Plague of Athens Greece

13
MOUNT SINAI JOURNAL OF MEDICINE 76:456-467, 2009 456 The Plague of Athens: Epidemiology and Paleopathology Robert J. Littman, MLitl, PhD Department of Languages and Literatures of Europe and the Americas, University of Hawaii. Honolulu, HI OUTLINE PLAGUF OF ATHENS: THUCYDIDES' ACCOUNT THE PLAGUE OF ATHENS: ANALYSES IN THE 20TH CENTURY EL'IDEMLOLOGICAL INTERPRETATION Person Place Time Other Epidemiological Information Thucydides' Text Common Source Acquisition Person-to-Person Transmission Mathematical Modeling Discussion Respiratory Transmission Reservoir Transmission PALEOPATHOLOGY ABSTRACT In 430 BC, a plague .struck the city of Athens, which was then under siege by Sparta during the Pelo- ponnesian War (431-404 BC). In the next 3 years, most of the population was infected, and perhaps as many as 75,000 to 100,000 people. 25% of the city's population, died. The Athenian general and historian Thucydides left an eye-witness account of this plague and a detailed description to allow future generations to identify the disease should it break out again. Iiecau.se of the importance of Thucydides and Athens in Western history and culture, the Plague of Athens has taken a prominent position in the history of the West for the past 2500 years. Address Correspondence to: Robert J. Littman Department of Languages and Literatures of Europe and the Americas University of Hawaii Honolulu, HI Email: [email protected] Despite Thucydides' careful description, in the past 100 years, scholars and physicians have disagreed about the identification of the disease. Based on clinical symptoms, 2 diagnoses have dominated the modern literature on the Athenian plague: smallpox and typhus. New methodologies, including forensic anthropology, demography, epidemiology, and pale- opathogy, including DNA analysis, have shed new light on the problem. Mathematical modeling has allowed the examination of the infection and attack rates and the determination of how long it takes a disease to spread in a city and how long it remains endemic. The highly contagious epidemic exhibited a pustular rash, high fever, and diarrhea. Originating in Ethiopia, it spread throughout the Mediterranean. It spared no segment of the population, includ- ing the statesman Pericles. The epidemic broke in early May 430 Bc, with another wave in the sum- mer of 428 BC and in the winter of 427-426 BC, and lasted 4.5 to 5 years. Thucydides portrays a virgin soil epidemic with a high attack rate and an unvary- ing course in persons of different ages, sexes, and nationalities. The epidemiological analysis excludes common source diseases and most respiratory diseases. The plague can be limited to either a reservoir diseases (zoonotic or vector-borne) or one of the respiratory diseases associated with an unusual means of persistence, either environmental/fomite persistence or adaptation to indolent transmission among dispersed niral populations. The first category includes typhus, arboviral diseases, and plague, and the second category includes smallpox. Both measles and explosive streptococcal disease appear to be much less likely candidates. In 2001, a mass grave was discovered that belonged to tile plague years. Ancient micro- bial typhoid {Salmonella entérica serovar Typhi) DNA was extracted from 3 skeletons. Because typhoid was endemic in the Greek world, it is not the likely cause of this sudden epidemic. Mt Sinai J Med 76:456^67, 2009. © 2009 Mount Sinai School ofMedicine Keywords: epidemic, Plague of Athens, smallpox, typhus. Publi.shed online in Wiley InlerScience (www.interscience.wiley.com). DOLlÜ,1002/msj.20137 © 2009 Mount Sinai School of Medicine

description

Scholarly article about plague in Athens Greece

Transcript of Plague of Athens Greece

Page 1: Plague of Athens Greece

MOUNT SINAI JOURNAL OF MEDICINE 76:456-467, 2009 456

The Plague of Athens:Epidemiology and Paleopathology

Robert J. Littman, MLitl, PhD

Department of Languages and Literatures of Europe and the Americas,University of Hawaii. Honolulu, HI

OUTLINE

PLAGUF OF ATHENS: THUCYDIDES' ACCOUNT

THE PLAGUE OF ATHENS: ANALYSES IN THE 20TH CENTURY

EL'IDEMLOLOGICAL INTERPRETATION

PersonPlaceTimeOther Epidemiological InformationThucydides' TextCommon Source AcquisitionPerson-to-Person TransmissionMathematical ModelingDiscussionRespiratory TransmissionReservoir Transmission

PALEOPATHOLOGY

ABSTRACT

In 430 BC, a plague .struck the city of Athens, whichwas then under siege by Sparta during the Pelo-ponnesian War (431-404 BC). In the next 3 years,most of the population was infected, and perhapsas many as 75,000 to 100,000 people. 25% of thecity's population, died. The Athenian general andhistorian Thucydides left an eye-witness account ofthis plague and a detailed description to allow futuregenerations to identify the disease should it breakout again. Iiecau.se of the importance of Thucydidesand Athens in Western history and culture, thePlague of Athens has taken a prominent positionin the history of the West for the past 2500 years.

Address Correspondence to:

Robert J. LittmanDepartment of Languages andLiteratures of Europe and theAmericas University of Hawaii

Honolulu, HIEmail: [email protected]

Despite Thucydides' careful description, in the past100 years, scholars and physicians have disagreedabout the identification of the disease. Based onclinical symptoms, 2 diagnoses have dominated themodern literature on the Athenian plague: smallpoxand typhus. New methodologies, including forensicanthropology, demography, epidemiology, and pale-opathogy, including DNA analysis, have shed newlight on the problem. Mathematical modeling hasallowed the examination of the infection and attackrates and the determination of how long it takes adisease to spread in a city and how long it remainsendemic. The highly contagious epidemic exhibiteda pustular rash, high fever, and diarrhea. Originatingin Ethiopia, it spread throughout the Mediterranean.It spared no segment of the population, includ-ing the statesman Pericles. The epidemic broke inearly May 430 Bc, with another wave in the sum-mer of 428 BC and in the winter of 427-426 BC, andlasted 4.5 to 5 years. Thucydides portrays a virginsoil epidemic with a high attack rate and an unvary-ing course in persons of different ages, sexes, andnationalities.The epidemiological analysis excludes commonsource diseases and most respiratory diseases.The plague can be limited to either a reservoirdiseases (zoonotic or vector-borne) or one of therespiratory diseases associated with an unusualmeans of persistence, either environmental/fomitepersistence or adaptation to indolent transmissionamong dispersed niral populations. The first categoryincludes typhus, arboviral diseases, and plague, andthe second category includes smallpox. Both measlesand explosive streptococcal disease appear to bemuch less likely candidates.In 2001, a mass grave was discovered thatbelonged to tile plague years. Ancient micro-bial typhoid {Salmonella entérica serovar Typhi)DNA was extracted from 3 skeletons. Becausetyphoid was endemic in the Greek world, it isnot the likely cause of this sudden epidemic.Mt Sinai J Med 76:456^67, 2009. © 2009Mount Sinai School of Medicine

