Carnac Stones Comparisons to Other Sites

About Carnac Stones Comparisons to Other Sites

The first thing to ask about Carnac is not what the stones mean but where they sit in time. For most of the twentieth century the Breton alignments were treated as a long, slow Neolithic project loosely contemporary with Stonehenge: a 4500–3300 BCE landscape, broadly Atlantic, broadly comparable. A Bayesian re-reading of the radiocarbon record has tightened that frame considerably. Bettina Schulz Paulsson's 2019 PNAS study modelled 2,410 dates from megalithic and pre-megalithic contexts across Europe and concluded that monumental stone construction emerged in northwest France in the second half of the fifth millennium cal BCE, then spread by sea — Iberia, the Mediterranean islands, the British Isles — within roughly 200 to 300 years. In June 2025, Audrey Blanchard and colleagues published rescue-excavation results from Le Plasker in Plouharnel (a previously unrecognised section of the Carnac complex) in Antiquity: a pre-megalithic monumental tomb dated to c. 4720 cal BCE and stone rows associated with cooking pits dated 4680–4250 cal BCE, with the alignments themselves built between approximately 4600 and 4300 cal BCE. The Bay of Morbihan is not a peer of the better-known megalithic centres. It is, on present evidence, the source.

That single chronological adjustment changes most of the comparisons that follow. The sections below pair Carnac with five peer sites — Stonehenge, Avebury, Newgrange, the Megalithic Temples of Malta, and Göbekli Tepe — along axes where the comparison is factually defensible: form (alignment versus circle versus enclosed temple), chronology, astronomical claims, and the social organisation a monumental project of this scale implies. Where named scholars sharpen the comparison they are cited; where the comparison breaks down, that breakdown is named.

The most basic distinction between Carnac and the British henges is geometry. Carnac's three principal groups — Le Ménec (1,169 m, originally about 1,100 stones in 11 rows), Kermario (about 1,300 m, 1,029 stones in 10 rows), and Kerlescan (880 m, about 555 stones in 13 rows) — are linear. Stonehenge and Avebury are circular. A processional row pointed at the horizon and a closed ring around an interior space encode different ideas about ritual movement, audience, and what counts as the inside of a monument. Aubrey Burl's From Carnac to Callanish: The Prehistoric Stone Rows and Avenues of Britain, Ireland, and Brittany (Yale University Press, 1993) traces the row tradition specifically — networks of paired and avenue-style stones from Dartmoor and the Scottish glens to the Breton coast — and argues that the rows belong to a distinct social grammar from the circles, even when their builders were neighbours and contemporaries.

The chronology pulls the contrast tighter. The earliest Le Ménec stones, on the Bayesian re-dating from Schulz Paulsson and the Le Plasker team, were standing by c. 4500 BCE. Avebury's henge, ditch, and stone settings — 421 metres in outer diameter, the largest stone circle in Europe by enclosed area — were built between approximately 2850 and 2200 BCE. Stonehenge's first earthwork phase begins around 3000 BCE; the recognisable sarsen circle, in Mike Parker Pearson's revised dating from the Stonehenge Riverside Project, was raised between roughly 2640 and 2480 BCE. Carnac is older than Avebury by about 1,500 years and older than the Stonehenge sarsens by about 1,800. The Atlantic megalithic tradition does not radiate from southern England outward; it radiates from Brittany, and the British circles are downstream developments.

That does not flatten the British monuments into late copies. Avebury's outer ditch was originally about nine metres deep — a colossal earthwork containing two inner stone circles and the West Kennet Avenue running south to the Sanctuary. Stonehenge's lintelled trilithons are unique in European prehistory; no Breton menhir was ever crowned with a horizontal stone. The point of the comparison is that the row-and-circle distinction is not a stylistic preference within one tradition but a long temporal sequence, with Carnac's open processional landscape coming first and the enclosed British rings coming much later. Burl reads the British rows (Callanish, Beaghmore, the Devon and Cornwall pairs) as a transmission of the Breton row idea northward; the circles are an Insular innovation grafted onto an inherited row tradition.

