Who Were the Druids?

And what were they doing?

Suppose you find yourself in the leafy, misty, and rainy beauty of the West Wales county town of Carmarthen. In that case, I recommend that you pop your head into the little-visited, strangely constructed, and fascinating ruins of the town's Roman amphitheatre. If you can't find it (it's a little way out of town), ask a friendly local, and they'll be happy to point you in the right direction. To be honest, if you hadn't been told that it was a Roman amphitheatre, you might have just dismissed it as an oddly shaped bit of hillside, as 2,000 years ago, the Roman-but-not-quite locals, the Demetae, just quarried it out of the landscape - a big bowl-shaped natural feature that they flattened, poked and prodded and stuck some seats around. But playing around those weird lumps and bumps was enough to set a certain young boy on a lifelong journey into exploring the Roman world.

The other thing Carmarthen is famous for is Merlin, the wizard of the young King Arthur fame. Merlin is, of course, an entirely fictitious figure, but the general development of his history can be traced back to the Welsh poet Taliesin, who was both a real person and a mythological figure.

The origin story of Taliesin goes something like this:

Gwion Bach was a servant boy on the shores of Bala Lake, where the giant Tegid Foel and his witch wife Ceridwen lived. Tegid and Ceridwen also have a beautiful daughter, Crearwy, and a son, Morfran, who is so ugly and stupid that nothing can be done for him.

So, Ceridwen brews a potion that will make him handsome and wise, and Gwion Bach is given the job of stirring it in a cauldron over a fire for a year and a day. A blind man, Morda, tends the fire beneath.

The first three drops of the potion give the person who drinks it untold wisdom, but the rest are poisonous. As Gwion stirs the concoction, three drops fall onto him. He instinctively puts his hand to his mouth to stop the burning, instantly gaining untold knowledge and wisdom.

Terrified of Ceridwen, Gwion flees. Ceridwen, hot on his heels, turns herself into a greyhound to catch him, but with his new powers, he becomes a hare and tears off ahead of her.

Gwion becomes a fish and jumps into a river, so she becomes an otter. Gwion turns into a bird; she becomes a hawk and continues her chase.

Finally, Gwion runs into a barn and turns himself into a single grain of corn. Ceridwen turns into a hen and eats him.

After Ceridwen becomes pregnant, she instinctively knows it is Gwion. When the baby is born, her first instinct is to drown the child, but the child is so beautiful she is unable to. Instead, she casts him into the ocean in a large leather bag.

Meanwhile, there is a prince called Elffin, son of Gwyddno Garanhir, a boy who is a constant failure in the eyes of his father. Everything his father gives him to do ends in disaster, and so, in the end, he is sent off to run a lowly fish trap on a weir over a remote river. Again, it is an absolute disaster, and the weir turns up virtually no fish at all until one day, one of his men finds a leather bag caught in the weir.

Inside the bag is the reborn Gwion Fach. When Elffin sets eyes on him, he is so stunned by the whiteness of the boy's brow and crystal blue eyes. He cries out ' yma yn dal iesin', meaning 'here's one with a shining brow!'

He stuffs the child into his backpack and sets off for home, sure that his father will be furious again. But suddenly, the child begins first to speak, then to recite poetry. The poem he recites tells Elffin that Taliesin has been sent to guide him, that he's not only a great poet but also a prophet, and that by using his gifts, all Elffin's enemies will be defeated and he will become the greatest King in Wales.

Elffin goes on to become a great and noble ruler, and Taliesin the greatest 'bard' in Britain, and all this is clearly the precursor to the much later medieval tale of Merlin.

The association of the myth of Merlin with Carmarthen comes from the Welsh name for the town - Caerfyrddin. The 'Caer' comes Latin and means 'fort' and 'fyrddin' is, supposedly, the Welsh name 'Myrddin' (Welsh has mutations where, in this instance, the M can mutate to an F), which means 'Merlin'.

