Malaria and Rome: A History of Malaria in Ancient Italy

by Robert Sallares

  The reason for the genesis of an economy based on mass chattel slavery in western central Italy (and large parts of the south of Italy) in antiquity is fundamentally exactly the same as the reason why it arose in the western hemisphere after Columbus. A spreading disease (in this case P. falciparum malaria) gradually, by a slow process of attrition, either killed or forced to emigrate the bulk of the indigenous farming population in Latium and southern Etruria, thus providing the manpower for Roman colonization elsewhere.

  That, in turn, created a vacuum, a massive labour shortage, on fertile agricultural land where free men were reluctant to work because of the disease. That labour shortage could only be filled by importing large numbers of chattel slaves. Even though the later slave revolts in Italy led by Spartacus and in Sicily have attracted much more attention from modern historians, it is very striking that the first attested slave revolt in Roman Italy occurred precisely in the Pontine Marshes, in 198 . Livy records that slaves, acquired Roman Campagna

  253

  as part of the proceeds of the Second Punic War, in the territories of Setia, Norba, and Circeii conspired with the slaves of Carthaginian hostages who were being held at Setia to attack these towns.

  Military action was required to suppress the revolt. Viewed in the light of the analysis given here, the location of this first slave revolt was not an accident. After the Second Punic War and such later acts of Roman aggrandisement as the destruction of Epirus in 167

  , slaves were extremely cheap. Their owners need not worry if the gangs of slaves employed on the land, such as those seen by Tiberius Gracchus, suffered extremely high mortality rates from malaria; slaves were very cheap and easy to replace. For as long as the slave trade to the western hemisphere continued, slave owners there took exactly the same attitude, tolerating very high mortality rates among their slave labour forces and assuming that imported slaves would not live more than a few years on average. Yet they were able to make large profits under those circumstances. The slaves who ended up in Setia were brought there because of the manpower shortage noted earlier. The slave revolt there in 198 

  shows that the slave economy witnessed by Tiberius Gracchus was already taking root in western central Italy two generations before his excursion through south Etruria which had such catastrophic consequences for the Roman Republic. The demand for labour preceded and called forth supply in Roman Italy, exactly as it did in the United States, according to the arguments of Benjamin Franklin. What happened in Setia was a microcosm of what happened throughout western central Italy and in large parts of southern Italy as well. The slaves who were brought to Latium and Etruria were forced to constitute the labour forces of the Roman villas which came to populate the landscape, villas such as Settefinestre, according to the villa-based slave mode of production characteristic of these regions during the Roman Empire, as described by Carandini.⁴⁸

  To replace the concept of the latifundium, which he maintains lacks clarity, Carandini postulated a typology of two types of Roman villas. He distinguished the villa centrale from the villa periferica. He suggested that the villa centrale was characteristic of the suburbana regio Italiae. It was relatively small and practised intensive ⁴⁸ Franklin (1751); Livy 32.26 for the events of 198 ; also 33.36.1–3 for another slave revolt in Etruria a couple of years later; Carandini (1985).

  254

  Roman Campagna

  agriculture using mainly slave labour. All this is acceptable. The point at issue here is the assertion that the villa centrale was found on fertile lands and in a healthy environment.⁴⁹ In contrast the villa periferica was found mainly in more outlying, isolated regions, located on less fertile land and in less healthy conditions.⁵⁰ Extensive agriculture which required a smaller labour input was the norm. That labour was supplied principally by serfs or tenants rather than slaves. This typology implies that unhealthy conditions were to be found mainly on the fringes of the agricultural landscape of central Italy in Roman times. However, there is plenty of evidence, as this book demonstrates, for malaria right at the heart of the Roman world, even in the city of Rome itself. There were villas in the extremely fertile territory of Setia, which undoubtedly employed the slaves attested during the slave revolt of 198  and later supplied wine for the personal consumption of the Roman emperors. Not a fringe area. Yet the land of Setia was pestilential.

  This example alone wrecks Carandini’s typology in so far as it concerns healthy/unhealthy conditions. Moreover, as we have already seen, in general it was fertile lowlands in particular that were likely to be unhealthy, while less fertile mountainous regions were usually healthy. The argument proposed here instead is that it was precisely because much of the best agricultural land of central Italy was unhealthy, owing to malaria, that the Roman élite was forced to import large numbers of slaves in order to get the land worked.