The transmission picture, on Schulz Paulsson's reading, is maritime rather than overland. Her 2019 model identifies a relatively narrow window — around 200 to 300 years in the second half of the fifth millennium BCE — during which megalithic construction appears at coastal sites in northwest France, southern France, Iberia, the western Mediterranean islands, and Italy. The geographic pattern follows shipping lanes rather than river systems or land routes, and the regional megalithic art (the carved orthostats at Gavrinis, the axe-and-crook motifs at the Table des Marchands, and similar engravings along the Iberian Atlantic coast) shares iconographic conventions consistent with knowledge moving between coastal communities by sea. The British and Irish circles arrive later in the same network: Newgrange c. 3200 BCE, Avebury beginning c. 2850 BCE, Stonehenge's earthwork phase c. 3000 BCE. Each was built by communities in contact with — and downstream of — the established Atlantic megalithic centre on the Morbihan coast.

Few claims in megalithic studies have generated as much methodological argument as Alexander Thom's hypothesis that the Carnac landscape encoded systematic lunar observation. In Megalithic Lunar Observatories (Clarendon Press, 1971) and a follow-up paper with Archibald Thom and J. M. Gorrie in the Journal for the History of Astronomy (1976, "The Two Megalithic Lunar Observatories at Carnac"), Thom proposed that the now-fallen Grand Menhir Brisé at Locmariaquer — 20.6 metres long and weighing approximately 330 tons, the largest known menhir ever erected in Europe — served as a universal foresight for observers stationed at backsight positions across the Carnac landscape. The horizon behind the menhir, on Thom's reading, marked the extreme rising and setting positions of the moon at major and minor standstills in the 18.6-year nodal cycle. The Grand Menhir was erected around 4700 BCE and broken (probably deliberately, possibly by earthquake) into four pieces around 4000 BCE.

Clive Ruggles, working through the same data with sharper statistical tools, was less persuaded. In Astronomy in Prehistoric Britain and Ireland (Yale University Press, 1999) and a sequence of earlier papers, Ruggles argued that with thousands of stones available, a small number of apparent alignments with lunar standstills will appear by chance, and that statistical significance must be tested against random horizon distributions before any single alignment can be treated as intentional. Ruggles found broadly consistent solar orientations — particularly an east-northeast / west-southwest axis approximating midsummer sunrise and midwinter sunset across all three principal Carnac groups — but the lunar standstill claim, on his reading, did not survive rigorous testing. The same methodology demolished a parallel claim about Thom's "megalithic yard" (a hypothetical 0.829 m unit Thom believed had been used at Carnac and elsewhere); P. R. Freeman's Bayesian re-analysis and David Kendall's frequentist analysis published in the Royal Society's Philosophical Transactions A found supporting evidence in some Scottish circles but not in the British or Breton data taken together.

Newgrange supplies the cleanest astronomical comparison. The Boyne Valley passage tomb, built around 3200 BCE on current calibrated dating, contains a roof-box opening above the entrance that admits direct sunlight along the 19-metre passage to the chamber for about 17 minutes at sunrise on the winter solstice. The alignment is unambiguous, repeatable, and survives the kind of statistical scrutiny Ruggles applied at Carnac. The relevant comparison is not whether Carnac's builders were less astronomical than Newgrange's, but that the Carnac alignments encode a different kind of astronomical engagement: an open processional axis very roughly oriented to the solar year, set in a landscape that also includes passage tombs (Gavrinis on its island in the Gulf of Morbihan, with carved orthostats and a midwinter-sunrise-aligned passage) where the Newgrange-style precision does occur. Brittany has both the row tradition and the chambered-passage tradition; Ireland borrowed the chamber and refined it.

The honest position on the lunar standstill claim is that Thom's measurements were careful but his statistical interpretation was weak; that Ruggles's critique is correct on its own terms but does not foreclose the possibility that some Carnac stones were erected with the lunar cycle in mind; and that the case-by-case alignment analysis at Gavrinis, Le Petit Mont, and the Table des Marchands continues to yield individually defensible solar orientations even where the Thom-style global lunar grid fails.

