Thunder Before the Plow

The bullroarer is a Palaeolithic instrument that reaches the Greek mysteries with roughly fifteen thousand years of documented depth behind it — and it entered the New World with a pre-agricultural grammar of weather-working and spirit-voice, set before farming existed. Farming, when it came, could recruit the instrument to the crop only in the Old World and only where gardens spread; in the Americas the same roarer was never re-aimed past rain-calling, because rain-calling was already in its inherited repertoire and crop-magic was not. The weather layer is old. The crop layer is a Neolithic overlay laid on top of it.

How old is the instrument?

Older than farming anywhere on earth. Three Ice Age horizons carry the instrument, each deeper than the last.

The Magdalenian is solid ground. The type-specimen is the ochred reindeer-antler blade from La Roche de Birol at Lalinde, which Denis Peyrony published in 1930 as a rhombe and Henri Breuil hailed as the first complete Magdalenian churinga — the Australian analogy that named the whole field. Behind the holotype stands a genuine corpus of about thirty-four bramaderas across Cantabrian Spain, the Pyrenees, and the Dordogne, catalogued by Ignacio Barandiáran Maestu: Laugerie-Basse, Bourrouilla (which yielded ten pieces alone), Lortet, Le Morin, El Pendo, Isturitz. Michel Dauvois settled the physics. A true rhombe, he showed, is thin — its thickness about a seventh of its width — so that as it circles the swinger it also spins on its own axis, and it is that double gyration, la double giration, that produces the instrument's characteristic drone. He built and swung experimental replicas; the Lalinde blade sounds. No user's account survives from the Ice Age — the ritual reading rests on shape, acoustics, and the instrument's worldwide behaviour — but the instrument reading itself is not in dispute.

Behind the Magdalenian, the trail runs deeper. The Solutrean is named in the literature, though its best-known piece, Badegoule, went into print as a rhomboidal bone pendant, and Dauvois's thickness rule prunes the look-alikes. The deep floor is Gravettian, and the claim is Dauvois's own. The much-repeated “oldest bullroarer, Ukraine, 17,000 BC” is a dead meme — it traces to a single uncited line and points at the Mezin mammoth-bone percussion set, not a roarer. The real claim is better: in western Europe the rhombes appear not much before the Solutrean, about 19,000 years ago, “alors qu'ils sont plus anciens en Europe Centrale, au Pavlovien (environ –24 000)” — older in central Europe, in the Pavlovian, around 24,000 BC, the culture of the Dolní Věstonice Venus and the first fired clay. Dauvois points to no single figured Pavlovian piece, and the reason is the instrument's oldest habit: a bullroarer parted from its cord is indistinguishable from an ornament, and the candidates hide in plain sight among the elongated, perforated ivory “pendants” of the Moravian sites.

Move east and some seven thousand years forward and the instrument re-appears, not obviously descended from the European corpus but strikingly convergent in form. Oliver Dietrich and Jens Notroff assembled thirteen elongated, one-end-perforated decorated bone “spatulae” from Pre-Pottery Neolithic sites and asked whether they are bullroarers, building a hardwood replica that hums. The most suggestive node is Körtik Tepe, where seven decorated plaques come from graves, perforated at the narrow end and carved with goats and, unmistakably, serpents — a single snake with a triangular head, a band of three snakes as hatched rhomboids. The earliest plausible Near-Eastern roarers are grave goods that carry serpent iconography, planting the serpent's-voice thesis in the Neolithic. At Göbekli Tepe three bone spatulae join them, one apparently depicting the site's own T-pillars; at Çatalhöyük an antler object is identified as a bullroarer in the site's own worked-bone report, complete with polish and cord-wear at the perforation. A second, wholly independent deep-time strand runs through Neolithic Northeast Asia, where end-pierced jade blades — some in large-and-small pairs at Chahai and Xinglongwa — show the same “end-pierced blade in a grave, function un-asked” signature; it is a parallel, not a link in the European chain.