Keywords: epidemic, Plague of Athens, smallpox,typhus.

Publi.shed online in Wiley InlerScience (www.interscience.wiley.com).DOLlÜ,1002/msj.20137

© 2009 Mount Sinai School of Medicine

Page 2: Plague of Athens Greece

MOUNT SINAI JOURNAL OF MEDICINE 457

Health and disease have played an importantpart in human religion and history. Although ourconquest of disease has extended the modernlifespan to 78 years in the Western world versus 25 tcj35 years in the ancient world, we are still frightenedby and concerned with plagues. In the modernworld, emerging strains of viruses and bacteria,such as influenza and acquired immune deficiencysyndr<ime, cause much anxiety and millions of deathson a yearly basis. These continuing tragedies arereminiscent <5f past centuries in which a husband andwife endured the deaths of half their children frominfectious disease.^ In addition to the normal patternsof disease in childhood, epidemics would often strikeout of nowhere and carry on large percentages ofthe population. For example, a plague that occurredin the reign of Marcus Aurelius (AD I6I-I8OX whichwas described extensively by Galen, killed a tenth ofthe population of the Roman Empire.̂ •'' The bubonicplague stmck in Byzantium in the 6th century andkilled a substantial part of the city's population.This plague, named the Plague of Justinian afterthe sitting emperor, was extensively described ina contemporary account, tnodeled on the work ofThucydides, by the historian Procopius in History ofthe Wars (11 22-33)- The worst plague in Iiumanhistory was the Black Death, which wiped outnearly half the population of Western Europe inthe I4th century.''^ A contemporary literary accountof the effects on society' survives in Boccaccio'sDecameron. Another famous literary account of aplague occurs in Daniel Defoe's Joiimcil of the PlagueYear, a semifictional account of the bubonic plaguein London in the 17th century (1665).

Over the pa.st 2000 years, we have learned muchabout infectious diseases, the existence of microor-ganisms, their causes, their symptoms, and how theyare spread. Since its beginning. Western medicine hasconcentrated on the clinical aspects of di.sease. In the19th centur>\ epidemiology began to emerge as ascientific discipline, although its roots go all the wayback to Hippocrates in tlie 5th century HC/'"** Thisnew discipline has proved invaluable in learning thecauses of diseases, thereby aiding modern medicinein formulating cures.

Despite some overlap, clinical medicine and epi-demiology are distinct. Clinical medicine is the careand study of sick individuals, the examination ofsigns and symptoms present in sick persons, andattempts to treat the patient. Epidemiology has ofienbeen called the study of the distribution and deter-minants of disease in human populations-, it seeksto determine who is infected, when, where, andwhy. It que.stions whether a di.sea.se strikes theyoung, the old, the sexually active, or those who

smoke. Epidemiology, unlike clinical medicine, stud-ies groups of people, not individuals. It also studiesthe well because those who do not contract a diseasecan often provide as many clues as those who do. Amore tiiodern definition of epidemiology is the studyof the distribution and detertninants of health events,including diseases, in human populations.

intimately, of course, the true test of publichealth, of which epidemiology is a component, iswhether it can help us prevent disease. Arcaneknowledge and intellectual formulations are of littleuse when the litmus test of lives saved and healthimproved stands visibly before us. Epidemiology hasbeen less than successful in some of its applica-tions. There is and perhaps always has been atendency for epidemiology to become intellectual-ized, in part because of its complex and philosophicalunderpinnings.

However, consideration of the epidemiologicalapproach with respect to the Athenian epidemic willhave, I hope, some modest by-products. Atiiong theseare lessons in the approach to public health problemsolving that can be applied to many new diseasesthat exist today and will be encountered in the future.What is needed is not only more technical knowledgeabout agents, hosts, and environments but alsoperspectives on and formulae for understanding thecomplex ways in which they interact, structures forassembling and synthesizing disparate facts, andapproaches for deriving conclusions. Although thiswill not, in and of itself, create a healthier world, andthe derived conclusions may not motivate humanbeings to change the behavioral risk components

Although modem epidemiologistsand physicians are mostconcerned with current disease,nonetheless it can also be valuableto examine the diseases of history,both to understand long-termsociological and demographicchanges and to better understandhow diseases work over centuries.

over which they theoretically have some control, itmay at least assist us in creating a framework and adatabase structure from which to proceed in makingpublic health decisions.

Although modern epidemiologists and physi-cians are most concerned with current disease,nonetheless it can also he valúatele to examine thediseases of history, both to understand long-term

DOl:10.1002/MSJ

Page 3: Plague of Athens Greece

458 R. J. LiTTMAN: THE PlAGHf- OF ATHENS

sociological and demographic changes and to bet-ter understand how diseases work over centuries,'̂

THE PLAGUE OF ATHENS:THUCYDIDES' ACCOUNT

One disease that has particulady attracted theattention of both laymen and scholars alike is thePlague of Athens. Although this was not the mostimportant plague in history in terms of mortalityand political and socioeconomic consequences, theplace of the 5th century BC Athenian hi.storian,Thucydides, as one of the fathers of hi.story has madehis description one of the best known passages inWestern literature and the history of medicine,

Thucydides presents a detailed description ofthe symptoms of the Plague of Athens that is markedby careful observation and is woven into a tersenarrative about the devastation of war and di.sease.In his History of the Peloponnesian War (2,47-55),Thucydides'" describes how the disease was accom-panied by a pustular rash, high fever, and diarrheaand how it usually resulted in the death of the victim.