The Megalithic Temples of Malta — Ġgantija on Gozo (c. 3600 BCE), Ħaġar Qim, Mnajdra, the Tarxien complex (in use until c. 2500 BCE) — are the most informative anti-comparison for Carnac. Both are Neolithic. Both rely on locally quarried limestone or granite raised without metal tools. Both belong to the wider Atlantic-Mediterranean megalithic horizon that Schulz Paulsson's diffusion model places in motion from northwest France in the late fifth millennium BCE. After that, the comparison breaks down — productively.

The Maltese temples are free-standing buildings: trefoil and pentafoil apsidal plans, doorways, paved interior courts, altars, libation holes, statuary. Ħaġar Qim's largest façade stone weighs approximately 57 tons and is set within an articulated architectural envelope, with thresholds, lintels, and finished interior faces. They were, until the Göbekli Tepe excavations made the comparison harder to sustain, regularly described as the oldest free-standing buildings in the world. Carnac, by contrast, has no buildings. The Ménec and Kerlescan alignments terminate at small stone enclosures (the Ménec enclosure is roughly 90 metres across, an oval of larger stones), but those enclosures are open to the sky and were not roofed. Brittany invested its labour in arranging stones across a landscape; Malta invested an arguably comparable labour budget in articulating interior space.

The contrast suggests a question that purely architectural comparison cannot resolve. If both regions inherit megalithic technique from the same fifth-millennium Atlantic source, why does the same toolkit produce a four-kilometre processional landscape in one place and apsidal temples in another? Mark Patton's Statements in Stone: Monuments and Society in Neolithic Brittany (Routledge, 1993) reads the Breton answer through landscape and exchange: the Carnac alignments belong to a region positioned at the centre of long-distance jadeitite axe networks (with Alpine stone axes recovered from passage graves more than 1,000 km from their source), where monumental statements were external, processional, and territorial. The Maltese answer was internal, architectural, and bounded — a temple culture rather than a landscape one. The contrast is real; the underlying technology is shared.

Göbekli Tepe in southeastern Turkey, dated c. 9600–8000 BCE, sits 5,000 years before Carnac and represents an entirely separate megalithic tradition: pre-pottery Neolithic, pre-agricultural in its earliest phases, with carved T-pillars set in circular enclosures by hunter-gatherer or proto-farming communities. The temptation to draw a continuous line from Göbekli Tepe through Carnac to Stonehenge is a temptation worth resisting. Bettina Schulz Paulsson's 2019 model finds no evidence for east-to-west diffusion of megalithic ideas from Anatolia to Atlantic Europe; the radiocarbon record places the Atlantic megalithic horizon's origin in Brittany, c. 4700–4500 BCE, with no plausible intermediate link to the Anatolian tradition five millennia earlier. Karahan Tepe and Çatalhöyük, both contemporary with or earlier than Göbekli Tepe in the Anatolian sequence, similarly do not feed forward into the Atlantic story.

What the two endpoints of the megalithic story share is a more interesting fact: in both places, stone-raising began before centralised states, before writing, and (at Göbekli Tepe) before settled agriculture. Klaus Schmidt's original interpretation of Göbekli Tepe — that "first the temple, then the city" — has been complicated by more recent excavations showing some domestic occupation alongside the ritual enclosures, but the basic point survives: monumental construction is older than urbanism. Carnac extends the same point into the fifth millennium. The Morbihan farming communities who raised the alignments had no metal, no writing, no cities, and no centralised political authority detectable in the burial record, yet they sustained a monumental programme across more than a millennium. Whatever explains the impulse to raise stones in coordinated formation, it predates the centralised states usually invoked to explain monumental labour.

The Grand Menhir Brisé at Locmariaquer — 20.6 metres long, approximately 330 tons, now in four pieces — is the largest single menhir ever erected in Europe and one of the few Atlantic megaliths that bear comparison with the largest stones at Stonehenge or in the Egyptian quarrying tradition. The four fragments lie where they fell on a low rise about 11 km from the principal Carnac alignments, the carefully prepared socket pit confirming that the menhir did stand upright before its breakage around 4000 BCE. The same Atlantic megalithic landscape contains the Cairn de Barnenez in Finistère — first phase 4850–4250 BCE on the radiocarbon dates, a 72-metre-long stone cairn enclosing eleven chambered passage graves — and the Tumulus Saint-Michel at Carnac itself, 125 metres long, 60 metres wide, and 10 metres high, with construction dates in the mid-fifth millennium BCE.

