Return to Wine

Lead Poisoning and Rome

"The bellows are burned, the lead is consumed of the fire; the founder melteth in vain: for the wicked are not plucked away. Reprobate silver shall men call them, because the Lord hath rejected them."

Jeremiah, 6:29-30

A 1983 article in the New England Journal of Medicine by Jerome Nriagu, a geochemist, reopened a debate that had been dormant for almost two decades. There, and in a book published later that year, he argued that "lead poisoning contributed to the decline of the Roman empire." Yet, a review by Scarborough, a pharmacist and classicist, criticized the book as being "so full of false evidence, miscitations, typographical errors, and a blatant flippancy regarding primary sources that the reader cannot trust the basic arguments." Scarborough concluded that, although ancient authorities were aware of lead poisoning, it was not endemic in the Roman empire nor caused its fall. More than thirty years later, the criticism still rankling, Nriagu retorted in an interview that "Scarborough knows nothing, absolutely nothing, about lead poisoning, Absolutely zero."

Lead (Pb) does not occur in an elemental state but is a by-product of silver mining. Extracted from galena ore (PbS, lead sulfide), which is crushed and smelted, the lead alloy was further refined by the Romans in a furnace made hotter still by blasts of air forced from a bellows (Pliny, Natural History, XXXIII.159). The oxidized lead (litharge, PbO), which was contained in a porous crucible of crushed bone ash, was absorbed, leaving behind a trace amount of silver in a process called "cupellation" (from the cupel used to collect the metal). The lead itself then was recovered by smelting the litharge again with galena, the lead oxide combining with lead sulfide to form metallic lead and sulfur dioxide (2PbO + PbS = 3Pb + SO2).

Readily abundant (in Britannia, the ore was so near the surface that restrictions limited the amount that could be produced, Pliny, XXXIV.164), easily malleable, and with a low melting point (low enough, in fact, to melt in a camp fire), lead (plumbum) was ideal for the production of water pipes, which were fabricated by plumbarii (plumbers) from fitted rolled sheets in a variety of diameters (Vitruvius, On Architecture, VIII.6.1ff; Frontinus, On the Aqueducts of Rome, XXXVIIff).

Sheets of lead were used to line Roman aqueducts (as was cement) and lead pipes to convey water. But lead also was known to be unwholesome and, for that reason, pipes made of clay were preferredas Vitruvius, who wrote during the time of Augustus, explains.

"Water conducted through earthen pipes is more wholesome than that through lead; indeed that conveyed in lead must be injurious, because from it white lead [ceruse or lead carbonate, PbCO3] is obtained, and this is said to be injurious to the human system. Hence, if what is generated from it is pernicious, there can be no doubt that itself cannot be a wholesome body. This may be verified by observing the workers in lead, who are of a pallid colour; for in casting lead, the fumes from it fixing on the different members, and daily burning them, destroy the vigour of the blood; water should therefore on no account be conducted in leaden pipes if we are desirous that it should be wholesome. That the flavour of that conveyed in earthen pipes is better, is shewn at our daily meals, for all those whose tables are furnished with silver vessels, nevertheless use those made of earth, from the purity of the flavour being preserved in them" (VIII.6.10-11).

Columella, too, remarks on the advantage of terracotta pipes. "Rain-water is after all most suitable to the body's health, and is regarded as uncommonly good if it is conveyed through earthen pipes into a covered cistern" (I.5.2).

Rain water was known to have the least amount of contamination (Celsus, On Medicine, II.18.12). But it also is slightly acidic and dissolves carbon dioxide in the atmosphere to form a weak solution of carbonic acid, which in turn reacts with calcium hydroxide to form calcium carbonate (CaCO3). In his treatise on the aqueducts of Rome, Frontinus complains that "the accumulation of deposit, which sometimes hardens into a crust, contracts the channel of the water" (CXXII.1). The aqueduct at Nmes, for example, had an accretion of calcium carbonate that accumulated approximately one millimeter per year (about four inches every century), constricting its channel by more than a third of its width. The frequent cleaning of this limestone encrustation (sinter, from the German) suggests that deposits of calcium carbonate in pipes and aqueduct channels protected against corrosion and insulated against the introduction of lead. Too, the water would have flowed continuously and not been in prolonged contact with the metal.