In his History of thePeloponnesian War (2.47-55),Thucydides describes how thedisease was accompanied by apustular rash, high fever, anddiarrhea and how it usuallyresulted in the death of the victim.

He relates that it originated in Ethiopia and spreadinto Egypt and Libya and through the Near Eastbefore arriving at the Piraeus, the port of Athens,From the Piraeus, the epidemic moved rapidly intothe city itself Thucydides indicates that it was highlycontagious and infected anyone who cared for thesick. He himself suffered from the plague but wasone of the lucky ones who survived. Others were notso fonunate. Entire families were wiped out, and thedead became so numerous that corpses could not beburied and were left where they had died,

Thucydides (2,48.3) states tliat he will set downthe nature of the disease and explain the symptomsby which it may be recognized if it should ever breakout again. Fie was influenced by Hippocrates,"''^who lived at the same time, both in his descriptionof the plague and also in his application of Hippo-cratic doctrine to his writing of history. In a .sense, hisaccount of the disease parallels the overall themes

of his work: history as an example and a force thatrepeats and the decay of Athens by both war anddi.sea.se.

The political ramifications of the Plague ofAthens were enormous because it stRick Athensshortly after the Peloponnesian War (431-404 îîc)between Athens and Sparta had begun (430 Bc).Although Ailiens was devastated by the outbreak,Spaita was relatively unscathed because the epidemicdid not penetrate the Spartan homeland in thePeloponnesus. The Plague of Athens carried offperhaps 75,000 to 100,000 people, about 25% ofthe population of Athens during the first few years ofthe Peloponnesian War, The Athenian leader Periclesperisiicd of the disease in 429 BC. The plague wasa major reason for the defeat of Athens in thePeloponnesian War. It was economically and sociallydevastating to Athens, both at the time and in thesubsequent centuries. The combination of diseaseand war depopulated Athens and changed Greekhistory, which might have been very different hadAthens won the war.

THE PLAGUE OE ATHENS:ANALYSES IN THE 20TH CENTURY

Numerous essays on the Plague of Athens appearedin the 20th century, attempting to identify the Athe-nian pestilence with 1 or more diseases. BecauseThucydides provided the symptoms, the resultantstudies were symptomatologies, which were corre-lated with a disease that presumably matched theancient description. Two diagnoses became promi-nent in the modern literature on the Athenian plague.The first diagnosis was smallpox, which .seemed toincorporate almost all of the Thucydidean details.'-^

The second most common retrospective diag-nosis has been endemic typhus, as represented byHans Zin.s.ser,''* who was one of the pioneers inthe study of typhus and developed the first typhusvaccine in 1934, Two other frequently cited studiesfollowed Zins.ser's diagnosis, those of MacAithur''̂and Scarborough.''' Scarborough especially calledattention to Thucydides' description as accurate inits own context but incapable of accounting for anyevolutionary shifts in the causative organisms. Heargued that any modem diagnosis is flawed becauseof possible mutations over time. Although Scarbor-ough might be correct if the disease described byThucydides was caused by a microorganism that wasprone to mutation (as typhus is), other microorgan-isms, such as smallpox, are very stable over time.

In 1994, Thomas E, Morgan'̂ analyzed Thucy-dides' medical knowledge, or the lack of it. A

DOI:10.1002/MSJ

Page 4: Plague of Athens Greece

MOUNT SINAI JOURNAL OF MEDICINE 459

physician-classicist, Morgan argued that Thucydidesdesired to contrast the tragedy of war and the pathosof the disease with the lofty ideals presented by Per-icles in the funeral oration. Thus, in his opinion, thedescriptions of the plague are quite imprecise notonly because the Hippocratic or other tenninologiesvaried from those of modern epidemiology but also,more importantly, because Thucydides employeddramatic license in his account of the plague. Theweakness in Morgan's argument is that his thesis issomewhat inconsistent with Thucydides' statementthat he describes the symptoms so that the diseasecan be identified if it breaks out again.

Other possible candidates for the identificationof the Plague of Athens have ranged from measles,typhoid fever, bubonic plague, anthrax, and ergotismto toxic shock syndrome with Staphylococcus, scarletfever. Rift Valley fever, and arboviruses, but none ofthese possibilities has gained many adherents amongstudents of ancient epidemics.̂ **'̂ ^ A number of thesesuggestions are diseases that may not have existedin the ancient world hut appeared only in the 20thcentury. Lassa fever tl965) and Ebola fever (1976)are viral hemorrhagic fevers caused by RNA viruses.Rift Valley fever (1915) is a viral zotjnosis whosevector is the bite of infected mosquitoes, typicallythe Aedes or Culex genera. Tularemia (1911) is bac-terial with ticks as a vector. Each of these diseaseshas problems either in its symptomatology or in itsepidemiok)gy, as we will see later. For a full dis-cussion of the various suggestions, see Morens andLittman,̂ '̂ Poole and Holladay,-" Cunha,^' Sallares,^-and Bruce-Chwatt and de Zulueta.^''

Beginning in the second half of the 20th century,new methock)logies were developed that allowedfresh examinations of diseases, plagues, and pan-demics in the ancient world. These methodologieswere based on new academic disciplines, includingforensic anthropology, demography, and epidemiol-ogy. Recent studies of the Plague of Athens reflectthese developments in methodology. Although theemergence of modem medicine in the 19th centuryand the beginnings of the discipline of epidemiologyled to a renewed interest in the Plague of Athens,it was not until 1992 that the first epidemiologicalapproach to the disease was brought to bear byDr, David Morens, now at the National Institute ofHealth, and my-self'̂ '̂ "* using such devices as math-ematical modeling to look at various possibilities.Mathematical modeling allowed us to examine infec-tion and attack rates based on the various candidatesand to determine how long it takes a particular dis-ease to spread in a city and how long it wouldbe able to remain endemic. Several diseases, such asmeasles, had to be eliminated as possibilities because

the population of Athens was too small to sustainan epidemic of them for more than a few months.Thus, the best possibilities were narrowed to typhus,t>'phus-like diseases, and smallpox.