By individual stone weight, the Grand Menhir Brisé exceeds anything at Stonehenge: the largest sarsen at Stonehenge — Stone 56 of the Great Trilithon — weighs an estimated 40 to 50 tons (the average sarsen circle upright is about 25 tons), still roughly seven times lighter than the Grand Menhir. By total monumental volume, the Tumulus Saint-Michel and Barnenez cairn are smaller than later state monuments but comparable to Avebury's Silbury Hill and rival the early phases of British long barrows. The numbers matter because they answer a recurring claim in popular literature that the Atlantic megalithic builders were technologically modest. They were not. They moved single stones heavier than anything Stonehenge's builders attempted, organised long-distance transport (the Grand Menhir's quarry has been provisionally identified some 10 km from its erection site), and built mortuary mounds at landscape scale — and they did so a millennium and a half before the British circles.

The Grand Menhir's breakage is a separate puzzle. Whether it fell from earthquake, ground subsidence, or deliberate ritual decommissioning is unresolved; the careful socket preparation makes simple engineering failure unlikely, and the four near-equal fragments suggest the stone may have been broken on the ground after toppling rather than failing in mid-erection. A separate 14-metre carved stele in the same Locmariaquer alignment was broken before erection and its fragments redistributed across the ritual landscape: matched pieces survive as the capstones of the Table des Marchands, Er-Grah, and Gavrinis dolmens, with the engraved decoration aligning across the breaks. Whether the Grand Menhir Brisé itself supplied any reused capstone material is less certain. Either way, the broken-stele tradition shows that nothing of value was wasted within the Locmariaquer ritual landscape.

Mark Patton's Statements in Stone (Routledge, 1993) reads the Carnac alignments alongside the prestige-axe networks centred on the Breton coast. Polished jadeitite axes, quarried in the western Alps, were carried more than 1,000 km to be deposited in passage graves at Tumiac, Mané er Hroëck, and other Morbihan tombs; some of the same axes were buried unworn, treated as ritual offerings rather than tools. The labour budget required to raise the alignments — thousands of stones erected over more than a millennium, with the larger stones requiring teams of fifty to two hundred people working with ropes and timber — is consistent with the social organisation a long-distance prestige-goods economy implies: not a centralised state, but coordinated communities with sustained access to surplus and the authority to direct collective labour across generations.

The comparison with the Maltese temple society sharpens the social inference. Malta's roughly 1,100-year temple-building phase (3600–2500 BCE) ended in a still-debated cultural collapse during which the temple sites were abandoned. Carnac's 1,200-year monument-building phase ended with a gradual transition into Bronze Age land-use patterns rather than a single rupture; the alignments were not destroyed by their builders' descendants, only allowed to decline as the ritual focus of the surrounding farming communities shifted. The contrast suggests two ways a Neolithic monumental tradition can wind down — collapse versus drift — and the Breton case arguably preserves more of its constructed landscape because the transition was less violent. Avebury and Stonehenge follow a third pattern: continued use and modification well into the Early Bronze Age, with sarsen settings raised, rearranged, and (at Stonehenge) progressively elaborated long after the original monumental impulse had passed.

The Carnac alignments are not a regional curiosity within a Stonehenge-centred Atlantic megalithic story. They are, on the present radiocarbon evidence, the early end of that story — the source from which the row tradition spread to the British Isles and the chamber tradition spread alongside it. The row form is older than the circle form; the Breton landscape encodes the original solution to the problem of monumental open-air ritual space, and the British monuments are subsequent regional dialects. The astronomical claims are mixed: Thom's global lunar grid does not survive Ruggles's statistical tests, but the broad solar orientation of the alignment axes is genuine, and the regional pattern of midwinter-aligned passage tombs (Gavrinis in Brittany, Newgrange in Ireland) is on firmer ground than the Carnac lunar standstill hypothesis. The contrast with Malta separates two different uses of the shared megalithic toolkit — landscape versus architecture — and the contrast with Göbekli Tepe restores the chronological depth of monumental construction, which began in stone-raising, hunter-gatherer Anatolia 5,000 years before the Bay of Morbihan was farming. Each comparison narrows the question rather than answering it: Carnac's purpose remains debated, but its position in the European Neolithic is no longer in serious doubt.

