Malaria and Rome: A History of Malaria in Ancient Italy (54 page)

metres above sea level.The annual
estatatura
(exodus) from Grosseto had been instituted by 1333 by the government of Siena, which ruled Grosseto at the time and decreed that government officials should leave Grosseto in July and August each year to avoid malaria. It was only abolished as recently as 1897, by which time the increasing use of quinine and widespread bonifications were having a significant impact on malaria in the region.²⁸ It is very striking that mortality in Grosseto was so high even with such a large proportion of the permanent population practising avoidance behaviour. Malaria was also endemic in Grosseto among migrant workers who only spent the summer there.

It was also possible to take avoidance measures at certain times of the day and night. In the nineteenth century Romans frequently made excursions to Fiumicino (now the site of Rome’s airport, near Ostia), whose small resident population was 100% infected with malaria, during the day to have a seaside meal, but they returned to Rome before nightfall and did not sleep there. Sambon noted that at Ostia in 1900 ‘in July and August the
Anopheles
used to appear very punctually a few minutes after sunset and disappear again a few minutes after sunrise’.²⁹ De Tournon, without any understanding of the aetiology of malaria, advised visitors to Rome itself to avoid in dangerous parts of the city the evening strolls which are such a prominent feature of everyday life in so many Mediterranean towns (with the effect of avoiding mosquitoes).³⁰ Tommasi-Crudeli noted that in Rome in the nineteenth century: in some parts of the city the inhabitants, during the hot weather, remain indoors after sunset, because experience has taught them that during the first hour of the evening there is a risk of infection, but later on they emerge from their houses³¹

²⁸ Del Panta (1989: 29–31); Celuzza (1993: 151); Santi (1996: 150); Bueti and Corti (1998).

²⁹ Blewitt (1843: 528); Sambon (1901
a
: 199).

³⁰ De Tournon (1831: i. 216): [sc. during the dangerous season]
sur toutes choses, il faut éviter, pendant la soirée, les promenades, soit dans la campagne, soit dans les parties désertes de la ville
.

³¹ Tommasi-Crudeli (1892: 80); Wrigley (2000: 219) quoted Berlioz’ observation that Romans disappeared from their promenade on the Pincio ‘like a cloud of gnats’ (a doubly appropriate metaphor) at seven o’clock in the evening, cf. Blewitt (1843: 165).

Geographical contrasts

281

37. The site of the Circus Maximus at Rome, where chariot races were held. It is situated in a lowlying area between the Aventine and Palatine hills where there was a risk of malaria infection in the past.

Knight had earlier also noted this advice which was given by Roman doctors, adding the comment that ‘it would not be prudent to sleep with an open window’.³² Female
Anopheles
mosquitoes generally bite at night, commonly in the period leading up to mid-night. As recently as 1893, Filippo Pacelli noted the custom of shut-ting windows in the district on the left bank of the Tiber, near Isola Tiberina, to avoid malarial fevers.³³ Conversely, Lapi noted that in some areas of the city people kept the windows open all the time and remained healthy, but this is merely stating the obvious, namely that malaria did not occur everywhere.³⁴ Lancisi had also recommended keeping windows and doors closed. Evidently not everyone took all the precautions. In the fourth century  Ammianus Marcellinus wrote that poor Romans set off at dawn for the Circus Maximus to watch the chariot races, but neglect of essential precautions helps to explain why malaria was common in imperial Rome.³⁵ Nevertheless in antiquity as well the accumulation of ³² Knight (1805: 3).

³³ Carcaterra (1998: 560):
per schivare le febbri malariche, si è costretti a chiudere le finestre un’ora prima dell’Ave-Maria
.

³⁴ Lapi (1749: 81).

³⁵ Ammianus Marcellinus 28.4.31.

