Monday, June 22, 2026 · 天火 · tianmu.org

Introduction to Continental Celtic Sources

✦ ─── ⟐ ─── ✦

The Discipline Of Fragments

The Continental Celtic room begins where smooth mythology fails.

A reader who comes from Irish saga or Welsh narrative may expect scenes, speeches, heroes, gods, cattle raids, enchantments, marriages, prophetic women, severed heads, otherworld journeys, and the deep pressure of story. Continental Celtic evidence rarely behaves that way. It gives a different kind of gift: a god's name on an altar, a broken line on lead, a public bronze whose law cannot yet be fully translated, a potter's account from a workshop, a Roman soldier's vow to Epona, a Greek geographer's sketch of Galatian political order, a Celtiberian hospitality tessera, a damaged Gaulish caption beneath a carved divine figure, a calendar reduced to bronze fragments, a river sanctuary where the goddess is visible because people left offerings.

This evidence is older in one sense than the medieval Irish and Welsh manuscripts. It often stands closer to ancient Celtic-speaking communities in time, place, and material practice. But it is also harder. The sources are fragmentary, multilingual, damaged, formulaic, hostile, Romanized, Hellenized, archaeologically displaced, or dependent on scholarly reconstruction. They do not give the reader a Celtic Bible, a Druidic catechism, or a complete Gaulish myth cycle.

That is why this room matters.

The Continental evidence teaches the reader the discipline of fragments. It says: do not despise a name because it is not a story. Do not turn a name into a story because you want one. Do not treat Roman Latin as proof that native religion vanished. Do not treat a Celtic divine epithet as proof that Roman categories are irrelevant. Do not make every broken lead tablet into a universal spell. Do not read Caesar as neutral ethnography. Do not read Celtiberian public texts as empty because they cannot yet be translated line by line. Do not mistake uncertainty for failure. In this field, uncertainty is often the most honest form of knowledge.

The Good Works Library gathers this room because the larger Celtic shelf would be distorted without it. Ancient Celtic-speaking worlds were not confined to the later Gaelic and Brittonic manuscript zones. Gaul, Celtiberia, northern Italy, Galatia in Anatolia, and the Roman western provinces all preserve pieces of Celtic language, religion, society, and identity. These pieces rarely assemble into a single picture. They form a field.

The reader's task is to learn how to walk that field without trampling the evidence.

I. What "Continental Celtic" Means Here

"Continental Celtic" is a practical shelf name, not a claim that all the sources here belong to one single people or religion.

The room covers Celtic-speaking or Celtic-associated evidence from ancient and Roman-period Europe and Anatolia outside the better-known medieval Irish, Welsh, Scottish, Manx, Cornish, and Breton literary traditions. It includes Gaulish inscriptions and calendars; Gallo-Roman Latin dedications preserving local divine names; Greek and Roman authors writing about Gauls, Celts, and Galatians; Galatian and Anatolian sources from the eastern Celtic world; and Celtiberian and Hispano-Celtic texts from the Iberian peninsula.

These source groups differ sharply.

Gaulish inscriptions are direct witnesses to a Celtic language, but many are short, damaged, or formulaic. Gallo-Roman dedications are usually in Latin, not Gaulish, but they preserve local gods, epithets, cult sites, dedicants, vows, and the social texture of worship under Roman rule. Classical witnesses are readable and continuous, but they are outsiders: generals, geographers, poets, historians, encyclopedists, and moralizers writing from Greek and Roman worlds. Galatian sources preserve the Celtic-speaking migration into Anatolia, but often through Greek and Latin narratives or Greek inscriptions rather than Galatian-language texts. Celtiberian sources preserve a western Celtic language in Palaeohispanic and later Latin scripts, but their syntax, institutional vocabulary, and legal formulas are often only partly understood.

This room therefore has no single reading method. It has a set of source methods.