This is, of course, utter poppycock. What is likely is that the Welsh name for Merlin - Myrddin - is an example of reverse etymology. Instead of Merlin giving his name to Carmarthen, Carmarthen has given its name to Merlin via this tenuous association. In reality, the original name for the town - the one used by the Romans - is Moridunum, a term that literally means 'Sea Fort' and 'Caer Moridunum' has just mutated over the centuries to Carmarthen. But don't discuss this with those friendly locals because the town makes a very good living, thank you very much, selling Merlin-themed everything to tourists eager to part with lots of money. And good luck to them, too. Just don't let the cat out of the bag.

*conspiratorial winking*

Bards, druids, shamans, wizards and all that sort of stuff have always been a great feature of Welsh history, from the well-meaning but clearly invented modern version of Druidry that originated in the imaginations of bored Victorians to the laughably silly fakery of 18th-century serial nonsense peddler 'Iolo Morgannwg' (real name Edward Williams) whose mania for making up the history of a supposed druidic Welsh past borders on the psychotic. But there is, of course, a historical link to what we call 'the druids' that goes back as far as the Romans and seems if we believe Tacitus, to have its epicentre in Wales.

But, to quote the immortal words of Spinal Tap, who were the 'Children of Stonehenge', and what were they doing? Let's find out!

The primary literary witnesses are Julius Caesar, whose Commentarii de Bello Gallico offers the earliest and most comprehensive Roman description of Gallic Druidism in the 1st century BC, and Tacitus, whose Annals and Agricola describe the British Druids encountered during Roman expansion under the Julio-Claudians. These authors, though Roman and therefore not unbiased, provide the only direct literary insight into Druidic roles, rites, and suppression.

Writing in the mid-1st century BC, Julius Caesar provides the most substantial description of the Druids in his Commentarii de Bello Gallico. As governor of Gaul and the leading Roman general during the Gallic Wars, Caesar observed the political and religious institutions of the peoples he conquered. In Book 6, he identifies the Druids as one of the two most important classes in Gallic society, the other being the warrior nobility (Caesar, B.G., 6.13).

“In all Gaul there are two classes of persons of definite account and dignity: the one is the class of Druids, the other of knights.”

— Commentarii de Bello Gallico 6.13

According to Caesar, Druids were responsible for public and private religious rites, education, legal arbitration, and moral instruction. They presided over sacrifices, taught doctrines about the immortality of the soul, and were exempt from military service and taxation (Caesar, B.G., 6.14). Most notably, Caesar reports that Druidic knowledge was transmitted orally, requiring students to memorise vast amounts of material over a course of up to twenty years (Caesar, B.G., 6.14). This emphasis on oral tradition has left modern historians with little direct testimony from the Druids themselves.

He also notes that Druidism had a pan-Gallic dimension, with a centre of training and authority based in Britain, suggesting an intellectual and spiritual exchange between continental and insular Celtic societies (Caesar, B.G., 6.13). This idea would later be echoed by Tacitus, who also emphasised Britain’s importance as a cultural heartland for Druidic practice.

Caesar's account, while detailed, is ultimately Roman in outlook. His portrayal of Druidic religion—particularly the accusation of human sacrifice—must be viewed critically. He states:

“They believe that unless the life of a man is offered for the life of a man, the mind of the immortal gods cannot be propitiated... and they have sacrifices of this kind regularly performed by the Druids.”

— Commentarii de Bello Gallico 6.16

These charges of human sacrifice served a dual rhetorical purpose: reinforcing Roman notions of Gallic savagery, while justifying Roman intervention as a civilising mission. Nonetheless, the presence of some archaeological evidence—discussed later—has prevented outright dismissal of this claim.

A century later, Tacitus further elaborated the role of the Druids, particularly in the context of Roman military operations in Britain. In his Annals and Agricola, Tacitus describes the siege of Mona (Anglesey) under the general Suetonius Paulinus in AD 60, portraying the Druids as active participants in native resistance against Roman rule.

In Annals 14.30, Tacitus vividly depicts the confrontation between Roman soldiers and Druids on the shores of Mona:

“On the shore stood the enemy in a close array, with women running between the ranks, their clothing black, like the Furies, with dishevelled hair, waving burning torches. The Druids around them, lifting their hands to heaven and pouring out dreadful prayers, struck the soldiers with such awe that, as though their limbs were paralysed, they stood motionless and exposed to wounds.”

