A very deep history of Halloween

Or, how far back can historical analysis take us?

You know the holiday: the one where people wear outlandish costumes and sweet things are eaten. It’s fun, but also otherworldly, with roots in an ancient belief that this evening — this one night at the change of the seasons — is when spirits roam the earth.

I’m referring, of course, to qāšoq-zani, celebrated in Iran on the eve of Chaharshanbe Suri, the beginning of the Persian New Year festivities.

And also Halloween.

And Día de Muertos.

The days of the calendar differ. But it intrigues me that there is a wide swathe of the world which celebrates a holiday marked by sweets, blurred boundaries between the living and the dead, votive candles or bonfires, and costumes.

Halloween, Día de Muertos, and Chaharshanbe Suri

How much can we really know about things like this — cultural patterns which might share a common origin in the distant past, but which have no documented historical connection?

One of my favorite historians, Carlo Ginzburg, is a specialist in this very question.¹ Much of Ginzburg’s work explores what you might call the history of liminal states — moments when the perceived divisions between the ordinary world and the spirit world (or between rule and misrule, or consciousness and unconsciousness) fall apart.

Ginzburg’s first book, The Night Battles (1966), dove into one specific example: the benandanti (“good walkers”) of the area around Renaissance Venice, a kind of agrarian secret society who believed that they battled witches in their sleep. At the end of the book, Ginzburg argues that these benandanti were likely a survival from a Neolithic fertility cult. This, in turn, may have had its origin in shamanistic practices from the Central Asian Steppe.

Is he right? There’s no smoking gun source. But Ginzburg is an extremely careful scholar, and he has spent decades amassing evidence. Personally, I find the evidence compelling. Familiar things are sometimes much older than we assumed.

For the rest of this post, I’ll be doing something in Ginsburg’s spirit — but more slapdash — as I do my best to follow the thread of Halloween back as far back as possible.

One thousand years?

Halloween has origins in Samhain, a major annual festival of the ancient Celts, which took place on November 1 and was distinguished by bonfires, feasting, and the belief that the spirits of the “Otherworld” roamed the earth on this night. By the ninth century CE, elements of Samhain had been Christianized via the Feast of All Saints, also known as All Hallow’s Day. Folk customs preserved pre-Christian practices on the evening before All Hallows’s Day (“All Hallow’s Eve”). These, it is claimed, developed into the rudiments of the modern holiday of Halloween in the early modern British Isles, especially Ireland.

Meanwhile, in 16th century Mexico, Aztec devotion to Mictecacihuatl, the Lady of the Dead, merged with the Feast of All Saints. This produced yet another variation on these customs: Día de Muertos.

So far, so good. But is this really the furthest back we can go?

The earliest reliable written reference that I could find is in a tenth century Irish epic poem which speaks of “Samhain, when the summer goes to rest.” A 12th-century Irish source records a week of feasting during this time, when “there would be nothing but meetings and games and amusements and entertainments and eating and feasting" (sounds fun!).² There is talk of kindling sacred fires, and of spirits and ghosts wandering the earth. But very little detail.

This is where archaeology lends a hand. Because it turns out that several passage tombs from Neolithic Ireland are oriented such that their entrances would’ve been lit by the sun on the exact day of Samhain. Here we have a fascinating example of two different types of historical evidence working in tandem to make a compelling case for a link across several millennia.

The Mound of the Hostages, a passage tomb from circa 3100 BCE in Ireland which is aligned with Samhain.

The archaeological evidence for this seems pretty robust, and at times is surprisingly detailed — for instance, this study theorizes that a specific standing stone in Scotland was a place where “local chieftains, titled after the [cuckoo] bird and ‘married’ to the local goddess of sovereignty that personified Venus, were tied to the stone and ritually sacrificed.” This, it is claimed, happened on Samhain “at eight-year intervals from the Bronze Age until the late Iron Age.”

That same study includes a fascinating detail:

The evidence suggests that as part of the ritual the victim was given a drugged drink which reduced him to a state of semi-consciousness; tied to a standing stone and despatched with a sacred weapon.

Now we’re getting somewhere. Because both of those things — the ritual drug and the ritual weapon — sound distinctly proto-Indo-European.

Meaning very, very old.