Rather than encrusted lead pipes, a more probable cause of chronic lead poisoning (plumbism or "saturnism" because its symptoms seemed indicative of the planet's characteristics) was the consumption of defrutum and sapa. Cato, Columella, Pliny, and Palladius (On Agriculture, XI.18) all describe how unfermented grape juice (mustum, must) was boiled to concentrate its natural sugars. ("Mustard" is derived from the word because must was added to the ground seeds to make a paste.) "A product of art, not of nature," says Pliny (XIV.80), the must was reduced to one half (defrutum) or even one third its volume (sapa), and the thickened syrup used to sweeten and preserve wine and fruit that otherwise was sour or would spoil. Cato, for example, recommends that olives and pears be preserved in boiled must (VII.4) as does Varro (I.59.3). And Columella indicates that defrutum should always be boiled with quinces or some other flavoring (XII.20.2). Apicius, in De Re Coquinaria offers directions for preserving quinces in defrutum and honey (I.21) and added the rich syrup to many of his sauces to enhance the color and flavor of almost every dish, whether meat, fish, fowl, or fruit. (The fact that the reduction was used to color food indicates that red wine was used rather than white.)

In De Agri Cultura, the earliest example of Latin prose (c.160 BC), Cato gives directions for reducing must in "a copper or lead vessel" over a slow fire, "stirring constantly to prevent scorching; continue the boiling, until you have boiled off a half" (CVII). Writing in the first century AD, Columella elaborates on the process.

"Some people put the must in leaden vessels and by boiling reduce it by a quarter, others by a third. There is no doubt that anyone who boiled it down to one-half would be likely to make a better thick form of must and therefore more profitable for use....But, before the must is poured into the boiling-vessels, it will be well that those which are made of lead should be coated inside with good oil and be well-rubbed, and that then the must should be put in....The vessels themselves in which the thickened and boiled-down must is boiled should be of lead rather than of brass; for, in the boiling, brazen vessels throw off copper rust, and spoil the flavour of the preservative....Must of the sweetest possible flavour will be boiled down to a third of its original volume and when boiled down, as I have said above, is called defrutum" (XII.19.1, 19.6, 20.1, 21.1; notice that the definition is not always consistent).

Pliny, too, recommends that the must be prepared in lead vessels.

"Also boiled-down must and must of new wine should be boiled when there is no moon, which means at the conjunction of that planet, and not on any other day; and moreover leaden and not copper jars should be used, and some walnuts should be thrown into the liquor, for those are said to absorb the smoke" (XIV.136).

It would seem therefore that must was boiled in cauldrons of lead, although Scarborough is reluctant to weaken his case, insisting that "one needs to read these texts carefully which mention a 'preference' for lead over bronze to realize that the Romans most often used bronze cauldrons (copper and tin in alloy), not those of lead" and that the short boiling time would not have contaminated the juice in any event. But copper and bronze are suspect as well. Not only, says Pliny, was the best bronze alloyed with ten percent lead and tin (XXXIV.95) but "When copper vessels are coated with stagnum [a lead alloy], the contents have a more agreeable taste and the formation of destructive verdigris is prevented" (XXXIV.160).

The bitter salt about which Pliny complains is verdigris or copper acetate, which first was described by Theophrastus (De Re Metallica, IX), a student of Aristotle, and is formed when acetic acid (the component that gives vinegar its sour taste and pungent smell) reacts with copper oxide. It was manufactured by exposing copper plates to strong vinegar or the soured grape skins and lees of wine dregs (Dioscorides, Materia Medica, V.91). Acetic acid also reacts with lead oxide (litharge) to create another metallic salt: lead acetate or sugar of lead. Although as sweet as sugar (which was unknown to the Romans), it did not add much to the sweetness of the defrutum, which derived from the concentrated glucose and fructose of the grapes themselves. It did import, however, an onerous burden of lead.