— Annals 14.30

The account captures the theatricality and emotional force of Druidic ritual, as well as the Roman view of it as dangerous and otherworldly. Tacitus notes that after the conquest of Mona, the sacred groves were destroyed and the Druids' altars, "drenched with the blood of prisoners," were demolished—a further allusion to human sacrifice (Tacitus, Annals, 14.30). His language reinforces the Roman portrayal of the Druids as obstacles to both military conquest and cultural assimilation.

While Julius Caesar pointed to Britain as the intellectual and ritual heartland of Druidic tradition, it was Tacitus who brought the island of Mona—modern Anglesey—into sharp focus as the symbolic and strategic core of resistance. According to Tacitus, the Roman assault on Mona in AD 60 under Suetonius Paulinus was not only a military campaign but also a direct blow to the spiritual and ideological leadership of native resistance (Tacitus, Annals, 14.30). Mona was, in Roman eyes, a sanctuary, a training ground, and a source of unity for the Druids.

"The troops cut down all who met them and wrapped the shrines in fire. The groves, devoted to inhuman superstitions, were destroyed: they considered it a duty to cover their altars with the blood of captives and to consult their gods by means of human entrails."

— Annals 14.30

Tacitus's description is highly charged, portraying the Druids as guardians of barbaric rites and as figures of deep resistance to Romanisation. Yet his language also suggests that Mona held more than symbolic value. It was a site of active practice and, likely, training—possibly something akin to an elite religious school.

Both literary and archaeological clues suggest that the Druids operated in a formalised educational structure, but there is no evidence of fixed architectural "colleges" in the Roman sense. Instead, Druidic learning appears to have been mobile, oral, and elite, following rules that preserved knowledge from Roman observation and ensured continuity through memorisation.

In Commentarii de Bello Gallico, Caesar remarks:

"It is said that they learn by heart a great number of verses. Therefore, some of them remain in the course of training for twenty years. Nor do they regard it as lawful to commit these to writing."

— B.G. 6.14

This refusal to write down doctrine, whether for reasons of secrecy or tradition, ensures that no Druidic texts have survived. Their educational model likely combined ritual memorisation with oral dialectic—a form of spiritual apprenticeship rather than scholastic instruction. This makes identifying specific training centres archaeologically difficult, yet Mona stands out as a candidate due to the literary focus placed upon it.

Caesar also adds that Druidic authority transcended tribal boundaries and that disputes were often resolved at a central meeting point, which he claims was located in Britain (Caesar, B.G., 6.13). Although the island is not named, Tacitus's later emphasis on Mona suggests continuity in Roman perceptions: that Britain, and Anglesey in particular, housed a pan-Celtic religious and judicial elite.

Despite the strong literary case for Mona’s centrality, archaeological confirmation remains limited and circumstantial. Anglesey does contain numerous prehistoric ritual sites, including megalithic tombs, stone circles, and Iron Age hillforts. However, identifying any of these as definitively “Druidic” is problematic.

The Roman military assault on Anglesey in AD 60, and again in AD 77 under Agricola (Tacitus, Agricola, 18), likely destroyed many surface-level features. However, several sites have yielded interesting finds:

• Din Lligwy, an enclosed settlement near the east coast of Anglesey, contains remains from the Iron Age into the Roman period. While not a religious site per se, it suggests continuity of occupation and potential ritual reuse.

• Barclodiad y Gawres, a Neolithic passage tomb reused in the Iron Age, shows signs of ritual activity and symbolic carvings. Though far earlier than the Druids’ time, it may have served as a sacred landscape appropriated by later elites.

• The Roman fort at Holyhead (Caer Gybi) was likely established to control the island post-conquest, underscoring Anglesey’s strategic and symbolic importance.

Yet no direct inscriptions, temples, or altars explicitly naming Druids have been discovered on Mona. The Roman destruction of sacred groves and possible oral prohibitions on inscribing religious identity may help explain the absence.