EPIDEMÍOLOGICAL INTERl^RETATION

PersonA number of prominent people contracted the Plagueof Athens. In the first outbreak, the historian Thucy-dides, who wrote a history' of this period entitledHistory of the Peloponnesian War, contracted the dis-ease. He survived and lived to write an account ofit. The great Athenian statesman Pericles was lessfortunate. He died in the first outbreak at the age ofapproximately 65, as did his sister, who was around60, and 2 of his children, who were around 30 (Xan-ttiippus) and 25 CParalos; see Plutarch, Life of Peri-cles, 36.3-4).^* The father of medicine, Hippocrates,according to legend, tried to combat the plague but inthe end ñcd Athens .so as not to contract it."'' Thucy-dides (2:51) states that the epidemic "carried awayall alike.'' Apparently, it spared no .segment of thepopulation. Thucydides relates that physicians hadincreased risk. The city of Athens was under siege,and the surrounding countryside was either burnedor .seized by the Spailans. Thucydides does not com-ment on the diet, although even with an unblockadedport, probably there were shortages of fruits andvegetables, with a possible concomitant deficiencyof ascorbic acid (vitamin C). Vitamin A would havebeen obtained through fish and fish products.

PlaceThucydides (2.48,1) relates that the plague originatedin Ethiopia and spread to Egypt and Libya and thenceto the territory of the king of Persia (all of theMiddle East and as far east as the Indus river). Fromthere, it reached Athens, which was the hardest hit.The Roman historian Lixy {History of Rome, 4.20-21,4.25.3-4, and 4.30.8-10), writing many centuries later,reported that epidemics occurred in Rome in 433and 438 BC. These outbreaks may well have beenpart of the same epidemic. The disease stmck first atAthens' port, the Piraeus, one of the major ports of theMediterranean. Premodern epidemics and pandemicsof many diseases (cholera, dengue, plague, andsmallpox) typically were spread by ships. AroundJuly 430 BC, at the height of the epidemic, theAthenian general Hagnon took a naval expeditionand sailed northward to Thrace with 4000 troops.The expedition was under sail for about 5 days, andby the time it had reached the city of Potidea, the

DOI:10.1002/MSI

Page 5: Plague of Athens Greece

460 R. J, LlTTMAN; T H E P L A G U E OE ATHENS

epidemic had struck. It killed about 25% of the armyand infected the Athenian besiegers of Potidea. Thecity of Potidea itself, sealed shut by the invaders, wasnot afïlicted by the epidemic.

Thucydides describes the directionality of spreadwhen the disease first attacked Athens. It began inthe port and spread inland into the area betweenthe port and the city, which was encompassed bya defensive wall, and into the city itself. Thucydidesis silent on the incidence rate (ie, number of newcases per population at risk per time), the high casefatality rate (proportion of those who died), andthe rapidity of epidemic progression. He does saythat the spread among the refugees in the city wascaused by "poor ventilation," which is consistent witheither airborne spread, poor hygiene and .sanitation,crowding, or a combination of these. The watersupply was probably uncontaminated. Potable waterin the Piraeus came from cisterns and in Athenscame from wells throughout the city. The city inaddition had river-fed aqueducts from the Ilissusriver and several springs. Thucydides mentions thatdespite frequent contact with the Spartan army,the disease did not spread much beyond the city.Thucydides (2,54.5) says that it did not enter thePeloponnesus to any extent and that its "full forcewas felt at Athens, and, after Athens, in the mo,stdensely populated of the other places," Perhaps thefact that the Peloponnesus was rather sparsely settledcontributed to a lower epidemic rate.

TimeThe epidemic broke out soon after the Spartansbesieged Athens in early May 430 BC, There wasanother wave in the summer of 428 BC and in thewinter of 427-426 BC. Apparently, the epidemic lastedaround 4.5 to 5 years. The population of Athens,^^'^^with reiiigees from the invading Spartan army,amounted to around 300,00 to 400,000. According toThucydides (3.87.3). the plague was unintermpted,but with exposing outbreaks;

The plague broke out among the Athenians for asecond time. In fact, it had never entirely .stopped . , .This second outbreak lasted for no less than a yearand the first outbreak for two years.

Other Epidemiological InformationThucydides portrays a virgin soil epidemic, whichThucydides confirms with his own words. It had ahigh attack rate and an unvarying course in personsof different ages, sexes, and nationalities. We can-not determine the incubation per- iod, attack rate,or case fatality rate. From July to August 430 BC,

Thucydides portrays a virgin soilepidemic . . . It had a high attackrate and an unvarying course inpersons of different ages, sexes,and nationalities.

26% of an expedition of 4000 hoplites were killedin 40 days. Thucydides (3.87,3) records a final deathtoll of 4400 hoplites (34%) and 300 cavalry deaths(30%). This suggests that the first wave was the worst.Survivors did not contract the disease a second time.This implies an attack rate of more than 25 and acase fatality rate of greater than 25%, Thucydides(2.50.1-2) also suggests ambiguously that there mighthave been a zoonotic component:

Though there were many dead bodies lying aroundunburied, the birds and animals that eat human flesheither did not come near them or, if they did tastethe flesh, died of it afterwards. Evidence for thismay be found in the fact that there was a completedisappearance of all birds of prey: they were not tobe seen either round the bodies or anywhere else.But dogs, being domestic animals, provided the bestopportunity of observing this eflect of the plague.

Thucydides did not claim that anyone saw aninfected dog or bird, but some scholars have usedthis to suggest epizootic involvement. However,Thucydides' description is so precise that if he sawan infected animal, he surely would have indicated it.