Eisinger found that must reduced to one-third its volume contained approximately one gram of lead per liter. If, as Columella recommends (XII.20.3), one sextarius of defrutum should be mixed with one amphora of wine, the resulting proportion would be one part in forty-eight or almost 21 milligrams per liter (mg/L). It is an extraordinary burden and certainly would induced symptoms of lead poisoningeven more so, if one follows Cato's recommendation of one part in thirty (XXIV).

Although there is no basis for such a quantification, Nriagu assumes the aristocracy of Rome to have consumed two liters of wine a dayalmost three bottles, which would seem to make alcoholism more suspect than lead poisoningand estimates the lead intake to have averaged 180 micrograms (g) daily. Although the relationship between ingested lead and blood lead levels varies, and there is no way to quantify what Romans actually ate or drank, he further estimates the total amount of lead absorbed in the blood from all sources to be 50 micrograms per deciliter (g/dL) or 0.5 mg/L. Three deciliters of such wine, the equivalent of two glasses, would therefore contain 150 g of lead. Wine poured from Columella's amphora would contain 21,000 g/L or 2,100 g/dL.

Such levels have significant physiological consequences. As reported by the U.S. Department of Health and Human Services, there is brain and kidney damage in adults with blood lead levels of 100 g/dL; gastrointestinal symptoms such as colic, with levels of approximately 60 g/dL; anemia with levels of 50 to 80 g/dL; neurological symptoms with levels of 40 to 60 g/dL; depressed sperm count with levels of 40 to 50 g/dL; and increased risk of preterm delivery, low birth weight, and impaired mental development with maternal blood lead levels of 10 to 15 g/dL. The physiological insult to children would have been even greater. In children five years or younger. Since 2012, the Centers for Disease Control and Prevention has recommended intervention when the level of lead in the blood is above 5 g/dL and has considered lowering that threshold to 3.5 g/dL.

Nriagu's contention that "a large number of Roman aristocrats ingested more than enough lead with their foods and drinks each day to put them at risk for lead poisoning" may be correct. But to infer that the impact on fertility was one of the principle causes for the decline of the Roman empire is not. Twenty years before, Gilfillan had insisted that "lead poisoning is to be reckoned the major influence in the ruin of the Roman culture, progressiveness, and genius," a thesis convincingly refuted by Needleman and Needleman, who demonstrate that the decline of the Roman aristocracy can as easily be explained by a simple desire not to marry or to rear few or no children. Indeed, as Waldon remarked, just as the decline of Rome was not unique, so "it is perhaps too facile to suppose it to have a unique cause."

Augustus sought to promote marriage and encourage procreation by legislation (the Julian laws of 18/17 BC and the Lex Papia Poppaea a generation later in AD 9). "And yet, marriages and the rearing of children did not become more frequent, so powerful were the attractions of a childless state" (Tacitus, Annals, III.25). In a review of the life span of emperors and aristocrats, Scheidel dismisses any impact of lead ingestion on fertility. "Nor is there any need to suspect that the incidence of marital sterility in the Roman ruling class might have been much higher than in other groups, times, and places." And Drasch found that the average lead burden in Rome was not significantly higher than in the legionary camp and provincial capital at Augsburg. In Britain, the skeletal lead burden was even higher than in Rome. Still, Nriagu insists that "one of the principle, probable causes of the internal weaknesses" of the Roman empire was lead poisoning of the aristocracy.