Then the instrument does what a secret instrument does: it vanishes from the record without vanishing from the world. The Bronze Age and the Iron Age show no Minoan, Mycenaean, Cypriot, or pharaonic bullroarer from about 6,500 BCE to 330 BCE — and no one who knows the living instrument will find that strange. Everywhere it is documented, the roarer is the concealed thing: whirled at night, buried with the dead, kept from women and children on pain of death, hidden in the kiva between rites. Its everyday body is wood, which rots; the Ice Age pieces survive only because hunters cut them from bone and antler. An instrument whose whole office is to be heard and never seen is the last object a palace fresco or a temple relief would show. One phantom must be cleared from the bridge: the “Tutankhamun bullroarer” is a modern factoid, absent from the excavation records; every securely Egyptian roarer is Ptolemaic — Hans Hickmann's Schwirrhölzer from Mit Rahina, and the whirled instrument named in the Orphic Gurôb papyrus — Greek mystery material that happens to sit physically in Egypt. And when the record finally resumes — Demosthenes sneering, in 330 BC, at a rival's mother whirling the rhombos in a backstreet rite — the instrument surfaces exactly where a secret survivor should: not in the open religion of the temples but in the mysteries, the one corner of Greek life organized around initiation, concealment, and terror in the dark. The object in that woman's hands already had the Ice Age behind it.

What did the Ice Age instrument do?

Depth alone does not tell you function. But there is one place on earth where the function was, in effect, put to a controlled experiment — the Americas. The chronology is not in dispute. The founding populations crossed from Beringia by roughly sixteen thousand years ago; agriculture appears nowhere on Earth before about twelve thousand years ago; maize was domesticated in the Balsas valley of Mexico some nine thousand years ago, the Andean and Eastern-Woodlands complexes in the same Holocene window or later. Whatever the first Americans carried across, they carried it as foragers. Then, in isolation, three great farming civilizations arose independently — Mesoamerican maize, the Andes, the Mississippian Eastern Woodlands. If a crop-roarer were simply what farmers make because they farm, the Americas should be full of them.

They contain none. Read every row across the American farming macroregions and the functions divide cleanly: the roarer is a child's toy among the Lacandon and Mopan Maya, the Cherokee, the Catawba; an evil-spirit driver circled around a Pipil house at dusk; a Maleku diviner's instrument; a Ticuna girls'-puberty voice. Eastern North America, the maize heartland, is a near-blank where the instrument survives at all it survives as a toy. Nowhere in the hemisphere is there a “mother of the yams,” a sound that makes the plant grow. The absence is the argument: the instrument entered with a grammar of weather and spirit-voice, and even millennia of intensive American farming could not add the crop idiom to it, because that idiom was never in the inherited kit.

The near-exception proves the rule. In the arid Pueblo Southwest the roarer is recruited toward agriculture — but only as far as its inherited thunder-capacity allows. At Zuni it is whirled to summon the u'wannami rain-makers and gather the clouds. At Hopi the tovokìnpi is twirled by the kalektaka warrior who trails the Antelope priests in the Snake dance, imitating the thunder that accompanies the rain; John Bourke was told plainly that by making this sound they compelled the wind and rain to come to the aid of the crops. The Navajo whirl it to make a mimic rain-storm; among the Tewa of San Ildefonso it is kept out of sight in the kiva. This is a weather instrument aimed at corn — thunder-mimicry to call rain — never a sound that grows the plant directly. And the tell is next door.