Thucydides' TextAlthough Thucydides proposed to tnake a descrip-tion that would allow others to identify the diseaseif it should break out again, he was not successful,although perhaps he was by ancient standards^^•• '̂̂Thucydides jumbled physical signs and symptoms,epidemiological ob,servations, and historical facts. Hewas not a physician, and a technical medical vocab-ulary probably did not exist in the 5th century BCwhen he was writing.-^" He thus applied commonwords to signs and .symptoms with the result that it isoften difficult to figure out precisely what he meant.For example, Thucydides used the term phlyktainai,a word to describe blisters and spots on a loaf ofbread. Scholars have translated this work with var-ious terms, including blains, blebs, blisters, buUae,eruptions pimples, pustules, vesicles, and whelks.Hippocrates used the word to describe burns andcontact dermatitis. Thus, questions of how to inter-pret the word have led scholars to develop theoriessuggesting smallpox, typhus, and syphilis. Tiying to

DOI:10.1002/MST

Page 6: Plague of Athens Greece

MOUNT SINAI JOURNAL OF MEDICINE 461

reach a definitive conclusion from the clinical descrip-tion alone has reached a limit of what it is able toachieve. Only when the clinical is melded with theepidemiological can more definitive conclusions bereached. Althougli the epidemiological approach can-not be definitive, it can clearly eliminate whole cate-gories of disease and some specific diseases, such asmeasles.

Common Source AcquisitionCommon source acquisition is not likely on the basisof Thucydides' description. Although ergotism hasbeen suggested, it is difficult to see how all grainsources could have been contaminated.

Waterborne disease is also unlikely becausethere was no centralized water system that couldhave become contaminated. Also, the disease spreadfrom the Piraeus uphill to Athens. There is no wayfor hundreds of Athenian wells to be simultaneouslycross-contaminated. Also, common source acquisi-tion of a single epidemic disease in 430-425 BC canbe ruled out, including such diseases as cholera,dysentery, ergotism, shigellosis, scurvy, and t>'phoidfever. However, recent DNA investigations of skele-tal remains from the plague years have suggested thepresence of typhoid. We will return to a discussionof these results later.

Person-to-Person Transmissionof the 3 main categories of person-to-person trans-mission, enteric and inoculation transmission can beruled out. Most consistent with the symptoms anddescription of the disease is aerosol/respiratory trans-mission, as suggested by the high attack rate andrapid spread. However, most respiratory diseasesrapidly die out in crowded populations, althoughsome, such as tuberculosis, can become endemic,Thucydides records that the epidemic lasted 2 yearsor more. In addition, the besieging Spartan amiy wasnot infected to any degree.

Mathematical ModelingBecause epidemic diseases vary in incubation peri-ods and in susceptible percentages of populations,the rate of disease spread varies. For example, if a dis-ease takes 2 days to become contagious and becomescapable of transmitting the disease to a third party,it might spread faster than an infectious disease thattakes 10 days to become contagious. Using mathe-matical models teased on these factors, we can predictfor any contagious disease mathematical patterns ofoccurrence over specified time periods for various

tJsing mathematical models basedon these factors, we can predict forany contagious diseasemathematical patterns ofoccurrence over specified timeperiods for various populationsizes and degrees of crowding.

population sizes and degrees of crowding. We canthen apply such theoretical patterns to the Athenianepidemic.

The equation used to generate the epidemiccurves is as

C,+i =S,(2~qCt')

where Q is the number of cases of the disease attime t (the beginning of the epidemic), C,+i is thenumber of cases of the disease at time t -\-\. t isthe chosen time interval (here the serial generationtime), S¡ is the number of susceptible persons at timet. and ^ = 1 — p is the probability of adequate con-tact (ie, adequate to cause infection) between any 2individuals per time interval t.

For each disease, the following assumptions aremade:

1. There is universal susceptibility at the outset withinfection either killing or conferring permanentimmunity.

2. The serial generation time is assumed to be4.5 days for influenza, 12 days for smallpox,14 days for measles, and 19 days for streptococcalinfections such as scarlet fever.

3. The total population of Athens was at lea.st 100,000,and most likely during the war years, the popu-lation climbed to 200,000 to 400,000 as refugeespoured Into the city.

4. Adequate contact numbers are based on conser-vative estimates of the frequency of contact underconditions of known severe crowding.

The area of the Piraeus, connected by wallsto the city of Athens, was about 4 square miles;this means a population density of 25,000 to 50,000per square mile. Thirty percent of the square milesof Athens proper, not including Attica, was proba-bly unoccupied.^^ This reflects a typical settlementpattern of cities with crowded populations; peo-ple are crammed into houses, rooms, and publicplaces, while at the same time there are large pock-ets of unoccupied spaces, There were probably about10.000 dwellings, each occupied by 10 to 40 personsduring the siege.̂ ^ Although the main outbreak of

DOl:10.1002/MSJ

Page 7: Plague of Athens Greece

462 R. J. LriTMAN: THE PI^GUF OF ATHENS

the plague may give some ambiguity about onset, wehave a clearer case with the expedition of Hagnon.The expedition would not have set out from thePiraeus with infected people on board. Among the4000 hoplites in the force and the 3000 hoplitesleft in Potidea, the epidemic curve from the indexcase to extinction was about 6 weeks. The results ofapplying the aforementioned formula can be .seen inFigures 1 to 3.

DiscussionEpidemiological analysis is consistent with ( 1) an ani-mal or insect reservoir or (2) respiratory transmissioncombined with a reservoir-like mechanistn of per-sistence. This is consistent with the fact that thebesieging army of Spartans was not infected.

Respiratory TransmissionDiseases transmitted only by a respiratory routewould not have "over-wintered'' and would havespread rapidly through the population and lastedover 2 years. Within a very short period, all thepopulation would have been exposed. For example,in 1918, influenza spread through and died out inNewark, NJ (population of 435,000) in 12 weeks.

Measles could not have persisted for 2 or moreyears: 300,000 to 500,000 people are considered tobe necessary to sustain measles transmission. Measlesv^ould have been extinguished in .several months.