Certainly, the Romans knew lead to be dangerous, even if they did not associate it with their lead cooking vessels. Pliny speaks of the "noxious and deadly vapour" (sulfur dioxide) of the lead furnace (XXXIV.167; there was a four-fold increase in atmospheric Pb pollution during the Greco-Roman period); white lead (cerussa) as a deadly poison (XXXIV.176), even though it was widely used as a medicine and cosmetic; and the power of sapa (and onion) to induce an abortion (XXIII.30). Dioscorides cautions against taking white lead internally, as it is deadly (Material Medica, V.103). Soranus in his Gynecology (I.19.61) recommends that the mouth of the uterus be smeared with white lead to prevent conception. Galen (De Antidotis, XIV.144) and Celsus (V.27.12b) both provide an antidote for poisoning by white lead, and Vitruvius remarks on the pernicious effects of water found near lead mines and its effect on the body (VIII.3.5, 6.11).

The earliest description of acute lead poisoning (mid-second century BC) is given in the Alexipharmaca of Nicander, who speaks of "gleaming, deadly white lead whose fresh colour is like milk which foams all over" (II.74ff). The poet describes a frothing mouth, asperity of the tongue, and dry throat, together with dry retching, chills, delusions, and overwhelming fatigue. But if lead poisoning had been endemic, it presumably would have been remarked upon at the time. And yet there is no mention of the fact until early in the seventh century AD, when Paul of Aegina, a Byzantine physician, described chronic lead poisoning (although he does not associate its symptoms with the disease). "I am of the opinion that the colic affection which now prevails is occasioned by such humours; the disease having taken its rise in the country of Italy, but raging also in many other regions of the Roman empire, like a pestilential contagion, which in many cases terminates in epilepsy, but in others in paralysis of the extremities, while the sensibility of them is preserved, and sometimes both these afflictions attacking together" (III.64).

It should be remembered that the Romans diluted their wine. Of the two liters that Nriagu estimates to have been drunk every day, if mixed with two parts water, only two-thirds of a liter actually would have been undiluted wine. An inscription from AD 153 indicates that junior members of the College of Asclepius and Hygia were allotted two sextarii of wine a day, and senior members, nine sextarii, the equivalent of approximately one to five liters. And Cato allotted his slaves seven quadrantals (amphorae) of wine a year, allocated according to the month (On Agriculture, LVII). During the summer months, the allowance was a sextarius (about half a liter) per day, with more for the Saturnalia and, a few days later, the Compitalia.

Smaller vessels have a larger surface area relative to their volume. The four-liter pot used by Patterson et al. to determine the amount of dissolved lead presumably would leach out more of the metal than the huge oiled cauldron mentioned by Columella (XII.20.3), in which ninety amphorae (approximately 2340 liters) of must remained after boiling.

Defrutum was only one of several remedies to sweeten or preserve potentially sour wine (Pliny, XIV.121). Sea water and resin also were added, as was sugar of lead. Martial accuses a wine merchant of Marseilles of shipping poisonous and overpriced wines to his friends and, indeed, being reluctant to visit Rome for fear of having to drink them himself (Epigrams, X.36). Pliny, too, complains that "genuine, unadulterated wine is not to be had now, not even by the nobility" (XXIII.1), ruefully remarking "So many poisons are employed to force wine to suit our tasteand we are surprised that it is not wholesome!" (XIV.130). Indeed, "So low has our commercial honesty sank that only the names of vintages are sold, the wines being adulterated as soon as they are poured into the vats. Accordingly, strange though it may seem, the more common the wine is today, the freer it is from impurities" (XXIII.34).

Columella regarded "as the best wine any kind which can keep without any preservative, nor should anything at all be mixed with it by which its natural savour would be obscured; for that wine is most excellent which has given pleasure by its own natural quality" (XII.19.2). Further, "care must be taken that the flavour of the preservative is not noticeable, for that drives away the purchaser" (XII.20.7).

Writing several decades later in the first century AD, Dioscorides says much the same thing. "Generally, all unmixed and simple wine (hard by nature) is warming, easily digested and good for the stomach. It encourages the appetite, is nourishing, induces sleep, and causes a good colour" (Material Medica, V.11). Those with sapa, however, "fill the head causing drunkenness." One should drink moderately and (because alcohol is a diuretic and dehydrates the body) drink water both before and afterwards, "for it brings some help in avoiding illness owing to drunkenness."