  Mass chattel slavery was an adaptation to malaria.

  One of the plays of Plautus explicitly mentions the idea of slaves dying rapidly in summer after being forced to perform agricultural labour on pestilential estates. The six months’ life expectancy on farms where malaria was endemic recalls the six months’ life expectancy mentioned in the traditional Italian proverb about the Maremma quoted in Chapter 7 above. Of course Plautus’ comments on a particularly undesirable farm, where everything that could go wrong did go wrong, were intended to be funny, but it was a type of comedy which could only be enjoyed by slave owners.

  Nevertheless it shows that the type of analysis advocated here was well within the consciousness of ancient Romans:

  moreover none of the Syrian [sc. slaves]- the most enduring of men—who ⁴⁹ Carandini (1995: 33): Si trova su terreni fertili e in ambiente salubre.

  ⁵⁰ Carandini (1995: 34): È posta su terreni meno fertili e in condizioni meno salubri.

  Roman Campagna

  255

  lived there for six months, is alive: all of them were killed by the disease that strikes at the summer solstice⁵¹

  There was an alternative, or at least a possible supplement, to slave labour which deserves some attention, bearing in mind that Mediterranean agriculture requires a lot of labour during harvest-ing, in the summer, which is not required for the rest of the year.

  The alternative is the employment of hired labour from seasonal migrant workers. Varro’s recommendation, quoted earlier (Ch. 4.

  2 above), that hired labour should be employed in unhealthy regions, rather than slaves, should be recalled. In the early modern period a considerable proportion of the labour input was provided by free labourers who migrated from the uplands of Abruzzo and Marche to gather in the harvest in the Roman Campagna, and from Liguria and Emilia to Tuscany, and also from Abruzzo to the Tavoliere. Bercé described how in 1593, for example, forty thousand labourers arrived in the vicinity of Rome, first to reap wheat and barley, then to thresh it, and afterwards to harvest the grapes.⁵²

  Those labourers were prepared to take the risk of catching malaria, sleeping out in the fields in the summer, because it was the only way in which they could make a living. External colonization gave poor Romans alternatives in antiquity during the Republic. The employment of hired labourers inevitably meant that large landowners did not have to bear the costs if the labourers died from malaria. Cipolla described from the reports of the Florentine health magistrates how seasonal workers who had gone from Liguria to work in the Tuscan Maremma had become ill in early autumn at Bibbona in 1614 on their way home and died during the winter months. In this way malaria was able to influence the demography of parts of Italy in which it did not occur. Del Panta attributed the fact that the territory of the Senese had very high mortality levels in the early modern period, even though it was a considerable distance from the coast, to the effects of malaria on labourers who migrated seasonally to the Maremma.⁵³ The ⁵¹ Plautus, Trinummus 542–4: tum autem Surorum, genus quod patientisumumst | hominum, nemo extat qui ibi sex menses vixerit: | ita cuncti
solstitiali morbo decidunt.

  ⁵² Bercé (1989: 241), citing Paolo Paruta; Delano Smith (1978: 145); Sorcinelli (1977: 95–6) linked malarial fevers in the Marche to seasonal migrations of farm labourers, and also to the construction of the Bologna–Ancona railway line.

  ⁵³ Cipolla (1992: 51–3); del Panta et al. (1996: 193–6); Scheidel (1994 a: 175, 187–8, 216) discussed wage labour in gravia loca.

  256

  Roman Campagna

  combination of slaves for the permanent labour forces of villas with seasonal labour for the harvest and associated tasks explains how the Roman élite was able to extract a substantial amount of agricultural production from a land which was shunned by free peasants because of the ‘reckoning with death’, ratio cum orco, mentioned by Varro. The reckoning with death from malaria was exactly the same in the Roman Campagna in the early modern period as it was in antiquity: the difference was that early modern Rome was not a slave society:

  The workman does not languish voluntarily where the cause of illness and death is close by and powerful.⁵⁴