The non-farming Apache use the identical roarer — tzi-ditindi, “the sounding wood,” cut from lightning-struck mountain timber and belonging to the wind god — with no crop attached at all. The thunder-and-wind-god function is the constant; the corn is the local add-on wherever people happen to farm. And the same instrument shows its pre-agricultural face most vividly where there was never any farming to recruit it: the Great Basin. Across the Numic-speaking Ute, Southern Paiute, and Shoshoni — foragers to a band — the roarer is the densest weather instrument in the entire catalog, and it has nothing to do with crops. The Moanunts and their neighbours make the wind blow and clear away the clouds; the Uncompahgre Ute stop the wind; the Surprise Valley Paiute whirl the kwi'mo to bring the warm south wind and melt the deep snow, and among the Harney Valley Paiute the wimo'to made the wind blow and “was no toy.” These peoples have no gardens. Their roarer is wind for hunting, warm wind to break the winter — not rain rationalized for a crop. Northward the forager weather-use runs on through the Subarctic, where Cree hunters sat up all night whirling it to bring the wind best for hunting, and it reaches across the strait to the Ainu rera-suyep, the “wind-swing,” sounded to raise a favourable wind for the deer. The two regions the peopling corridor joins — the Americas and Siberia — share one profile: weather-heavy, crop-empty. William Buckner, assembling the same forager wind-roarers, allowed cautiously that “we cannot necessarily rule out… diffusion across Beringia.” The catalog lets us say it harder: the Beringian instrument's function-set was weather-and-spirit-voice, and that set is exactly what its American descendants kept.

The Neolithic re-aiming

If the crop-roarer is a Neolithic overlay rather than a universal reflex of farming, it should appear precisely where farming spread out of the Old-World Neolithic — and it does. In the New Guinea lowlands every horticultural society with a flagged roarer uses it as a crop instrument. The Kiwai madubu is whirled so the gardens fill with yams, sweet potatoes, and bananas; in the garden rite a husband's and wife's mingled fluids are smeared on the blade so its swinging scatters the medicine across the field, and one roarer sent to England bore the label “the mother of the yams.” The Marind carry ritual semen home to spread through houses, plants, and bodies; the Bukawa and Yabim whirl it for the increase of taro while naming the dead; the Nara swing the beriwa over the gardens to make the crops grow. Where the roarer is agricultural here it is a growth instrument, pointed at the plant, not at the sky.

The same overlay lies over the African farming belt. Where the African roarer is agricultural it clusters in the Sudanic and Guinea cultures: the Uukwaluudhi omathila rite that joins rain to sowing, the Verre harvest-cycle held to be the direct source of good yields. And monsoon Asia keeps the field-guard variant — Naga and Bornean and Kassena boys whirling the roarer to scare birds and monkeys off the ripening rice, a sound that “could pass for a lion's roar.” Against that whole belt sits Australia, where there is no crop-roarer at all: the Australian instrument is initiation, ancestor-voice, and, in the southeast, rain-making. The Sahul interior is a natural experiment of its own, holding foragers and farmers side by side, and the function tracks subsistence exactly — Australia kept the pre-agricultural uses, New Guinea's horticulture grew the crop function on top.

The most telling places are the seams, where you can watch the recruitment happen. The Pueblo Southwest is one: farmers who did re-aim the roarer toward agriculture, yet only as rain-for-corn, the exact ceiling of an inherited thunder-function. Torres Strait is its Pacific mirror. Between farming New Guinea and foraging Australia, the gardener-fisher islanders show both functions on adjacent islands: Yam Island's bigu “belonged to sweet-potatoes and yams” and was whirled to make the crops grow, while the Meriam of Mer whirl the bigo in rain-making and the Moa islanders raise and quench the wind. The same transition, on both sides of the Pacific, at exactly the farming/foraging boundary.

Africa also preserves the two layers stacked in plain view. Beneath the farming-belt crop rows sits an older forager rain-layer, concentrated in the San and Khoisan south: the San rain-and-hunting roarer, and, in the rock art of the Cederberg, eight painted figures each swinging a bullroarer, its sound tied in San practice to rain-making and to !Khwa, the Rain, with a Later Stone Age specimen in the Matjes River layers to anchor it in the ground. And the only rites in the whole catalog that fuse weather and crop in a single ceremony are African — the Uukwaluudhi rite that is at once rain and sowing, the Gwana Dodo cult that works rain, sowing, and harvest together. That is exactly what thunder-before-plow should look like where the two strata meet: a forager rain-roarer underneath, farming-belt crop-roarers on top, and a visible seam where they fuse.