Smallpox remains a strong possibility. In themathematical modeling (see Figures 1-3), a 12"dayserial generation time is assumed as well as 2 to 20adequate contacts and a minimal base population of100,(X)0 people per 4 square miles. Under this model,with 2 contacts, the epidemic would last 11 mtinths.Smallpox outbreaks in virgin populations tend to bebrief. For example, in Aztec Mexico in 1530, between3 5 and 15 million out of a population of 25 to30 million died within 6 months.• '̂* However, in thecase of smallpox, unlike other respiratory diseases,viruses in dried smallpox secretions can survive for atleast several months in clothes or bed linen.^^ Thereis even a case of smallpox being used for biologicalwarfare against the Indians in the French and IndianWar, when the commander of British forces gavethe Indians blankets contaminated with smallpox.•^''It is possible in Athens that smallpox was carriedhack to the countryside and dispersed in an indolenttransmission among the population of Attica in theremote areas in which the Spartan army was notpresent, although most of the population had comewithin the city walls.-̂ ^

Reservoir TransmissionThe epidemiology of the epidemic is consistentwith re.servoir transmission (insect or animal vector).Anthrax might be a po.ssibility because by a combina-tion of inhalation, inoculation, and ingestion, it couldbe consistent with the clinical picture described byThucydides. However, no large-scale anthrax epi-demics are known to have ever occurred. Also,there is no easily identified reservoir in ancientAthens. Although sheep and cattle are very sus-ceptible to anthrax, these animals were sent out ofAttica to Euboea. Dogs, rodents, and birds remainedin the city, but the.se are generally resistant toanthrax, although vultures and flies can spread thedisea.se after feasting on contaminated carca.s.ses. asis the case with dogs. Infected birds could contami-nate the water supply, but Thucydides mentions theabsence of birds. Anthrax is difficult to reconcile withthese conditions.

Other resei-voir diseases of epidemic proportioninclude various insect-borne diseases and diseasesspread by various zoonotic reservoir hosts, suchas malaria, plague, typhus, and various arboviraldiseases (eg, dengue, yellow fever, and Rift Valleyfever). All except Rift Valley fever are linked to war,refugees, and overcrowding. Dengue and Rift Valleyfever both have explosive behavior consistent withthe Plague of Athens. However, the clinical symptomsof both dengue and Rift Valley fever seem to differgreatly from Thucydides' description, especially inthe nature of the rash.

Malaria is unlikely as a cause, particularlybecause it was not an unknown disease,-^^ Althoughthe ancient worid did not distinguish betweentyphus and t>'phoid. it is clear that it was preva-lent in ancient Cîreece and Rome. Hippocrates in the5th century BC describes at least 6 cases in EpidemicsI and n.̂ 9

Of the reservoir diseases, typhus is the best fit.It is cla.ssically associated with war and crowding,and some symptoms described in Thucydides aresugge.sti\ e, such as gangrene (if in fact Thucydides isde.scribing gangrene) and blindness. Also, the diseaseis persistent in crowded populations for prolongedperiods. The main argument against typhus isthe pre.sence of bullae in the Athenian epidemic.Although not regular, they are seen occasionallyin typhus,'"* and their occurrence in a mild formof the disease, rickettsial pox, might suggest thatan ancestral rickettsial agent could have caused thePlague of Athens.

In sumniar>-. the epidemiological argumentexcludes all common source diseases and mostrespiratory disea.ses. By a process of elimination,the Plague of Athens can be limited to either a

DOT:I0.1002/MSJ

Page 8: Plague of Athens Greece

MOUNT SiNAt JOURNAL OF MEDICINE 463

10.000

1.000

lûû

10

Af/ /

1 1

i/ ••"

i /// / /i 1 ••

/ / / ' •

/ / / •'

/ / / ••

11 ••'/ / ' •'

/ / ' . ' •

1/ •'lu :A'/il;il;1;t

////

^ — — - =30 AdequMe C— = 30AilM|ii»leC

^,(20) •' ' .

\ , •' '• (IOJ

' ^ '•••l ^\ '> •••

\ \ '\ \\ \i \1 \\ \

\ \\ \\ \\ >l \\ 1

\ \

\ \

\ \

\ \\ \\ \\ \\ \\ \

ntacis

01 MHS

'•

'•

t\

\1i

per i : D.j.per 12 D.I.pt! 12 DBVS

ptt 12 Oip

\

\

',

;

'.

',

5 10 15 20 25 30 35 40 45 50 55 60 65 70DAYS AFTER EPIDEMIC ONSET

Fig 1, Theoretical epidemic curves of a transmissible respiratory disease witlia 12-day serial generation time (eg, smallpox) in Hagnon's naval expeditionto Potidea in 430 BC {a clcsed population of 4000 persons) according to thedeterministic mathematical model of Maia. Four separate curves have beencomputed with adequate contact numbers of 10, 20. 30, and 40 per 12 days.

By a process of elimination, thePlague of Athens can be limited toeither a reservoirdiseases (zoonotic or vector-borne)or one of the respiratory diseasesassociated with an unusualmeans of persistence,either environ men tal/fomitepersistence or adaptation toindolent transmission amongdispersed rural

populations. The first categoryincludes typhus, arboviraldiseases, and plague, and thesecond category includessmallpox. Both measles andexplosive streptococcal diseaseappear to be much less likelycandidates.

reservoir diseases (zoonotic or vector-borne) or oneof the respiratory diseases associated with an unusualmeans of persistence, either environmental/fomite

DOI:10.1002/MSJ

Page 9: Plague of Athens Greece

464 R. J. LiTTMAN: T H E P L A G I ' F OF ATHENS

100.000

iÙOOQ

1.000

10 20 30 « 50 60 70 80 90 100 110 120 130 140 150DAYS A F T E R E P I D E M I C O N S E T

Fig 2. Theoretical epidemic curves of influenza A, smallpox, measles,and streptococcal disease in a closed population of 100,000 susceptiblepersons according to the mathematical model of Maia. The curves assumean adequate contact number of 10.

persistence or adaptation to indolent transmi.ssionamong dispersed rural populations. The first categoryincludes typhus, arboviral diseases, and plague, andthe second category includes smallpox. Both measle.sand explosive streptococcal disease appear to bemuch less likely candidates.