The strongest first rule is to name the source type before interpreting it. A line in Caesar is not a Gaulish self-description. A Latin dedication to Mars with a Celtic epithet is not simply a Roman god and not simply an untouched Celtic god. A Gaulish lead tablet is not a myth. A Celtiberian bronze is not a folktale. A Galatian Zeus epithet in Greek is not a Galatian hymn, but it may preserve local cult identity. A potter's account from La Graufesenque is not religious literature, but it is direct evidence that Gaulish language could be written in an economic setting.

The second rule is to keep scale. Some sources tell us about institutions, migrations, and war. Others tell us about one dedicator, one vow, one sanctuary, one word, one woman in a magical text, one military unit, one local goddess, one legal formula. A small source is not a failed large source. It is evidence at the scale it survived.

The third rule is to resist both romance and flattening. Continental Celtic evidence has often been used by modern readers to feed a fantasy of ancient Druids, warrior mysticism, lost calendars, pure nature religion, or universal Celtic spirituality. The corrective should not be a dead, purely technical reading. The sources are alive with religious acts: vows fulfilled, names spoken, springs honored, gods paired, mothers invoked, curses written, calendars kept, public laws displayed, and communities negotiating empire. The point is not to drain the wonder. The point is to locate it exactly.

II. Classical Witnesses: Readable, Necessary, Dangerous

The easiest sources in this room to read are often the most dangerous to trust.

Caesar, Strabo, Diodorus, Livy, Pliny, Lucan, Pomponius Mela, Polybius, Plutarch, Athenaeus, Pausanias, and Ammianus give continuous Greek and Latin prose or poetry about Celtic peoples. They describe Druids, warriors, migrations, feasting, law, sacrifice, gods, plants, souls, battles, tribal geography, and encounters with Rome and Greece. Without them, the field would be far poorer. But none of them is a neutral window.

Caesar writes as a conqueror. His Gallic War gives precious information about Gaulish society, assemblies, Druids, sacrifice, and conflict, but it also explains and justifies Roman war. Caesar's Gaul is a political field he is subduing and narrating to Rome. His clarity is part of his power. A reader should learn from him, but never forget the hand that writes.

Strabo writes as a geographer under the Roman world order. His chapter on Galatia is one of the best compact descriptions of the three Galatian peoples, their tetrarchies, Drynemetum, Pessinus, and the Roman provincial aftermath. It is invaluable, especially because it preserves political and geographic structure. Yet Strabo too writes from outside. His Galatia is a region in a Greek geographical system and Roman imperial present.

Diodorus and Athenaeus preserve ethnographic and Posidonian material on Celtic feasting, warriors, poets, and customs. Livy tells Roman stories of Gauls in Italy and Galatians in Asia, with all the drama and prejudice of Roman historical writing. Lucan's passage on the gods of Gaul and the Druids is powerful poetry, not a field report. Pliny's sacred plant lore and mistletoe account are essential, but he writes as an encyclopedist collecting marvels, natural history, and inherited report. Pomponius Mela's statements about the souls of the dead and the Gallizenae are enticing, but they belong to a Roman geographical imagination of the edges of the world.

The classical witnesses therefore need a double reading.

First, read them for the data they may preserve: names, institutions, reported practices, geography, social categories, cult details, and comparisons. Second, read them as Greek and Roman texts: what does the author need the Celt to be? Savage, noble, terrifying, philosophical, decadent, brave, excessive, exotic, conquered, unconquered, useful, or morally instructive?

The page on Caesar on Druids, Sacrifice, and the Gods of Gaul should not be treated as a Druidic handbook. The page on Lucan's gods of Gaul should not be treated as a calm pantheon list. The page on Pomponius Mela's Gallizenae should not be turned into a complete priestess tradition. The page on Livy's Galatians preserves Roman rhetoric about degeneration and terror; the translation keeps the hostility visible because it is part of the source.

Classical texts are doors, not authorities that end the question. They give the reader narrative continuity, but continuity can seduce. In this room, the smoothest prose is often the least direct witness.