While Anglesey lacks clear material confirmation of Druidic institutions, continental Gaul provides valuable context. At Gournay-sur-Aronde, a major sanctuary site in northern France, archaeologists have uncovered a large Iron Age ritual complex that includes:

• Weapon deposits, including deliberately bent swords and spears

• Animal bones, arranged in structured patterns

• Human remains, some with signs of decapitation or display

• A surrounding ritual enclosure

While no explicit Druidic markers are present, the site matches Caesar’s description of sacrificial ritual and ceremonial centres presided over by Druids. Gournay may represent the type of sacred site described by Caesar in B.G. 6.17–6.18, where Druids made offerings and presided over religious law.

“They believe that the power of the immortal gods is exercised over all, that they see all things... Accordingly, they consecrate to the gods those things which they intend to offer: animals, and in times of great danger, even men.”

— B.G. 6.17

In this sense, Mona and Gournay-sur-Aronde serve as geographical poles in the Roman imagination of Druidic power: Mona, the remote and defiant sanctuary; Gournay, the continental and institutionalised sacred space.

From the Roman viewpoint, Druids were not merely religious officiants but a multi-functional elite stratum of society—priests, philosophers, judges, teachers, and advisors. Caesar and Tacitus both depict the Druids as holding significant social, moral, and political authority, suggesting a concentration of intellectual and spiritual power within a distinct, hereditary or trained class.

In B.G. 6.14, Caesar states that they taught their pupils about "the stars and their motion, the size of the cosmos and the earth, the nature of things, and the power and authority of the immortal gods."

“They hold many discussions as touching the stars and their movement, the size of the universe and of the earth, the nature of things, the power and the might of the immortal gods...”

— Commentarii de Bello Gallico 6.14

This passage has often been interpreted to suggest that Druids functioned like natural philosophers in the Greek tradition, though there is no direct evidence they developed written cosmological theories. Nonetheless, the comparison implies a sophisticated body of knowledge and the existence of structured, intergenerational teaching. Druids, in this capacity, were the intellectual class of their societies, distinct from warriors and artisans.

Caesar goes on to note that Druids were also judges:

“Whenever there is any dispute, either in a matter of public or private concern, and when any crime has been committed, when a murder has occurred, or when there is a dispute about inheritance or boundaries, the Druids settle the matter and decree rewards and punishments.”

— B.G. 6.13

Their decisions were apparently binding. Those who disobeyed Druidic rulings could be excluded from religious rites—a form of social excommunication akin to Roman exsilium sacrorum (banishment from the rites of the gods), which carried immense stigma.

This judicial function likely elevated the Druids above tribal kings in many respects, especially as they presided over intertribal disputes. It may explain their role in resistance movements, such as during the Roman invasion of Britain, where Druids likely offered not only spiritual leadership but also a legal and ethical justification for armed resistance.

Roman authors consistently associate Druids with religious ritual, often focusing on sacrifice. According to Caesar (B.G. 6.16), they oversaw both animal and human sacrifices, especially during times of communal danger. His most famous—and widely debated—claim is that of constructing wicker men in which humans were burned alive:

“Others have figures of immense size, the limbs of which, formed of osiers, they fill with living men, which being set on fire, the men perish, encompassed by the flames.”

— B.G. 6.16

There is no direct archaeological evidence for wicker men, though human remains bearing signs of violent ritual death—such as decapitation, dismemberment, or cranial trauma—have been found at sanctuaries like Gournay-sur-Aronde, Ribemont-sur-Ancre, and Alésia in Gaul, and Lindow Moss and Tollund in the British Isles. These bodies are often interpreted as ritual killings, though definitive links to Druidism remain speculative.

The divination by entrails (haruspicy), while better known from Roman and Etruscan practice, may have had Celtic parallels, though again, no definitive material evidence exists.

