Jagailloux (1986: 262) regarded the debilitating effects of P. vivax malaria as an important factor in mortality in Egypt.
Demography of malaria
155
expectancy of 33 in malarial parishes in England and the life expectancy at birth of 20 in Grosseto only a few years later represents the difference between P. vivax and the more virulent P. falciparum (see also discussion below). However, P. vivax was also very common in Mediterranean countries in antiquity and will have made its mark on mortality patterns there as well. Where both species of malaria coexist in an endemic form, as in western central Italy in the past, P. falciparum often tends to outcompete P. vivax because of its higher rate of reproduction, especially in very hot years when environmental conditions are most favourable for it.¹²¹
Some recent research in Vanuatu in the Pacific has generated the hypothesis that prior infections with P. vivax may confer some immunity to subsequent infections with P. falciparum. The modern experience that P. vivax is a relatively mild disease is then invoked to suggest that cross-immunity could give rise to a low incidence of severe malaria even under holoendemic epidemiological conditions. It is not clear if these results can be generalized. The artificial infection experiments at Horton Hospital in England yielded different results, suggesting that P. vivax is unable to prevent subsequent infections with P. falciparum, although P. falciparum does inhibit P. vivax.¹²² In any case this particular idea does not seem relevant to European historical populations, such as Grosseto and Croton (see below), where both P. vivax and P. falciparum were undoubtedly present yet the demographic effects of malaria as a whole on the human population were very severe. Mathematical modelling of the interaction of P. falciparum and P. vivax in the human bloodstream yields complicated results depending on the relative timing of the various infections. An existing P. vivax infection can reduce the parasite load in the blood of a subsequent P. falciparum infection by as much as 50%, but on the other hand, a subsequent P. vivax infection or even relapse can actually exacerbate an earlier low-level, asymptomatic, P. falciparum infection. In ¹²¹ Gill (1938: 36–9) noted that in northern Italy P. vivax infections in autumn tended to produce acute attacks immediately after the normal incubation period. In other words there was no delay of the primary attack until the spring as in Holland, for example. Consequently the spring wave of P. vivax malaria in northern Mediterranean countries was probably mainly composed of relapses, not primary attacks. The hotter the year, the more likely it was that P. falciparum would overshadow P. vivax and P. malariae in the late summer and autumn in Mediterranean countries. Sorgoni (1832) noted that pernicious symptoms were more frequent around Narni the greater the difference between the daytime and the night-time temperatures.
¹²² Maitland et al. (1997); contrast Covel and Nicol (1951) and Shute (1951).
156
Demography of malaria
the case of simultaneous infections with the two different species, P. vivax often fares poorly.¹²³ It is difficult to generalize about the demographic consequences at the population level of the P. vivax–P. falciparum interaction because of the complicated results found in individuals. Nevertheless the fact remains that the evidence from more recent periods of European history discussed in this chapter suggests that being infected with P. vivax in the past was not necessarily any better in the long run than being infected with P. falciparum. As far as antiquity is concerned, Galen’s comments in the second century indicate that P. falciparum infections tended to occur at an early age then (Ch. 8 below). This suggests a different epidemiology from that described recently in Vanuatu. This Pacific island group seems to lack the full range of genotypes of P.
falciparum found in Africa and Eurasia, as is suggested by a greatly reduced range of polymorphism in the merozoite surface protein genes.¹²⁴ Presumably a large proportion of the organism’s range of genetic polymorphism failed to make it across the ocean during the human population movements which colonized the Pacific islands.
Alternatively they might have been eliminated when the human population of Vanuatu passed through a severe bottleneck within the last two hundred years. Either way, the evolution of malaria in Vanuatu, an example of evolution in an isolated island population, is not directly relevant to the historical situation in Europe, although it is certainly of considerable intrinsic interest.
In passing, it should be noted that P. vivax probably already existed and operated in the way described by Dobson in Britain in classical antiquity. The disease of the marshes that severely affected the army of Septimius Severus in Scotland in 208 may well have been P. vivax malaria, as Bordier suggested a long time ago.¹²⁵ In the seventeenth century Doni expressed the opinion that quartan fever was unknown in Scotland, implying that he believed that tertian fever did occur there.¹²⁶ An inscription, now lost, from Risingham ¹²³ Mason and McKenzie (1999). Bruce et al. (2000) suggested that in tropical regions where malaria is continuously active there is density-dependent regulation of infections that transcends species as well as genotype, resulting in non-independent sequential episodes of infection with each species.