PALEOPATHOLOGY

A challenge in historical epidemiology and the historyof medicine is working with written history with littleor no physical evidence. In some cases, particularlyancient Egypt, mummies provide actual physicalevidence, as do skeletal remains. The new science

of DNA analysis has opened new windows in theinvestigation of the past.

Beginning in the second half of the 20thcentury, new methodologies were developed thatallowed fresh examinations of diseases in history.These methodologies were based on new academicdisciplines, including forensic anthropology, demog-raphy, and epidemiology. Scientific discoveries andtechnology alloweci us new insights into the veryessence of life. Chief among these were the discov-ery of DNA and the ability to analyze it, magneticresonance imaging, and computed axial tomography(ie, three-dimensional pictures of hard and soft tissueinside the human body). These disciplines, discover-ies, and technology have been brought to bear on the

DOl:10.1002/MST

Page 10: Plague of Athens Greece

MOUNT SINAI JOURNAL OF MEDIONE 465

(JO.Ú0O

10.Ü00

1.000

100

10

0

(20

:l

¡il

il

•J /

v

MM'

• • i l; /•• í

• i

¡

ii

0)

/ 1• / 1

•i

í".;

/ ;

(5)

11

11

1

/

/

Ii

j

i

i

i

t

i\i\

\

\

\

\

\

_ ' 5 Adcquile Conlacts

• 20 Adujuile CDnlaciï

1

\

\

\

\

\

\

\

0 25 50 75 100 125 150 175 200 225 250 275 300 325 350 375D A Y S A F T E R E P I D E M I C O N S E T

Fig 3. Theoretical epidemic cundes of smallpox in a closed population of100,000 .susceptible persons according to the mathematical model of Maia.The cur\-es assume adequate contact numbers of 2, 5, 10. and 20 personsper serial generation time of 12 days.

study of di.seases in antiquity, including the natureand impact of plagues and pandemics in the ancientworld.

The study of ancient DNA has begun to providenew clues in studying plagues of antiquity. Ancientmicrobial DNA has been successfully extracted'*'̂ ;400-year-old DNA of Yersinia pestis, the microorgan-ism that causes bubonic plague, has been extractedfrom dental pulp,*^ The extraction of human DNA ismore problernatic, There are claims that NeanderthalDNA has been extracted, and other claims that itis not possible to extract DNA more than 800 yearsold. In any event, for both the Plague of Athens andother ancient diseases, microbial DNA might providea useful tool.

in 2001. a mass grave was discovered at thecemetery of the Kerameikos, which belonged to

the plague years. The excavation was conductedby Effie Baziotopoulou-Valavani,^^ a member ofthe Greek Archaeological Service. The study ofthe skeletal material was undertaken by ProfessorManolis Papagrigorakis''^ of the l!niversit>' of Athens.He was able to extract ancient microbial t>'phoid (S.entérica serovar Typhi) DNA from the remains. Heposited that the probable cause of the Plague ofAthens was this disease. It is premature, however,to draw this conclusion. We know from Hippocratesthat typhoid was most likely endemic in the Greekworld. The presence of an endemic disease does notnecessarily indicate that it was the cause of death, Asecond problem is that the sampling was extremelysmall: 3 skeletons. A third difficulty- is that the analysisof Papagrigorakis has been challenged, Shapiroet al.'^'^ argued that Papagrigorakis reported a 7%

DOl:10,1002/MSJ

Page 11: Plague of Athens Greece

466 R. 1. LiTTMAN: THE PLAGUE OF ATHENS

divergence between their sequences and S. entéricaserovar Typhi and therefore by simple phylogeneticanalysis the ancient sequence falls outside both S.entérica and S. typhimurium. Therefore, the obtained.sequences do not allow an identification of typhoidin these skeletons, although it is possibly a relatedSalmonella. Shapiro et al. further suggested that thesequence might have come from soil surrounding theburial and that the analysis is therefore compromised.Papagrigorakis et al.'*'' responded that soil wash wasused as a negative control and that there was nocontamination, a phylogenetic analysis of only 1 genemight be insufficient or misleading, and Sal mo i wildspecies do not survive for long in soil. A DNA analysisof the skeletal materials is an intriguing .start that onecould anticipate will lead to new information ai:)outpaleopathology and the plague years. Because theplague staick Egypt according to Thucydides, whatwe most hope for are Egyptian mummified remainsof someone who died of the Plague of Athens.

DISCLOSURES

Potential conßict of interest: Nothing to report.

REFERENCES

1. Bagnall R, Frier BW. The Demography of Roman Egypt.Cambridge, F.ngland; Cambridge University Press;2008.; Cambridge Studies in Population, Economy andSociety in Past Time.

2. Littman RJ. Litltimn ML. Gaien and the Antonine Plague.AmJPhiïot 1973; 94: 243-255.

3. Nutton V. Áncietít Medicine. London. England;Routledgtr; 2004.

4. Niittrtn V, ed. Pestilential Complexities: UnderstandingMedieval Plague. London, England; Wellcome TmstCentre for the History of Medicine at UCL 2008.

5. Hatcher J. Plague, Population, and the English Econ-omy 1348-1530. Basingstoke, England: Macmillan;1977.

6. Hippocrates f W. H. S. Jone.s trans. ) Hippocrates: Natureof Man rv (Loeb Classical IJhraiy). Boston: HarvardUniversity Press; 1931; 27-29.

7. Hippocrates {W.D. Smith trans.) Hippocrates:Epidemics Book 2, 4- 7 (Loeh Classical Library) Boston:Harvard University Press; 1994.

8. Hippocrate.s (W.H.S. Jones transj Hippocrates AncientMedicine - Airs. Wate/y and Places - Epidemics I and///Boston; Harvard University Press; 1984.

9. Ferngren GB. Medicine &amp: Heatth Care in HarfyChristianity. Baltimore, MD; Johns Hopkins UniversityPress; 2009: 13-16.

10. Thucydides. Thucydides: the History of thePeloponnesian War Vols I and II. CF Smith, trans.Boston, MA: Harvard University Press; 1919 and 1921:Loeh Classical Library.