---

If you’ve read this far, please consider a paid subscription. Reader support is essential for allowing me to write Res Obscura — one of the few places online with ad-free, original longform historical writing. You’ll receive the same weekly posts as free subscribers, plus:

• Occasional special posts

• Reader Q&As

• Early access to works in progress

---

Ten thousand years?

Once we get back to proto-Indo-European religion, things invariably get weird (phrases like “ritual human-mare intercourse” are not uncommon). And, since we’re dealing with a society that existed eight thousand or more years ago, any claim we can make is deeply uncertain.

But claims can still be made. In her book Celtic Cosmology and the Otherworld, Sharon Paice MacLeod draws on comparative mythology, linguistic evidence, and archaeology to argue for an

Indo-European cosmology in which deities inhabited the bright sky realm, and human beings — those that “belonged to the earth” — lived below them. In Celtic cultures, at least, there appears to have been a shift; human beings lived in the realm above, and the gods lived below in the Underworld. This may be the result of changes in religious belief and practice which started in the late Bronze Age in Europe, in which the religious focus on the heavens (as evident from pre-Iron Age monuments and alignments) changed to focus on the lower world. All across Europe, offerings began to be made into bodies of water (and ostensibly, into the earth as well). It has been theorized that this took place due to deterioration in climate in the later Bronze Age.³

As an example of post-christian survivals of this cosmology, MacLeod cites the case of an Irish priest living in 8th century Germany — Virgil by name — who angered the Pope himself with his heretical preaching that “there was another world and other men beneath the earth.”

Back to a potential link between the Persian rites around Nowruz and the modern Halloween.

It at least seems possible that there was a set of proto-Indo-European beliefs and rituals involving an evening which briefly parted the veil between the human world and the world of the dead/spirits/deities (a word which, originally, meant “shining like the sun”). Among the ancient Iranians, this realm of the divine continued to be the sky. And, by extension, the ritual fires which moved from earth to the sky. It is notable that both Samhain and the festivities before Nowruz involve ritual bonfires.

At some point, as MacLeod writes, the Celtic branch of this ritual tradition shifted the location of the spirits/dead/gods from the sky to the earth or water (though still emphasizing links to the sky, via the passage tombs oriented toward the sun). This may be reflected in the archaeological record of the la Tené culture, famous for its votive offerings deposited in wetlands and bogs. In fact, it has been theorized that the famous “bog bodies” of proto-Celtic Europe were ritual human sacrifices that reflected this tradition — and, indeed, that some bog bodies may have been killed on or around the date of Samhain.

Is it coincidence, then, that the term jack-o’-lantern “comes from the phenomenon of strange lights flickering over peat bogs”?

Maybe so. But then again, maybe not. One thing I love about history is that some things really do go back ten thousand years — the myth of a dog that guards the gates of the underworld, for instance. We can’t always prove the link, but we can find little moments of connection across time that are more than coincidental.

I, for one, find these links beautiful. And also a little spooky.

Weekly links

• A mysterious bronze lion has sat at the top of a pillar in Venice’s St. Mark’s Square since at least the twelfth century. Is it actually Chinese? (Archaeology Magazine)

• “The Frankfurt Kitchen, as it was known—rational, unpretentious, and socially oriented—was conceived as one of the first steps toward building a better, more egalitarian world in the late 1920s.” (Museum of Modern Art blog)

• One of the best academic article titles I’ve ever encountered, found while researching this post: “Kings Dying on Tuesday” by G. F. Dalton, 1972.

1. Ginzburg’s best book is The Cheese and the Worms, which details the tragic story of Menocchio, a 16th century Italian miller and autodidact whose homespun theories of the nature of reality put him fatally at odds with the Inquisition. Highly recommended for anyone interested in unusual thinkers in history.

2. The quote is from Ronald Hutton’s interesting book The Stations of the Sun: A History of the Ritual Year in Britain (Oxford, 1996), pg. 361.

3. Sharon Paice MacLeod, Celtic Cosmology and the Otherworld: Mythic Origins, Sovereignty and Liminality (McFarland, 2018), pg 15. The same book also provides one potential explanation for the night-time focus of Halloween and its kindred holidays today. Samhain, summer's end, began at the moment of the sun’s setting (pg 49).