A recurrent feature in Roman accounts is the Druidic use of sacred groves, referred to in Latin as nemeta (sing. nemeton). These were often described as sites of sacrifice and communion with the gods. Tacitus mentions the destruction of these groves on Anglesey. Lucan, in his poem Pharsalia, offers a highly stylised, almost Gothic description of such a site in Gaul:

“Not far from the walls [of Massilia] stands a hideous wood... no birds dwell in its branches, no beasts make their lairs in it... even the trees shudder at the sight of themselves.”

— Lucan, Pharsalia 3.399–405

While poetic, Lucan’s account underscores Roman unease with Celtic religious landscapes. Archaeologically, sacred groves are elusive—being natural features—but enclosed ritual spaces such as at Gournay, Ribemont, and Roquepertuse may have fulfilled similar roles. These sites often contain ritual weapon deposits, animal bones, and sometimes human remains, arranged in structured ways.

Roman accounts give little detail on Druidic clothing, but Pliny the Elder (Natural History 24.103) describes the ritual use of mistletoe, gathered by Druids in white robes, using a golden sickle:

“The Druids hold nothing more sacred than mistletoe... they cut it with a golden sickle and catch it in a white cloth. They sacrifice two white bulls and pray that the gods will make their gift beneficial.”

— Pliny, Natural History 24.103

This is the origin of many romanticised modern images of Druids, but Pliny was not a direct observer and likely drew on secondary or stylised sources. Still, the colour white, ritual purity, and nature symbolism appear to have been strongly associated with Druidic practice in Roman perception.

The Roman campaign against the Druids was not a coincidence of conquest—it was a deliberate and ideologically motivated act of suppression. Unlike other religious or cultural groups the Romans encountered, the Druids were subject to intentional eradication. Their sacred groves were cut down, their altars were desecrated, and they themselves were executed or expelled.

Why? What made the Druids such a threat to Rome?

To answer this, we must examine the political, religious, and cultural roles Druids held in Celtic society, and how those roles directly challenged Roman hegemony.

Though primarily described as religious figures, the Druids also acted as political advisors and arbitrators. Their authority transcended tribal boundaries, and they were capable of organising pan-regional assemblies. Caesar notes in B.G. 6.13 that disputes were referred to a central authority in Britain:

“They decide all disputes, public and private... and they determine rewards and punishments. If any person does not abide by their decision, they forbid him to attend the sacrifices, which is the heaviest punishment among them.”

— Commentarii de Bello Gallico 6.13

Such power made the Druids a unifying force among otherwise fragmented Celtic tribes. In a culture without centralised kingship or imperial administration, Druidic networks may have served as a de facto national infrastructure—an intellectual and spiritual framework capable of coordinating resistance.

From the Roman point of view, that made them politically subversive. If tribal leaders were divided and in competition, the Romans could divide and conquer. But a shared spiritual elite with pan-tribal authority presented a major threat to imperial control.

Roman imperialism was not only military and economic—it was cultural and religious. After conquest, Rome typically allowed local cults to persist, often incorporating them into the imperial cult structure through syncretism. But Druidism proved resistant to this.

Unlike civic cults devoted to local deities (which could be Romanised), Druidism was non-iconic, non-literary, and resistant to imperial forms. It was transmitted orally, centred on sacred groves rather than temples, and refused to acknowledge the emperor’s divine authority.

This placed the Druids outside the client-religion model Rome preferred. As non-participants in the imperial cult, they were incompatible with Roman religious policy. Tacitus’s language in Annals 14.30—describing the burning of altars “drenched in blood”—was not only polemical but part of a broader Roman effort to delegitimise Druidic ritual.

Furthermore, Druidic teachings on immortality of the soul and cosmic law may have offered a moral and metaphysical worldview rival to Roman Stoicism and civic religion, giving people under occupation a spiritual identity separate from their imperial rulers.

The most direct motivation for Rome’s assault on the Druids was their association with rebellion.

In Britain, the suppression of the Druids occurred during some of the most volatile uprisings in the Roman conquest. Tacitus’s account of the assault on Mona in AD 60 comes just prior to Boudica’s revolt, the most famous uprising in Roman Britain.

“Suetonius prepared to attack the island of Mona, which had a considerable population and was a refuge for fugitives.”