The best tasting wine likely would not have been adulterated and presumably was the very wine drunk by the nobility, who supposedly were most at risk for lead poisoning. More than wine or water transported through lead pipes, the dainties and elaborate sauces prepared with defrutum by gourmands such as Apicius are likely to have been the primary source of ingested lead by the Roman aristocracy.

"What I tell you three times is true."

Lewis Carroll, The Hunting of the Snark (I.8)

A cautionary note: One reads that Hippocrates (died c. 370 BC) was the first to describe lead colic in a metal worker. Although a primary source almost never is provided, the Encyclopaedia Britannica Online does say, in its entry on Occupational Disease: The Preindustrial Era, that "The first recorded observation of an occupational disease may be a case of severe lead colic suffered by a worker who extracted metals. It is described in the third book of Epidemics, attributed to Hippocrates."

In fact, the reference is from Epidemics IV.25, where there is brief mention of a man "from the mines" who, among his other symptoms, was pale. And, indeed, miners often were described as pallidus. Vitruvius remarks on the pallid color of lead workers. Lucan (Pharsalia, IV.298), Silius Italicus (Punica, I.233), and Statius (Silvae, IV.7.15) all comment on the pallor of gold miners, especially those from Asturias in northern Spain, the mining operations which Pliny may have witnessed for himself (XXXIII.66ff). But it is facile to assume that, because Romans reported symptoms concomitant with lead poisoning, they necessarily were caused by leador, to phrase it another way, that if lead poisoning has certain symptom, the same symptom can be attributed to it. While Hippocrates may have known about lead poisoning, Waldron cautions that he "did not describe it in any of the books which have come down to us."

"But the Goths kept pressing vigorously upon them, shooting many missiles at the battlements, and they were already about to set their ladders against the wall, having practically surrounded those who were fighting from the tomb; for whenever the Goths advanced they always got in the rear of the Romans on both flanks; and for a short time consternation fell upon the Romans, who knew not what means of defence they should employ to save themselves, but afterwards by common agreement they broke in pieces the most of the statues, which were very large, and taking up great numbers of stones thus secured, threw them with both hands down upon the heads of the enemy, who gave way before this shower of missiles."

Procopius, The Gothic War (V.22)

Procopius is relating the Siege of Rome by the Goths in AD 537. The Barberini Faun (above) was discovered early in the seventeenth century at Hadrian's Tomb (Castel Sant'Angelo) and may have been one of the pieces thrown from the wall. Named for the cardinal who owned it early in the seventeenth century, it now is in the Glyptothek (Munich), which was constructed to house the sculpture collection of Ludwig I of Bavaria, who had a room designed specifically for his prize statue.

