We can consider the ecology of disease to be the sum total of all the influences on pathogens and their hosts and, because of the interdependence of the two for disease expression, the internal structures and systems of both that bear upon a disease question.

Thus we are clearly considering a structure of rela­tions whose complexity surpasses all comprehen­sion. Such may also be said of a single disease. No one has yet fully defined all that constitutes the expression of even one disease, and such an explana­tion would be its ecology.

With such cautions in mind, I should like to offer a few general comments on human disease and its expression, and then set forth a very limited number of aspects of the totality of my subject, which I feel can be discussed. For most of the sojourn of human­kind on Earth, we have only skeletal remains upon which to build any concept of disease in the past. Even after the advent of agriculture and the earliest civilizations, we have little upon which to develop a coherent view. General trends are perceptible in clas­sical times, and very sketchy numbers can be offered for population in the late medieval period. Modern times have, of course, brought masses of statistics, but these as often confuse as enlighten. Still, it is possible, at least, to discern some of the impact of disease upon demography and to infer some aspects of the influences that played upon certain diseases. Because it is so difficult to identify diseases in the past, I confine myself largely to the specific diseases of plague, leprosy, tuberculosis, and smallpox. Other­wise, the influence of epidemic disease of unknown type will be the other main consideration, along with some presumed diseases of childhood. There is at least some reason to feel that the four specific diseases may be identified as factors in human de­mography in the past, as can epidemic and endemic diseases whose exact nature is unidentifiable.

All human populations share some common fea­tures, and European populations carry some particu­lar features, which can help us at least conjecture about factors influencing the expression of disease. For example, about 105 males are born per 100 fe­males (Russell 1977). Although malnutrition can cause disease, major population losses do not neces­sarily reflect malnourishment (Wilson 1957; Wrig­ley 1962; Rotberg and Rabb 1983). Certainly the diseases of postindustrial society are those of an abundance of food. Despite the fascination since an­cient times with food as a cause, predisposing factor, and therapeutic agent in disease, very little is actu­ally known about the relationship between food and health beyond the effects of some vitamin-deficient states, and full-blown protein-energy malnutrition.

The ability of human populations to replenish themselves is limited. A birthrate above 50 per 1,000 population per year is truly exceptional, and above 30 is quite uncommon (Wrigley 1962). There are well- known age and sex predilections for a variety of dis­eases, but they are as etiologically obscure as they are epidemiologically well documented (Wilson 1957). For example, although infants are extremely suscepti­ble to tuberculosis, children between the ages of 5 and 15 exhibit a relative immunity to the disease. Then a sex predilection occurs, with women much more com­monly the victims from about 25 years up to around age 45, at which time men are the predominant vic­tims of the disease until late in life (Wilson 1957). These observations are based on data gathered mainly in the British Isles during the nineteenth century, but no one knows the extent to which they represent the result of metabolic versus environmen­tal factors in pathogen and host. It is also generally felt that societies tended to have more males than females among adults because more women than men die between the ages of 25 and 45 (Russell 1977). This generalization, however, is based to a very great extent on the perceived epidemiology of tuberculosis, a disease neither universal nor of equal importance in all societies.

In Europe, several generalizations are usually ac­cepted regarding human populations. Since classical times until the Industrial Revolution, cities have harbored more women than men. Cities have also been unable to replenish their own populations, at least from medieval times until the nineteenth cen­tury, because of the unfavorable disease climate of urban life (McNeill 1976; Russell 1977). Before the nineteenth century, the vast majority of humans in Europe lived in villages or on farms, with only a very small fraction residing in cities (Guillaume and Poussou 1970; Russell 1977). The total population of preindustrial societies contained at least one-third children, of whom less than half lived to adulthood from medieval until modern times (Flinn 1985).

Humans also affect their environment and in turn affect their own disease ecology. Practices we take for granted, such as the domestication of animals, change another species’ disease environment, and the interaction of domesticated animals with hu­mans has often led to human disease (Lovell 1957). An example of a disease that became of at least minor importance among humans primarily in this way is brucellosis, usually introduced to humans through goats. During the Crimean War, for exam­ple, Malta became an endemic area for brucellosis; in 1906 the means of human infection was described (Alausa 1983).

Disease in turn may alter the character of human populations. After major epidemics of infectious dis­ease associated with high mortality, a period of mark­edly reduced mortality generally begins. It is felt that such epidemics “weed out,” at least in some cases, the weaker elements in a population, leaving a healthier one behind (Wrigley 1962). On the other hand, diseases such as plague and some rickettsial infections often kill the healthy.

If we limit our generalizations about humans in Europe, our knowledge of their pathogens is even sketchier. The whole question of virulence among various bacteria is extremely complicated, and, since it is not well understood in modem infections and since organisms can mutate, virulence can be assessed only by effect. The effect of an organism in turn is related to host, pathogen, and environment, so that any comments about organisms in the past is unrealistic.

In general, organisms are tissue specific. Clos­tridium tetani, for example, is harmless to intact skin, but can be deadly when in contact with abraided or necrotic tissue (Wilson 1957). The longer an organism is in contact with the environment and outside a host, the likelier it will be damaged. Organ­isms that affect the gastrointestinal tract, such as Salmonella typhi, are able to withstand environmen­tal pressures longer than those not waterborne or foodbome (as a generalization) (Wilson 1957).

Despite the staggering complexity of host-patho­gen relationships, the presence of certain diseases can reveal important aspects of human populations. Measles is a human disease only and requires at least a population of some hundred thousands in some con­tact to remain endemic (Bailey 1954, 1957; Bartlett 1957,1960; Becker 1979; Smith 1983). Diseases such as measles and smallpox (another human disease whose pathogen needs a fairly large population to remain active) tend to become childhood diseases (Mc­Neill 1976; Smith 1983). Thus the first observation of measles would tell us when a communicating popula­tion of humans reached a certain level. Unfortu­nately, we do not know when this occurred in Europe because we do not know when measles first appeared. The clinical expression of a given disease can also tell us about genetic factors in a population. Susceptibil­ity to leprosy, in both its polar forms (although more in tuberculoid leprosy), is in part genetically medi­ated (Smith 1979; Eden et al. 1980; Keyu et al. 1985). When measles and TB have fulminant courses with high and dramatic mortality, they are present in a previously unexposed population (Wilson 1957; Mc­Neill 1976). The same may have been the case when syphilis appeared in Europe at the end of the fifteenth century, but this is hardly an undisputed subject. It is also becoming apparent that some resistance to par­ticular diseases reflects prior exposure to others that confer cross-immunity (Ell 1984a).

Characteristics of childhood diseases, especially measles, have become the subject of mathematical modeling. Birthrate has been shown to be an impor­tant factor in determining whether or not a disease like measles can remain endemic and in predicting interepidemic periods (Bailey 1957; Bartlett 1957; Becker 1979).

In the following pages, I try to discern, within what is known about the demographic history of Europe, the effects of disease on that history. Partly because war has been an invariable aspect of human behav­ior, and partly because it reflects so much else about a society, I focus often on military strategy, on the com­position and size of armies, as well as on the direct effects of disease upon armies. The time frame consid­ered extends from the hunter-gatherers of prehistoric Europe to late twentieth-century society. I have very little to say about medicine, because I do not believe it has had a significant effect on the ecology of disease except in very limited circumstances.

Geographically, I define “Europe” as extending as far east as western Poland down through Greece, as far north as Scandinavia (but excluding Iceland), as far west as the British Isles, France, and Spain, and as far south as Greece, Italy, and Malta.