¹²⁴ Maitland et al. (2000).
¹²⁵ Bordier, cited by Fraccaro (1919: 86 n. 2), drew attention to Cassius Dio 77.13.2: ËpÏ
t0n Ëd3twn dein0ß ƒkakoınto (they were badly affected by the waters). Herodian 3.14.6–8
emphasized the marshiness of Britain in the early third century .
¹²⁶ Doni (1667: 7): in Scotia quartanam febrim ignotam esse credi potest. It will be remembered that the British mosquito vector A. atroparvus is a poor carrier of P. malariae (Shute (1951) ), although it could have been transmitted by A. plumbeus instead.
Demography of malaria
157
(Roman Habitancum), a Roman fort north of Hadrian’s Wall, has been frequently cited as a dedication to the goddess of tertian fever.
Unfortunately this reading of the text is a guess made in the early seventeenth century. It appears to be unreliable.¹²⁷ Dobson noted that the opium poppy was widely cultivated in East Anglia in the early modern period to provide opium, which was used to relieve the symptoms of P. vivax malaria. It did not create addiction under those circumstances. Archaeobotanical finds of opium poppy ( Papaver somniferum) from Iron Age–Roman archaeological sites in East Anglia show that it was already being cultivated there by Roman times, probably for the same reason. Localized extreme variation in mortality patterns caused by malaria was probably already occurring in Roman Britain.¹²⁸
Although there is no direct evidence available, owing to the shortage of written sources for Roman Britain, the balance of probability is that the constant movements of soldiers, merchants, slaves, and administrators between Britain and other parts of the Roman Empire introduced malaria to Roman Britain, if it was not already present. As was seen earlier, the evidence of Gregory of Tours shows that malaria was frequent and familiar in France in the sixth century (Ch. 4. 1 above). Moreover Pliny mentions tertian fever in the territory of the Tungri in Belgium and Holland in the first century .¹²⁹ P. vivax malaria was present in these areas by then, only a short distance from Britain. Alcuin in the late eighth century is a specific case of an individual who travelled from Britain to Rome, became infected with ‘Roman fever’, and brought the parasites back to northern Europe with him. He contracted malaria during his visit to Rome in 798 and was severely affected by it thereafter. Alcuin presumably had mixed infections, ¹²⁷ The text Deae Tertianae sacrum Ael(ia) ( CIL 7.999) was accepted by Weinstock in Pauly-Wissowa RE s.v. Tertiana (vol. v A.1 (1934), column 822), Schaffner in Cancik and Schneider (1998), s.v. Febris, and Burke (1996: col. 455, p. 2269). However, it was rejected in favour of the alternative reading ( Deae Dianae sacrum Aelia Timo posuit votum solvens laeta libens merito) found in the manuscript tradition by Collingwood and Wright (1995: 397 no. 1209), who rejected the lectio difficilior. The inscription CIL 12.3129 is a dedication to quartan fever
in the third century from Nemausus in Gallia Narbonensis ( quartanae votum reddet libens merito Byrria Sev-erilla).
¹²⁸ Dobson (1997: 304–6); Pryor (1991: 130); Cameron (1993: 10, 54–5) described malaria as common in Anglo-Saxon England; the lencten-ádl of Old English texts probably signified spring relapses of P. vivax malaria, according to Bonser (1963: 403–5); Darby (1983: 52, 95, 107, 112, 146, 150–2, 176) on malaria in the East Anglian fens and its disappearance in the nineteenth century; Nicholls (2000).
¹²⁹ Pliny, NH 31.8.12.