Europe keeps the thunder in the names

If weather is an ancient layer, it should leave fossils in the words where the cult has otherwise drained away — a storm stranded in a noun after it left the verb. It does. The clearest is Scots. In Aberdeenshire and Kincardineshire boys whirled a notched slat during thunderstorms in the belief that the thunder would cease or the bolt would spare them, and they called it the thunner-spell — the thunder-spell — a usage Walter Gregor set down first-hand in 1881. On the Croatian island of Krk, five wooden blades at Vrbnik carried names built on grmljavina, thunder — grmavica, germjavica — and were whirled overhead to make a thunder-like sound. At Villefranche-de-Lauragais the local rhombe survives with a museum notice recording only that it imitates the rumble of thunder. Across northern and Mediterranean Europe the demoted toy keeps the storm in its name centuries after the cult that gave it meaning was gone. The European vector is generally defensive — the instrument turned against the weather, to make the thunder stop — but the weather is what stayed in the word.

Behind the folk toy stands the sacred original, and it too roars. The Orphic-Dionysiac mysteries staged the murder of the child Dionysos-Zagreus, lured to his death by the Titans with a basket of toys and avenged by Zeus's thunderbolt; the initiate re-enacted the sacra, and the rhombos is named among them in two independent witnesses. The Gurôb papyrus, an early-third-century-BC ritual roll, gives the instruction itself — these things are to be put into the basket: “cone, bull-roarer (rhombos), knucklebones” and then “mirror” — the earliest document on earth to name a bullroarer among sacred objects being assembled for a rite. Clement of Alexandria, around AD 190, quotes the Orphic verse of the same toys: “kōnos and rhombos and a limb-bending doll and golden apples.” The god who presides is Bromios, “the Roarer, the Thunderer,” and rhombos itself means the whirling, roaring, rumbling thing: the roaring instrument, the roaring god, and the thunder-myth converge in one rite. Older still is the eeriest passage on the Greek stage. Aeschylus, in the lost Edonians, staging the arrival of Dionysos's rites in Thrace, conjures “bull-voiced” bellowers, unseen, answering the drum whose echo rolls like subterranean thunder. It was Jane Harrison who put a name to that invisible bull-thunder, and she did not hedge: “We know the very instrument with which in ancient days mimic thunder was manufactured, the famous Bull-roarer or rhombos.” Her reading has become the standard apparatus of the fragment — a disembodied bull-roar out of nowhere is what precisely one instrument makes, and the mysteries owned that instrument by name.

What survives in Europe is therefore not the instrument's origin but its long afterlife: a Palaeolithic technology whose sacred charge decayed into a children's storm-toy that still, in Aberdeen and on Krk, answers to the name of thunder. For the full arc of that demotion — the mystery instrument becoming a plaything — see God to Toy; for why the sound was heard as a god in the first place, The Mechanical Scream; for the wider storm complex the weather-roarer belongs to, The Whirlwind; and for the paired large-and-small roarers that recur from Chahai to New Guinea, Two Voices.

A dated stratigraphy

Stacked into one column, the vertical history runs from the Ice Age to the mystery cults.

A parallel East-Asian jade strand — end-pierced blades at Chahai, Xinglongwa, and Beifudi, some in graded pairs — runs alongside this column without joining it: the same deep signature, arising independently. The Siberian Mal'ta plaque (centre-pierced, snake-engraved) is a waypoint on the Beringian route, not a roarer.

Read the column bottom to top and the essay's claim is simply what the column shows. At the top of it, in Greece, the instrument is a god's voice and its god is called the Roarer. In the middle, wherever farming spread out of the Old-World Neolithic, the roarer is bent to the crop — the yams' mother in the Fly delta, the harvest's source on the Benue. And beneath all of that, in the Dordogne caves and on the Beringian trail and in the hands of every forager people who never planted a seed, the instrument does what it did first: it speaks with the sky's voice, raises the wind, calls the rain, and stands in the dark for the dead. The plow, when it arrived, borrowed an instrument that was already old. The thunder never had to borrow anything.