.
Another critical factor for such an estimate is the question of pre-modern infant mortality rates; these figures are very difficult to estimate for ancient times due to a lack of accurate records. Haub (1995) estimates that around 40% of those who have ever lived did not survive beyond their first birthday. Haub also stated that "life expectancy at birth probably averaged only about ten years for most of human history".