Significance

Comparison work is what moved Carnac from "regional Atlantic curiosity" to "source point of the European megalithic tradition." Bettina Schulz Paulsson's PNAS 2019 Bayesian model of 2,410 radiocarbon dates and Audrey Blanchard's 2025 Antiquity publication on Le Plasker together place the Carnac alignments at c. 4600–4300 cal BCE — earlier than any megalithic monument in the British Isles, older than the Maltese temples, and the apparent origin of the maritime diffusion network that carried stone-raising technology around the Atlantic and Mediterranean coasts within roughly two to three centuries.

The site's place in archaeoastronomy is also defined by comparison. Alexander Thom's lunar standstill hypothesis at Locmariaquer set the methodological terms; Clive Ruggles's statistical critique answered them. The resulting framework — that astronomical claims at any megalithic site must survive testing against random horizon distributions — became standard practice from Stonehenge to the Nazca Lines.

Connections

Carnac Stones — the parent entity. This sub-page focuses on cross-site comparisons; the parent covers the alignments, Tumulus Saint-Michel, Grand Menhir Brisé, and the broader Morbihan landscape in standalone depth.

Stonehenge — the principal Atlantic megalithic peer. Stonehenge's sarsen circle (c. 2640–2480 BCE on Mike Parker Pearson's dating) post-dates the Carnac alignments by roughly 1,800 years; the row tradition at Carnac and the circle-and-trilithon tradition at Stonehenge are sequential, not simultaneous, developments within the same Atlantic horizon.

Avebury — Britain's largest stone circle (421 m outer diameter), built 2850–2200 BCE. Like Stonehenge, Avebury demonstrates the Insular preference for enclosed circular space; its henge bank and ditch system has no parallel at Carnac.

Newgrange — the closest astronomical comparison. The Boyne Valley passage tomb's roof-box midwinter sunrise alignment (c. 3200 BCE) is technically related to the passage-grave tradition that produced Gavrinis in the Carnac landscape; both belong to the chamber side of the Breton-Irish megalithic exchange.

Megalithic Temples of Malta — Ġgantija (c. 3600 BCE), Ħaġar Qim, Mnajdra, Tarxien (in use until 2500 BCE). The most useful anti-comparison: shared megalithic technique, opposite architectural choice — Malta builds enclosed apsidal temples, Carnac arranges stones across an open landscape.

Göbekli Tepe — c. 9600–8000 BCE, southeastern Turkey. Pre-agricultural megalithic construction in a tradition unrelated to the Atlantic horizon; the comparison establishes the chronological depth of monumental stone-raising rather than direct cultural connection.

Karahan Tepe — sister site to Göbekli Tepe in the Tas Tepeler complex (c. 9400 BCE). Like Göbekli, separated from Carnac by 5,000 years and a continent of cultural distance.

Çatalhöyük — c. 7500–5700 BCE, Anatolia. Predates Carnac by roughly two millennia. A settled farming community without monumental stone construction; the Anatolian megalithic and Anatolian farming traditions did not run on the same track.

Easter Island (Rapa Nui) — moai construction c. 1250–1500 CE. An independent stone-raising tradition with no historical connection to the Atlantic megalithic complex; the comparison separates "monumental stone in lines" as an Atlantic Neolithic phenomenon from "monumental stone as Polynesian ancestor figures" five millennia later and a hemisphere away.

Archaeoastronomy — the discipline that took its modern form through the Thom–Ruggles exchange at Carnac. Statistical testing of horizon alignments against random baselines, now standard practice, was sharpened on Carnac data first.

Further Reading