References: "Saturnine Gout Among Roman Aristocrats: Did Lead Poisoning Contribute to the Fall of the Empire?" (1983) by Jerome O. Nriagu, New England Journal of Medicine, 308, 660-663; Lead and Lead Poisoning in Antiquity (1983) by Jerome O. Nriagu; "Occupational Exposure to Lead in Ancient Times" (1983) by Jerome O. Nriagu, The Science of the Total Environment, 31(2), 105-116; "The Myth of Lead Poisoning Among the Romans: An Essay Review" (1984) by John Scarborough, Journal of the History of Medicine, 39, 469-475; "Lead Poisoning and the Fall of Rome" (1965) by S. C. Gilfillan, Journal of Occupational Medicine, 7, 53-60; "Lead and Wine: Eberhard Gockel and the Colica Pictonum" (1982) by Josef Eisinger, Medical History, 26, 279-302; "Lead in Ancient Human Bones and Its Relevance to Historical Developments of Social Problems with Lead" (1987) by C. C. Patterson, H. Shirahata, and J. E. Ericson, Science of the Total Environment, 61, 167-200; "Lead Poisoning in the Ancient World" (1973) by H. A. Waldron, Medical History, 17, 391-399; "Vitruvius, Lead Pipes and Lead Poisoning" (1981) by A. Treavor Hodge, American Journal of Archaeology, 85, 486-491; "Lead Poisoning and the Decline of the Roman Aristocracy" (1985) by Lionel and Diane Needleman, Classical Views, 4(1), 63-94; Lead Compounds: Hazard Summary (1992, revised 2000) by the U.S. Environmental Protection Agency, Technology Transfer Network, Air Toxics Website; "Greenland Ice Evidence of Hemispheric Lead Pollution Two Millennia Ago by Greek and Roman Civilizations" (1994) by Sungmin Hong, Jean-Pierre Candelone, Clair C. Patterson, and Claude F. Boutron, Science, 265, 1841-1843; "Hippocrates and Lead" (1973) by H. A. Waldron, in The Lancet, 2(7829), 626; "Emperors, Aristocrats, and the Grim Reaper: Towards a Demographic Profile of the Roman Elite" (1999) by Walter Scheidel, The Classical Quarterly, 49(1), 254-281; Roman Aqueducts & Water Supply (2002) by A. Trevor Hodge; "'Gleaming, white and deadly': Using Lead to Track Human Exposure and Geographic Origins in the Roman Period in Britain" (2010) by Janet Montgomery, Jane Evans, Simon Chenery, Vanessa Pashley, and Kristina Killgrove, Journal of Roman Archaeology, (Suppl. 78: Roman Diasporas), 199-226; "Lead Poisoning in Ancient Rome" (2006) by F. P. Retief and L. Cilliers, Acta Theologica, 26(2), Suppl. 7, 147-164; "Lead Burden in Prehistorical, Historical and Modern Human Bodies" (1982) by Gustave A. Drasch, The Science of the Total Environment, 24(3), 199-231; "Exposure to Lead in Ancient Populations" (1979) by Tony Waldron and Calvin Wells, Transactions and Studies of the College of Physicians of Philadelphia, 1(2), 102-115; "A Perspective of Lead Poisoning in Antiquity and the Present" (1984) by D. E. Woolley, Neurotoxicology, 5(3), 353-361; "Lead Poisoning and the Fall of Rome" by Lenny Bernstein, Washington Post, February 17, 2016.

Vitruvius: Ten Books on Architecture (2001) edited by Ingrid D. Rowland and Thomas Noble Howe; The Architecture of Marcus Vitruvius Pollio (1826) translated by Joseph Gwilt; Lucius Junius Moderatus Columella: On Agriculture (1954) translated by E. S. Forster and Edward H. Heffner (Loeb Classical Library); Marcus Porcius Cato: On Agriculture and Marcus Terentius Varro: On Agriculture (1935) translated by William Davis Hooper, revised by Harrison Boyd Ash; Pliny: Natural History (1945) translated by H. Rackham (Loeb Classical Library); Frontius: Stratagems, Aquducts (1925) translated by Charles E. Bennett (Loeb Classical Library); The Seven Books of Paulus Aegineta (1844) translated by Francis Adams; Celsus: On Medicine (1938) translated by W. G. Spencer (Loeb Classical Library); Apicius: De Re Coquinaria (1936/1977) translated by Joseph Vehling (Dover Books); Soranus' Gynecology (1956) translated by Owsei Temkin; Hippocrates: Epidemics I & III (1923) translated by W. H. S. Jones (Loeb Classical Library); The Fourteen Books by Palladius Rutilius Taurus milianus on Agriculture (1807) translated by T. Owen; Procopius: The Gothic War (1919) translated by H. B. Dewing (Loeb Classical Library); Dioscorides Pedanius of Anazarbus: De Materia Medica (2005) translated by Lily Y. Beck; Tacitus: The Annals of Imperial Rome (1959) translated by Michael Grant (Penguin Classics) Hippocrates: Epidemics (Vol. VII) (1994) translated by Wesley D. Smith (Loeb Classical Library).

See also Lead and the Roman Harbor of Portus.

Return to Top of Page