  As was noted earlier, Celli advocated ‘the theory that . . . periods of prosperity coincided with periods of attenuation in the severity of the malarial fever’.⁵⁵ He knew that many Roman villas were constructed in the Roman Campagna, above all in the period from Augustus to the Antonines, and thought that the economic prosperity indicated by these villas could only be explained on the assumption that malaria, which was certainly present, was less virulent at that time. Celli proposed a cycle of fluctuations of virulence of P. falciparum malaria in the Roman Campagna throughout history. The idea was accepted by other leading Italian malariologists, such as Missiroli for example, and by medical historians like Bercé and North. It has in fact never been subsequently seriously re-examined.⁵⁶ Yet the whole theory is quite weak. Of course Celli, writing at the end of the last century, had no direct scientific evidence for fluctuations in the virulence of P. falciparum, and there is little available today; as was observed in Chapter 3 above, current scientific research into parasite evolution and epidemiology suggests that extreme virulence is adaptive for P. falciparum. Moreover the modern populations of regions with endemic malaria in the past tend to have high frequencies of human genetic mutations which give some resistance to malaria (see Ch. 5. 3 above, and the discussion of Ravenna in Ch. 4. 2 above). This implies intense pressure by severe malaria as an agent of natural selection on ⁵⁴ F. Giordano, Condizione topografiche e fisice . . ., in Monografia (1881: lxiii): dove è prossima e forte la causa di malattia e di morte, non si perita volentieri il lavoratore.

  ⁵⁵ Celli (1933: 109).

  ⁵⁶ North (1896: 86); Missiroli (1938: 5–6); Bercé (1989); Hofmann (1956: cols. 1203–6) on Celli’s theory of cycles of malaria as applied to antiquity.

  Roman Campagna

  257

  human populations in Italy in the past and contradicts Celli’s theory of mild malaria.⁵⁷

  Celli’s evidence was entirely indirect, basically the remains of villas in the Early Empire (and churches in the Late Empire) as signs of prosperity. A typical example is Cicero’s villa at Astura on the coast of southern Latium, a beautiful location. However, it must be noted that Virgil hinted that the marshes of Astura were unhealthy.⁵⁸ Moreover, according to his letters, Cicero stayed in his villa at Astura principally in the spring—the season of the year when transmission of malaria ceased or was very low. After the end of the Republic most of the villas on the coasts of Etruria and Latium eventually became the emperor’s property and probably rarely saw their owner. Domitian’s villa by the side of the Lago di Paola in the Pontine Marshes has left very imposing archaeological remains which have recently been restored. However, it appears to have been a single-phase site which was neglected after Domitian’s death. The letter of Pliny the Younger quoted earlier (Ch. 8 above) about the villa of Regulus on the Tiber confirms that some Roman villas were located in areas which were known to be unhealthy.

  Moreover Celli did not consider who built these villas, or ask who were the people who formed the labour force working out of these villas afterwards. He did not consider the importance of the fact that these villas had a labour force made up of slaves. He did not pay any attention, as a possible parallel, to the slave societies of the western hemisphere, where plantation owners were quite happy to make big profits by employing large numbers of slaves with a low life expectancy in very unhealthy environments.⁵⁹ He did not pay ⁵⁷ In using this line of argument it is necessary to take account of human population movements and migrations. The modern populations of some of the regions of Italy which formerly had endemic malaria have moved there recently from other areas. For example, the modern population of the new towns of the Pontine plain is largely descended from colonists sent there from the north of Italy by Mussolini in the 1930s (Gaspari 1985). Under such circumstances it would obviously be foolish to use the genetics of the modern population of this region to attempt to shed light on ancient malaria. Similar considerations apply to the population of the city of Rome itself, which has been a magnet for migrants not just in modern times but throughout history (see Ch. 11 below).

  ⁵⁸ Cicero, Letters to Atticus, 257, ed. Shackleton-Bailey (1965–70) (written on 14 March 45

  ) described Astura as a locus amoenus (a pleasant place) (unlike the Pontine Marshes); Virgil, Aeneid 7.801.

  ⁵⁹ Giglioli (1972) studied the large reductions in mortality following the eradication of malaria from the sugar plantations of Guyana. For comparative evidence from North America see also J. F. Smith (1985: 7, 136–7); Joyner (1984: 35–7, 70); Savitt (1978: 17–35); Merrens and Terry (1984); Dubisch (1985); Duffy (1988); Dobson (1989); Dusinberre (1996); Rutman and Rutman (1997).