158
Demography of malaria
since he emphasizes that he had quotidian fevers. The timing of one episode of fever, after Easter 801, sounds like a typical spring relapse of vivax malaria, but it is possible that he contracted falciparum malaria as well in Rome.¹³⁰ There were undoubtedly other such individuals, since Bede and Paulus Diaconus both commented on the popularity among the Anglo-Saxons of pilgrimages to Rome in the early medieval period.¹³¹ More than one Anglo-Saxon king died in Rome. Bede related the story of the miracle of a young boy who was cured of a long-running intermittent fever by standing on the tomb of St. Oswald, king of Northumbria (633–642), at Bardney Abbey in Lincolnshire. This text constitutes some direct evidence for malaria in England c. 700 (at that time in the same monastery a little boy was severely affected by a prolonged fever. One day he was awaiting the hour of the attack,when one of the brothers entered and said to him: ‘my son, shall I teach you how to cure yourself from the trouble of this debilitating disease? Get up, enter the church, and go to Oswald’s tomb, remain there, stay quiet and remain by the tomb. Make sure that you don’t leave the church, or move from the spot, until the hour when your fever is scheduled to leave you. Then I will enter the church, and lead you away.’ The boy followed the brother’s instructions. While he sat by the tomb of the saint the disease never dared to touch him; indeed it was so terrified that it flew away, and did not dare to touch him on the second day, or the third day, or ever again.’)¹³²
¹³⁰ e.g. febrium flagellatione . . . remanet cotidianus labor eiusdem castigationis; tamen febrium castigatio cotidianis diebus nos non reliquit (By the lashing of fevers . . . the daily labour of the very same chastisement continues; the daily chastisement of fevers does not leave me.) Alcuin, Epistolae 218, 221, ed. Duemmler (1895), Monumenta Germaniae Historica. Epistolae, iv. 362, 365, cf. Epistolae 146, febris et infirmitas me fatigatum habet (fever and weakness keep me tired), and 149 ( febric-itantem); Gaskoin (1904: 99, 110, and 124). Peter of Blois is another example of a person who returned to northern Europe after contracting malaria in Italy. He developed semitertian fever in Sicily in 1169, was treated at Salerno, and then returned home with a detailed knowledge of malaria to the Loire valley region of France, where in a letter to his friend Peter medicus written c. 1170–5 he described another case of semitertian fever in the knight Geldewin, which may also have been imported—Peter of Blois, Letter 43, discussed by Holmes and Weedon (1962).
¹³¹ Bede, Historia Ecclesiastica, 5.7, ed. Plummer (1896); Paulus Diaconus 6.37.
¹³² Bede, HE iii.12: tempore fuit in eodem monasterio puerulus quidam longo febrium incommodo graviter vexatus: qui cum die quodam sollicitus horam accessionis exspectaret, ingressus ad eum quidam de fratribus: ‘Vis,’ inquit, ‘mi nate, doceam te quomodo cureris ab huius molestia languoris? Surge, ingredere ecclesiam, et accedens ad sepulcrum Osualdi, ibi reside, et quietus manens adhaere tumbae. Vide ne exeas inde, nec de loco movearis, donec hora recessionis febrium transierit. Tunc ipse intrabo, et educam te inde.’ Fecit ut ille suaserat, sedentemque ad tumbam sancti, infirmitas tangere nequaquam praesumpsit; quin in tantum timens aufugit, ut nec secunda die, nec tertia, neque umquam exinde eum auderet contingere.
Demography of malaria
159
In the late medieval period there continued to be regular contact between England and Rome. The resident English community in Rome suffered from malaria there as well as the numerous pilgrims, the so-called Romipetae. Gervase, a monk in Canterbury, described a severe epidemic of ‘bad air’ in Rome in 1188. It badly affected both visitors from England and the Roman cardinals who were patrons of the Canterbury church. It started its ravages in July, but continued to claim victims as late as October in that particular year. (‘For a terrible epidemic, arising from the excessive heat of summer and atmospheric conditions which appeared after the festival of St John the Baptist, devastated the Roman people to such an extent, and especially foreign visitors, that several thousand of the clerics and people died. Indeed the cardinals who were the patrons of the Canterbury church were killed by this abomin-able pestilence, as well as five monks who were companions of the prior and many of the servants. All the others were so ill that no one was able to pass on a drink of cold water to anyone else. However the prior Honorius, who was already on the point of death, was taken, through the good offices of the bishop of Ostia, to a mountainous location in the province of Velletri so that he could be cured by breathing purer air. Nevertheless that execrable poison of bad air, now occupying his vital organs, finally killed him on 21
October . . . Other monks, who had died in July, were buried in various churches.’)¹³³ A noted English doctor, Hugh of Evesham, was elevated to the status of cardinal of San Lorenzo in Lucina by Pope Martin IV (1281–5) to protect him from malaria, but Hugh himself died from Roman fever.¹³⁴ In view of these contacts it is ¹³³ The historical works of Gervase of Canterbury, ed. Stubbs (1879) i.: The chronicle of the reigns of Stephen, Henry II and Richard I, 42: Horrida enim pestilentia ex ardore aestatis nimio, et diversis aeris pas-sionibus quae post festum Sancti Johannis Baptistae emerserunt, Romanum adeo vastavit populum et maxime peregrinos, ut non nulla milia cleri et populi spiritus exhalerent. Ex hac pestilentia detestanda cardinales quidem Cantuariensis ecclesiae patroni extincti sunt, et ex sociis prioris monachi quinque et plurimi servientes.