— Annals 14.29

The timing of Suetonius Paulinus’s campaign and the symbolic importance of targeting Mona strongly suggest the Romans believed the Druids were facilitating or sheltering resistance. Though Tacitus does not explicitly link the Druids to Boudica, the Roman military’s rapid retreat from Mona in the face of her rebellion implies that both events were seen as connected. When the Romans talk of 'fugitives', are they discussing people who won't bend the knee or are these actual outlaws on the run from justice? It would seem that the terminology being used is primarily focused on propaganda - Tacitus loves to deal in broadly painted imagery of good guys and bad guys.

Moreover, in Gaul, while Caesar does not record a direct Druidic role in armed rebellion, their ability to influence legal and moral judgments may have enabled them to encourage resistance indirectly, by sanctifying opposition as lawful or sacred.

The suppression of the Druids thus formed part of a wider Roman strategy of cultural erasure. Just as the Romans eradicated the Carthaginian priesthood after 146 BC and sought to replace Egyptian temple authority with Roman-controlled cults, they saw the Druids as the ideological spine of native identity.

Destroying that spine was essential to the Romanisation of Britain and Gaul. And indeed, following the campaigns of Agricola in Britain (as recorded by Tacitus in Agricola, 18), Druidic influence appears to have declined markedly. By the second century AD, there were no more references to active Druidic communities in Roman Britain or Gaul.

The Druids were not absorbed—they were eliminated.

The Druids’ own voices were never recorded. They left no books, no temples with dedicatory inscriptions, and no direct testimonies. As Caesar noted, “They do not consider it proper to entrust their doctrines to writing” (B.G. 6.14). This presents a significant challenge: while Roman authors wrote vividly of Druidic rituals, the archaeological record is largely mute on the matter.

And yet, not entirely.

While no site has been conclusively identified as “Druidic,” certain ritual structures, human remains, and religious altars from Iron Age Gaul and Britain exhibit patterns consistent with Roman descriptions—especially regarding sacrificial rites, the use of groves or enclosures, and elite religious activity.

Unlike Roman, Greek, or even Egyptian religious officials, the Druids left no inscriptions naming themselves. No one refers to themselves as Druis or Archidruis in Latin or Celtic inscriptions. This absence is striking, but likely deliberate.

Caesar claimed that Druidic doctrine was never written down, and while administrative texts and votive inscriptions survive from Romanised Gauls, none name a Druid. This may reflect:

• The Druids’ deliberate avoidance of literacy

• The Roman criminalisation of Druidism under emperors like Tiberius and Claudius

• The transition of native religion into syncretised forms, often replacing Druids with Roman-style sacerdotes

Indeed, under Claudius (r. AD 41–54), Druidism was legally banned. According to Suetonius, Claudius “abolished the barbarous and inhuman religion of the Druids” (Suetonius, Claudius, 25). This would have further encouraged the erasure of Druidic identity from the public record.

What we can say with confidence is that Druidism, as described by Roman authors, was likely a real cultural institution, not an invention. It was religious, legal, and pedagogical in nature, forming a pillar of non-Roman Celtic identity. And it was, for that reason, methodically targeted and dismantled by the Roman Empire.

However, one thing we can be certain of is that almost everything we know about the ancient druids comes from the Romans. How they saw themselves remains a mystery.

References and Further Reading

Caesar, J. (trans. Hammond, C.). (1988). The Gallic War. Oxford University Press.

Tacitus. (trans. Mattingly, H.). (2009). Annals & Agricola. Penguin Classics.

Pliny the Elder. (trans. Rackham, H.). (1963). Natural History, Volume VI: Books 20–23. Harvard University Press.

Suetonius. (trans. Coverley, J.). (2025). The Twelve Caesars.

Lucan. (trans. Braund, S.). (1992). Pharsalia (The Civil War). Oxford University Press.

Ross, A. (1967). Pagan Celtic Britain. Routledge & Kegan Paul.

Green, M. (1997). Exploring the World of the Druids. Thames and Hudson.

Please follow my latest project on Kickstarter! There will be a discount for all subscribers (either free or paid!)