  258

  Roman Campagna

  enough attention to the fact, accepted by all historians writing about the Roman Campagna from Ashby to Brunt, that there were very few significant towns or villages in Latium populated by free people during the Roman Empire. Free people had a choice of where to live, but slaves did not have any choice. The verdict of free people is much more significant in assessing the problem of the desirability of living in Latium. Above all, Celli, surprising as it may seem in a work of medical history, did not pay enough attention to one crucial category of literary evidence from antiquity, namely the evidence provided by the medical writers. It is worth quoting one paragraph from the English translation of Celli’s book to illustrate the problem:

  The best description of the character of the different forms of fever is given by Galenus; he . . . describes vividly the . . . aestivo-autumnal fevers again recurrent in our days and which in those days were called ‘Emitritea fevers’, and were widely spread in Rome in summer and in autumn. The heavy occurrence of jaundice and dropsy could be daily observed, symptoms undoubtedly produced by malaria. This disease was widely spread at a time when we know from many signs and proofs [sc. villas, etc.] that the pest was diminishing in virulence.⁶⁰

  The massive problem with this line of argument is that Galen does not say anything whatsoever about the pest diminishing in virulence! On the contrary, Galen explicitly described the semitertian fevers, which were so common in Rome, as extremely dangerous (kindunodvstatoß), as has already been seen (Ch. 8 above).

  Moreover the symptoms of jaundice and dropsy noted by Celli himself indicate a severe disease. The direct testimony of Galen with regard to the virulence of the disease is much more significant than the extremely indirect evidence of villas which were largely populated by slaves. Celli’s argument for an attenuation of the severity of malaria during the Roman Empire, as a cyclical down-turn after the ravages which he argued it caused during the Late Republic, is very weak. There is no space to examine here in detail the possibility of fluctuations in the virulence of malaria in the Roman Campagna during more recent periods of history. That would require another book, which would have to be based on extensive research in archives and libraries in Italy, but the
weak-

  ⁶⁰ Celli (1933: 47, 111–17).

  Roman Campagna

  259

  ness of Celli’s arguments about antiquity suggests that the enterprise would be worth undertaking.

  There is only space here to note that some of the evidence presented by Celli himself for later periods contradicts his own theory in exactly the same way that Galen contradicted it in antiquity. For example, Celli argued, again on the basis of construction work (analogous to the villas of antiquity), that another period of attenuation of the severity of malaria occurred from the mid-fourteenth to the seventeenth century . However, he noted that the leading seventeenth-century historian of the city of Rome, Father Alexander Donatus, observed that the villas of that period were built in hilly locations precisely because the lowlands were unhealthy: The reason is to be sought in the unhealthy and noxious wildness of the air. For the opinion of the doctor Alexander Petronius, expressed in notable works, is confirmed by experience, with everyone’s agreement: the summer residences of the citizens in the vineyards around the city are unhealthy, and are not far away from the danger of ill-health. Consequently very few villas can be counted not only on the land along the Tiber, but even on the land around the city, despite the presence of so many noblemen and the abundance of wealth. The villas are located instead a little further away, on the ridges of Tibur, Tusculum, and Mt.

  Albanus.⁶¹

  Donatus’ evidence undermines Celli’s own theory. It is very reveal-ing to compare Donatus’ account with the archaeological evidence for the distribution of ancient Roman villas around Praeneste, for example, as described by Andreussi:

  All these villas arose either on the flat summits of hills separated by deep ravines, or on the southern slopes of the same hills, with a good aspect and view.⁶²

  The similarity to the situation described by Donatus in the seventeenth century is obvious.⁶³ It is likely that Donatus was right, ⁶¹ Donatus (1694: bk iii. ch. 21, p. 272): Causa rejicienda est in aëris intemperiem insalubrem, et gravem. Nam quod Alexander Petronius Medicus insignis typis evulgavit, omnium assensu, et experientia com-probatur; insalubres esse Civibus circum Urbem aestivas in vineis stationes, nec procul a periculo valetudinis abesse. Itaque non modo secundum Tiberim, sed etiam in agro suburbano, in tanta Principum et divitiarum copia, paucissimae numerantur Villae, quae paulo remotiora Tiburis, Tusculi et Albae juga insederunt.