Was ancient wine more alcoholic than modern wine?

Emlyn Dodd


Ancient wines were considerably more alcoholic than modern wine, and that is why they were watered down in Graeco-Roman cultures.




Wine permeated Greek and Roman culture. It was used in religion, medicine, and cooking, and played a crucial role in trade and the economy. It was, however, perhaps most embedded across everyday life – drunk in wine bars and taverns, at home in domestic contexts, or at the infamous Greek symposion (similar to the Roman convivium, though with notable points of difference).

It is in these contexts that we most often hear of serving watered down wine; a practice ostensibly to maintain appropriately sober, civilised and intellectual behaviour, thus avoiding association with certain barbarians who drunk wine neat.

This raises two questions. First, were ancient wines significantly stronger than modern wines? And second, were these wines then watered down to lower the alcohol content?

Ancient yeasts and viniculture

While there is a popularising trend for modern wines to be made using “ancient” methods, the vast majority use specifically curated yeasts and inoculation processes for a specified end product. The most favourable type of yeast that produces reliable and consistently good wine is Saccharomyces cerevisiae. This leads to a product that, on average today, has between 10-14% abv (alcohol by volume).

Ancient wines, by comparison, could not be so selective with the type of yeast used. They were restricted to using wild strains, blowing around the vineyard and found on the skins of grapes. Wild yeasts often struggle to continue converting sugars to alcohol at about 6% and, all things going well, secondary yeasts then kick in to push abv higher.

But even if an ancient Greek or Roman was lucky enough to have S. cerevisiae floating around their vineyard and end up in their must, as well as excellent external and internal vintage conditions (and some luck), it is still difficult to achieve much more than 15% abv. At best, they might be able to push it up to 18% with specialist knowledge, care and pre-treated grapes (e.g. using a late harvest or dried grapes to increase sugar content).

However, here it must also be remembered that ancient winemakers did not have knowledge of modern stabilising processes, like adding sulphur dioxide – without these, relatively alcoholic wine would have low natural acidity and very high pH. It would taste horrible and spoil easily.

Death by alcohol?

So then, what of Pliny’s reference to Roman Falernian wine lighting with a flame (Natural History 14.8)? Or Plutarch’s description of Alexander the Great’s wine drinking contest where more than 40 men died from excessive consumption (Life of Alexander 70.1)?

Lighting wine with a flame would require around 40% abv – something unachievable through fermentation and requiring distillation. Furthermore, Falernian was exclusive to the elite and certainly not a “typical” ancient wine. To Plutarch’s comment – at best, a wide array of variables unrelated to the alcoholic content of wine consumed could be at play, and, at worst, perhaps some literary license over a 300-year time lapse.

It seems quite clear, from the highly variable and unreliable nature of ancient viniculture as well as stratified market demands, that a broad spectrum of alcohol levels existed within Graeco-Roman wines. From something we would consider a strong dry wine to one much weaker, with very little alcohol, for the lower classes. (Ancient sources that discuss a range of wines in antiquity, include Hippocrates, Pliny the Elder, Columella, Palladius, Galen, the Geoponika, as well as a range of Egyptian papyri.)

That is to say, ancient wines were almost certainly not more alcoholic than wine consumed today.

Many reasons to water down

We know there were ambitions amongst the elite to avoid drunkenness and maintain some sort of intellectual civility. Yet there is equal evidence across social strata of excess and little care for sobriety (see, for example, Martial’s Epigrams 11.6). This makes the habit of watering down wine much more interesting. Something tied closely to a combination of cultural tradition, fashionable taste and societal preference, sanitisation, and nutrition (wine was a key source of calories in antiquity).

The mixing of water and wine in ancient Graeco-Roman culture(s) comes across as a sort of “gastronomic tradition” for citizens of all ages, including children (my thanks also to Dylan Thomas for this phrasing). Within the symposion we are also told it allowed discussion to be drawn out over a longer period of time (Aristotle, On Coming-to be and passing-away; Plutarch, Moralia; Xenophanes, PLG 4, fr. 4). In Greek (mostly Athenian) symposion contexts, wine was mixed in the communal krater and typically drunk following the meal. At the Roman convivium, however, there is much more evidence for the mixing of wine in individual’s cups to personal preference and, it seems, a greater emphasis on drinking throughout the meal (Dunbabin 1993).

Hesiod recommends hard-working farmers mix three parts spring water with one part wine and drink after a meal (Works & Days 596). Rather than primarily aiming to lower alcoholic content, this seems much more likely to serve the dual purpose of providing much-needed calories (in the form of alcohol and sugars from the wine) and sanitising water to provide safe hydration in the heat of the day.

Indeed, modern studies show that even low concentrations of wine in water prove effective at inactivating pathogens and damaging bacterial microorganisms (through alcohol, pH and organic acids). The study notes, with a 1/5 (wine/water) ratio, it took 24-48 hours to achieve 100% inactivation of Salmonella and E. coli – an unlikely scenario in antiquity, when wine and water were usually mixed immediately before drinking. Though increasing the ratio of wine/water would speed up this process.

These same studies indicate that water mixed with a low concentration of wine produced a more refreshing beverage than water alone. It is also clear that very small quantities of wine are detectable when added to water, perhaps positively influencing the taste of ancient water sources – though notions of ancient “taste” are inherently problematic. Taking into account the flavoured nature of many Graeco-Roman wines, often adding salt water, herbs and spices, resin, honey, capers and even cheese, this effect is only amplified.

Concluding remarks

Our literary sources make it clear that notions of “barbarianism” and civility played a role in the tradition of watering down wine, particularly amongst certain societal circles. Yet this is over-emphasised and a broader range of factors, likely more influential and relevant to the majority of society, must be acknowledged. It is clear that wine was watered down, but not because it was any more alcoholic (and inebriating) than what we drink today.

As a result, we rate this claim as false.

Related claims


  • J.P. Brun, Le vin et l’huile dans la Médeiterranée antique: Viticulture, oléiculture et procédés de fabrication (2003).
  • E. Dodd, Roman and Late Antique wine production in the eastern Mediterranean (2020).
  • E. Dodd, “Pompeii is famous for its ruins and bodies, but what about its wine?The Conversation (5 November 2020).
  • K. Dunbabin, “Wine and water at the Roman convivium”, Journal of Roman Archaeology 6 (1993), pp. 116-141.
  • P. McGovern, Ancient Wine: The Search for the Origins of Viniculture (2003).
  • O. Murray, “The culture of the symposion”, in: K. Raaflaub and H. Van Wees (eds), A Companion to Archaic Greece (2009), pp. 508-523.
  • K. Raff, “The Roman banquet”, The Met Museum: Essays (October 2011).