Caeteri omnes adversa valitudine adeo detenti sunt, ut nec unus alteri vel aquam frigidam valeret propinare.
Prior autem Honorius iam fere praemortuus, beneficio episcopi Hostiensis in montana Velletrensis provinciae absportatus est, ut ibidem liberiori refectus aere respiraret. Sed illa corrupti aeris detestanda infectio, iam ipsius occupans vitalia, ad extrema perductum XIIo Kalendas Novembris compulit exspirare . . . Alii vero monachi, qui mense Julio mortui sunt, in diversis ecclesiis sepulti quiescunt.
¹³⁴ Brentano (1974: 50, 89). For the history of medieval pilgrimage to Rome and malaria see Birch (2000: esp. 56–8), citing numerous sources, e.g. Peter the Venerable, abbot of Cluny, Letter 118, in The Letters of Peter the Venerable, ed. Constable (1967) ( = Harvard Historical Studies, 78) i. 311: mortem ipsam, quam Romanus aer nostratibus celeriter inferre solet (death itself, which the Roman air is accustomed to bring rapidly to our colleagues). Peter suffered from malaria himself and made several other references to the unhealthiness of the Roman air in his letters written in the twelfth century (Constable, (1967) ii. 247–51).
160
Demography of malaria
likely that P. vivax malaria in Britain was regularly replenished and refuelled directly from Rome during the Roman and medieval periods, in exactly the same way that during the First World War the return of infected British soldiers from Greece led to a resurg-ence of malaria in the English marshlands. Rome exported diseases.
Dobson’s demographic data from the Kent and Essex marshlands can be directly compared to the data from Grosseto studied by del Panta to illustrate the profound deviations in the age-structure of mortality produced by malaria. Del Panta compared the population of Grosseto to Coale and Demeny Model South Level 2 (for males) because of the similarity in infant mortality in the first year of life.¹³⁵ He pointed out that adult mortality, especially in the age-group 20–50, was much higher in Grosseto relative to the (high) level of infant mortality than the model life-tables predict ( Table 3).
Table 3. Probability of death (qx) at various ages (in %
) Interval
Grosseto
Treppio
South 2
South 1
East 1
1q0
31.7
19.6
31.1
33.6
50.5
5q1
34.0
16.5
31.6
34.7
24.7
50q20
60.0
26.5
43.1
46.0
46.3
e 0
20.0
37.0
22.3
19.9
1 7.4
The data in Table 3 show that the effect of malaria on the population of Grosseto was to produce a much higher level of adult mortality between the ages of 20 and 50 than even the ‘worst’
model life-tables used by demographers (and Coale and Demeny Model East Level 1 is a theoretical construct). In plain language, conditions in Grosseto were so bad that adult mortality went right off the bottom end of the scale generally used by demographers.
This is the full magnitude of Varro’s ‘reckoning with death’, ratio cum orco (see Ch. 9 below). Historians who have attempted to minimize the role of malaria in Italian history have completely failed to appreciate the magnitude of the ‘reckoning with death’. Del Panta, a leading Italian historical demographer, stated that numerous places in Italy, especially in the Mezzogiorno, had demographic patterns similar to those of Grosseto.¹³⁶ One example from the ¹³⁵ Del Panta (1989: 22).
¹³⁶ Del Panta (1989: 23).
Demography of malaria
161
Mezzogiorno is the territory of the Crotonese. Arlacchi described the excess adult mortality: [sc. in the early twentieth century] ‘for every 100 deaths in the Crotonese 15 befell persons between the ages of 20 and 40, compared to 7 to 8 for Calabria and 6 to 7 for Italy as a whole’.¹³⁷ According to Bonelli 12.3% of all deaths in the Crotonese in 1882 were directly attributed to malaria. This was substantially less than the direct mortality from malaria at Grosseto in the same year. Nevertheless Arlacchi’s description shows that malaria had severe effects on the entire population of the Crotonese. This once again demonstrates that the overall effects of malaria stretch far beyond the proportion of deaths directly attributed to it. In the nineteenth century the crude death rate reached 60 per 1,000, while as recently as 1890 life expectancy at birth in the Crotonese was no higher than 20. It is likely that this was the fate of the once prosperous populations of the great coastal cities of Magna Graecia, such as Croton and Metapontum, during the Hellenistic and Roman periods following the spread of malaria.¹³⁸
Malaria and Rome: A History of Malaria in Ancient Italy Page 23