11. Kechenaur G. Thukydides und die HippokratischeMedizin. Hilde.sheim, Germany; Georg Ohms;1991.

12. Weidauer K. V^ukydides und die HippokratischenSchriften. Heidelberg, Germany; Carl Winter: 1954.

13. Littman R|. Littman Ml.. The Athenian plague; smallpox,Trans Am Ph(tot Assoc 1969: 100: 261-275.

14. Zinsser H. Rats, Lice and History. Boston, MA: Linie,Brown and Company: 1934.

15. MacArthur WP. The Athenian plague: a medical note.Classical Q 1954; 4: 171-174.

16. Scarborough J. Thucydides, Greek medicine and theplague at Athens; a summary of possibilities. Episteme1970; 4: 77-90.

17. Morgan TE. Plague or poetry? Thucydides on theepidemical Athens. Trans Am Pbilol Assoc 1994; 124:197-209.

18. Longrigg ]. The great plague ai Athens, ¡list Sei 1980;18: 209-225.

19. Morens DM, Littman Rj. Thutydides' .syndromereconsidered; new thoughts on the "Plague of Athens.'Am/Epidemiol 1994: 140; 621-28.

20. Poole JCE, Holladay AJ. Thucydides and the Phigue ofAthens. Classical Q 19^9; 29; 282-300.

21. Cunha BA. The cause of the Plague of Athens; plague,typhoid, typhus, smallpox, or measles? Lnfect Dis ClinNorth Am 2004; 18; 29-43.

22. Sallares R. The Ecology of tbe Ancient Greek World.Ithaca, NY; Cornell University Press; 1991.

23. Plutarch (B. Perrin trans.) Phitarcb: Paratlel Lives, III;(Loeh Classical Library). Cambridge, Massachusetts:Harvard University Press: 1995.

24. Bruce-Chwatt LJ, de Zulueta J. 'Ihe Rise and Pallof Malaria in Europe: A Historico-EpiiiemiologicalStudy. Oxford, England; Oxford University Pre.ss;1980.

25. Morens DM, Littman RJ. Epidemiology of the Plague atAthens. Trans Am Philol Assoc 1992; 122; 271-304.

26. PinauhJR. IJip/xjcratic lives and Legends. Leiden, TheNetherlands: Brill; 1992; chap 2.

27. Han.sen MH. Athenian population losses 431-403B.C. and the number of Athenian citizens in 431B.C. In; Hansen MH, ed. Ihree Studies in AthenianDemography. Copenhagen, Denmark: Munksgaard;1988; Histonsk-PilosopfiskeMeddelelser- vol LVI.

28. Gomme AW. 'The Population of Athens in the Pifth andPourth CenturiesBC. Oxford, EnglantI: Basil Blackwell;19.33; 22-44.

29. Jouanna J. Hippocrates. DeBevoise MB, trans.Baltimore, MD: lohns Hopkins University Press: 1999;207-209.

30. Dumortier J. Le Vocabulaire Medical d'Eschyle et les&Eacute; crits Hippocratiqties. Paris, France: Les BellesLetters; 1975.

31. Costa Maia JDO. Some mathematical developments onthe epidemic theory formuiated by Reed and Frost.Hum Biol 1952; 24; 167-200.

32. Web.ster TBL. Athenian Culture and Society. Berkeley,CA: University ot California Pre.ss; 1973.

33. Siegel E. Epidemics and infectious diseases a[ the timeof Hippocrates. Their relation to modern accounts.Gesnerus I960; 17: 77-98.

34. Magner JA. Men of Mexico. Milwaukee. Wl; BaicePublishing; 1942.

35. Young M. Small pox conveyed by raw cotton. BrMed./1951: 25: 324-341.

DOI;10.K)02/MSI

Page 12: Plague of Athens Greece

T SINAI JOURNAL OF MEDICINE 467

36. Duffy I, Smallpox and the Indians in the Americancolonies. Bull Hist Med 1951; 25: 324- .^! .

57. Hopkins DR. Princes and Peasants: Smallpox inHistory. Chicago, IL: University of Chicago Press: 198.-Î;234-294.

38. Sallares R. Malaria and Greek Histoiy. Manchester,England: Manchester University Press; 2002.

39. Cunha BA. Osier on typhoid fever: differentiatingtyphoid from t>'phus and malaria. Infect Dis Clin NorthAm 2004• 18: 111-125.

40. Zink AR, Sola C. Rcischl U, et al. Characterizationof Mycohacterium tuberculosis complex DNAs fromEgyptian mummies by spoligotyping. / Clin Micmbiol2003; 41: 359-367.

41. Dnincourt M, Aboudharam G, Signoli M, et al.Detection of 400-year-old Yersinia peslis DNA inhuman dental pulp: an approach to the diagnosis ofancient septicemia. Proc Nati Acad Sei USA 1998: 95:12637-12640.

42. Baziotopoulou-Valavani E. A mas.s burial from thecemetery of Kerameikos. In Stamatopoulou M,Yeroulanou M, ed.s, Kxcavating Classical Culture:Recent Archaeological Discoveries in Greece. Oxford.England: Archat-opress; 2002; 187-201. Studies inClassical Archaeology: vol I.

43. Papagrigorakis MJ, Yapijakis C, Synodinos PN,Baziotopoulou-Valavani H. DNA examination ofancient dental pulp indicates typhoid fever as aprobable cause of the Plague of Athens, ¡nt J InfectD/5 2006; 10: 206-214.

44. Shapiro B, Rambaut A, Gilbert TP. No proof thattyphoid caused the Plague of Athens. Imj Infect Dis2006; 10: 3.M-335.

45. Papagrigorakis MJ, Yapijakis C, Synodinos PN,Baziotopoulou-Valavani E, Insufficient phylogeneticanalysis may not exclude candidacy of typhoid feveras a probable cause of the Plague of Athens. Int J InfectPis 2006; 10: 335-336.

DOI:10.1002/MSJ

Page 13: Plague of Athens Greece