282

Geographical contrasts

empirical knowledge would have permitted the practice of avoidance behaviour, to try to minimize the known risks, as Herodotus shows when he describes sleeping in towers in Egypt to avoid mosquitoes (Ch. 4. 3 above). Apparently a hieroglyphic text from a temple at Denderah in Egypt advised people not to go outside their houses after the sunset in the weeks following the Nile flood.³⁶

This ancient Egyptian text recalls the statement of Athenaios that people in the Greek colony of Sybaris who wanted to avoid dying young must not see either the setting or the rising sun.³⁷ Tommasi-Crudeli, followed by W. H. S. Jones, interpreted this text as a reference to malaria in southern Italy.³⁸ This remains a possibility, since Lancisi recommended staying indoors until sunrise to avoid malaria, although the text of Athenaios could equally well simply be a reference to the legendary life of luxury enjoyed by the citizens of Sybaris.

Regardless of the correct interpretation of this text, it was indeed well known in antiquity that the average duration of life was much lower in marshy areas where malaria was endemic than in mountainous areas (like Tifernum) where it did not occur at all. The following question in the pseudo-Aristotelian
Problems
proves it beyond doubt:

Why do men grow old slowly in places with fresh and pure air, while those in hollow and marshy places grow old rapidly?³⁹

As a comparison, the early modern tradition about Ninfa, which was abandoned because of malaria in the seventeenth century, should be recalled:

The modern legend that an evil spirit waits in ambush for passing young adults, to make them at once grow old miserably, is a personification of the wickedness of the air, which makes the youthfulness of those who breath its exhalations decay.⁴⁰

³⁶ Capasso (1985: 304) quoted this Italian translation of the Egyptian text:
non uscire di casa dopo il tramonto del sole nelle settimane che seguono l’ingrossamento del Nilo.

³⁷ Athenaios 12.520a: tÏn boulÎmenon ƒn Sub3rei m¶ prÏ mo≤raß åpoqane∏n oÇte duÎmenon oÇte ån≤sconta tÏn ~lion Ør$n de∏.

³⁸ Dunbabin (1948: 80, 216–17) emphasized the scarcity of documentary evidence with regard to the possibility of malaria in Sybaris in the archaic period.

³⁹ [Aristotle,]
Problems
14.7.909b: Di¤ t≤ oÈ m†n ƒn to∏ß eÛpnÎoiß tÎpoiß bradvwß ghr3skousin, oÈ d† ƒn to∏ß ko≤loiß ka≥ ‰l*desi tacvwß.

⁴⁰ Tomassetti (1910: ii. 394):
la leggenda odierna che una fata malignamente sta in agguato dei giovani che passano per farli subito miseramente invecchiare è una personificazione della malignità dell’aria, che fa intristire la giovinezza di chi ne contrae l’effluvio
. See also Hadermann-Misguich (1986: 23–46, esp.

Geographical contrasts

283

The prevailing view among ancient historians is to assume a fairly low life expectancy at birth across the whole of Roman Italy. Lo Cascio, following Hopkins, is a typical example of the traditional view:

Life-expectancy at birth cannot have been higher than, say, 25 for males . . . life-expectancy at birth for females must not have been higher than 25

and is likely to have been a bit lower . . . [this is] the assumption that we are entitled to make.⁴¹

The reality of Roman Italy was more subtle, more complicated, and more interesting than this bland uniformity. There were some extremely unhealthy localities,with a life expectancy at birth hover-ing around 20, in some cases no more than a few kilometres away from other localities, where life expectancy at birth may have been as high as 40 or 50.⁴² As was noted at the beginning of this book, the Italian demographers del Panta and Rettaroli found that the population of Grosseto, ravaged by malaria in the nineteenth century, did not have an age-structure corresponding to any of the main types described in the standard sets of model life-tables and based on data from modern European populations. The closest parallels instead come from
African populations
. African population structures in central Italy as recently as the middle of the nineteenth century ? African population structures in central Italy in antiquity? This will come as a tremendous shock to those historians who are accustomed to rely upon sets of model life-tables derived from modern European populations in their research on classical antiquity. A major flaw of the bulk of research carried out into the demography of the Roman Empire over the last thirty years has been the failure to realize that the standard sets of model life-tables do not encompass the entire universe of possibilities. In fact, they 43–5 on malaria). She noted that the eighteenth-century historian Pietro Pantanelli stated that no one in the region survived past the age of forty.

