Howell's study of the !Kung surprised many by reporting a substantial number of older people, and a mortality schedule well above the lowest in the standard models. Hill and Hurtado's study of the Ache showed the same thing. Were our traditional beliefs about hunter-gatherers wrong, or were the !Kung and the Ache just two fortunate exceptions? Detailed information from another population should help us decide. Hence, my purpose in this chapter is to report the mortality suffered by the Hadza during our study period. I will also look at evidence about mortality at previous times in the twentieth century so that we can set Hadza mortality in the larger context of historical and geographical variation in hunter-gatherer mortality.

Commonly used measures of mortality include crude death rate (CDR), average age at death, life expectancy, age-specific probability of death, and survival. The proportion of old people in the population and the maximum lifespan are also commonly reported. Each measure has its advantages and disadvantages, and some of them can mislead the unwary. For example, life expectancy at birth (e0) is heavily influenced by the mortality of infants and children, and thus is a good reflection of a widely accepted standard of life. However, e0 tells us less about life expectancy at later ages (ex), or about the proportion of old people in the population. Life expectancy at birth also reflects fertility (with high fertility, there are more of the vulnerable youngsters). Much of the increase in life expectancy in the industrialized nations during the past 200 years has arisen from a decrease in infant and child mortality. Only very recently has there been much change in life expectancy at age 60, for instance (Oeppen and Vaupel, 2002, and see discussion in Hawkes, 2006; Hawkes and Blurton Jones, 2005; and Gurven and Kaplan, 2007).

Population increase and decrease affect some of the commonly used measures such as average age at death. The proportion of old people in the population also cannot show us whether it signifies high or low mortality. A high proportion of old people can arise if the population is declining, as well as because old people survive quite well. By assembling as many of these measures as we can, we can offer a useful picture of mortality in any population.