⁴¹ Lo Cascio (1994: 36).

⁴² Herlihy and Klapisch-Zuber (1985: 199–201) noted the existence of populations with an older age structure in highland areas in the territory of Renaissance Florence. They explained this in terms of emigration of the young from the uplands. Although this did undoubtedly happen, numerous other studies leave no doubt that in general uplands were healthier than lowlands (see Ch. 4. 2 above). In antiquity it was believed that fertility levels were high in mountain populations in central Italy, according to [Aristotle,]
peri thaumasion akousmaton
80.836a. This is highly plausible even though the author of this work had no statistical evidence whatsoever for his assertion.

284

Geographical contrasts

do not cover populations heavily ravaged by malaria, or tuberculosis, or smallpox, for example.⁴³ Thus the results of a general assessment of the applicability of these model life-tables to ancient populations, as called for by Hopkins in 1966, are intrinsically likely to be negative for many ancient populations, especially in southern and central Italy. It is worth recalling that Herlihy and Klapisch-Zuber refrained from calculating a life-table for the population of Florence as reconstructed from the
Catasto
of  1427 for the following reason:

Our age pyramids emphasize the instability of the Tuscan population described in the
Catasto
. At the same time, they reveal the obstacles involved in utilizing for historical purposes model life tables which are based on stable or stationary populations . . . the age pyramid of 1427 does not fit any of the theoretical distributions closely.⁴⁴

Of course it can be argued that the situation in Florence in 1427

was exceptional because of the effects of massive plague epidemics.

Nevertheless the fact remains that the first major source of quanti-fiable demographic information in Italian history provides no support whatsoever for the hypothesis that model life-tables based on data from normal modern populations are likely to be useful for ancient populations afflicted by diseases which are no longer endemic in modern Europe. Moreover the situation in antiquity was itself not static, since disease evolution and history is a never-ending process. The spread of malaria in Italy in the first millennium  created new demographic patterns in both time and space. The demographic patterns of much of Latium and Etruria in the first century  were probably completely different from what they had been in the seventh century .

There is yet another methodological issue of fundamental importance at stake here. Hopkins entitled his well-known article ‘
On the probable age-structure of the Roman population
’. This begs the question of whether the Roman population had an age-structure.

Or, to put the question more precisely, making explicit a crucial assumption implicitly made by Hopkins, did the Roman population
as a whole
have
a single age-structure
? The answer to that question ⁴³ Coale and Demeny (1983: 33) stated that ‘any extraordinary incidence of a cause of death that is highly age-and-sex specific produces a mortality schedule that naturally does not conform to the model tables’. Populations affected by malaria evidently fall into this category.

⁴⁴ Herlihy and Klapisch-Zuber (1985: 195).

Geographical contrasts

285

must be no, taking account of the prevalence in historical populations of the type of regional variation which has been discussed here. The inference is clear. The age-structure of the Roman population is an illegitimate concept. The homogeneity of modern populations is a consequence of the process of modernization.

Only an anachronistic, modernizing approach to ancient history can imagine such homogeneity in ancient populations.

This page intentionally left blank

R E F E R E N C E S

Adekile, A. D. (1992). ‘Anthropology of the bs gene-flow from west Africa to north Africa, the Mediterranean, and southern Europe’,
Hemoglobin
, 16: 105–21.

Agricoltura e società nella Maremma grossetana dell ’800
(1980).
Biblioteca Storica Toscana
, iv. (Florence).

Aitken, L. (1873). ‘The sanitary state of Rome’,
British Medical Journal
, 1: 311–12.

Alexander, D. (1984). ‘The reclamation of Val-di-Chiana (Tuscany)’, Annals of the Association of American Geographers
, 74: 527–50.

Amadouny, V. M. (1997). ‘The campaign against malaria in Transjordan, 1926–1946: epidemiology, geography, and politics’,
Journal of the History of Medicine and Allied Sciences
, 52: 453–84.