III. Gaulish Inscriptions: Direct Voices, Damaged Sentences

The Gaulish inscriptions are among the room's most precious sources because they preserve a Celtic language directly.

They appear in different scripts and settings: Gallo-Greek, Gallo-Latin, occasional Gallo-Etruscan or related northern Italian contexts, stone inscriptions, lead tablets, pottery, instrumentum, monument labels, dedications, accounts, calendars, and magical texts. The RIIG project, building on the older Recueil des inscriptions gauloises, is crucial because it gives updated critical editions, archaeological context, linguistic analysis, and a digital way to approach the corpus. For ordinary readers, this means the Good Works page should never pretend that a Gaulish inscription is a loose piece of poetic inspiration. It is an edited object with a findspot, script, material, readings, bibliography, and interpretive limits.

The Larzac lead tablet is the great example. It is one of the longest Gaulish texts, written on lead, across several faces, by more than one hand. It appears to concern women's spell-work, names, binding or counter-binding, underworld language, and magical action. It is irresistible to readers because it feels close to ancient Celtic magic. But it is not a clean spell in English. The source is damaged and difficult; the syntax and vocabulary remain debated. Good Works therefore presents the Larzac Lead Tablet as a source dossier: face by face, zone by zone, with cautious English where the evidence can bear it. That is not timidity. It is fidelity.

The Pillar of the Boatmen shows another kind of Gaulish source. Its Latin dedication to Jupiter Best and Greatest stands beside labels for Roman and indigenous figures: Esus, Tarvos Trigaranus, Cernunnos, Smertrios, and others. The Pillar of the Boatmen does not give a myth about Cernunnos. It gives a public Gallo-Roman monument in which the Parisian boatmen set up a religious object under Tiberius, mixing Latin dedication, Roman gods, Gaulish divine names, and damaged ritual captions. It is a visual and linguistic program, not a storybook.

The Reze lead account is different again. It is a port ledger, not an occult tablet. The Reze Lead Account preserves rows, numerals, money signs, names, erasures, and account notation. Its value is precisely that it is ordinary. It shows Gaulish used for written economic activity in a port setting. A religious library should care about that because religion does not live only in myths. Language, literacy, trade, status, and public life are part of the world in which cult also lived.

The La Graufesenque workshop ledgers likewise matter because they place Gaulish in the life of labor, production, firing accounts, and names. Votive clusters from Bouches-du-Rhone, Nimes, Gard, Glanum, Vaucluse, Vallauris, Marseille, Alise, Autun, Nevers, and other sites preserve divine names, sacred places, dedications, and formulae. The Coligny calendar, although it deserves a fuller dedicated treatment, is the monumental reminder that Celtic timekeeping could be written, technical, and astronomically structured, even if the surviving bronze is fragmentary and scholarly reconstruction remains contested.

The rule for Gaulish inscriptions is simple: let them be inscriptions. Do not overtranslate them into fluent revelation. Preserve the source text, script, object, place, damage, readings, and uncertainty. If the inscription gives a god's name, honor the name. If it gives an account, preserve the account. If it gives a damaged spell, do not complete the spell with imagination and call it translation.

IV. Gallo-Roman Dedications: Local Gods In Roman Formulae

The Gallo-Roman Latin dedication dossiers are among the clearest ways to meet living cult in this room.

Most of these inscriptions are not in a Celtic language. They are Latin texts using Roman epigraphic formulae: sacred to, to the god or goddess, in honor of the divine house, for the safety of someone, fulfilled a vow willingly and deservedly, from their own resources, by command of the deity. A careless reader might therefore say: these are Roman, not Celtic. That is too simple. The Latin language and Roman formulae are part of the evidence, but so are the divine names, epithets, cult sites, dedicants, military units, households, sanctuaries, rivers, springs, and regional clusters.

Epona is the best-known example. The Epona dedication cluster shows her not through a single myth but through use: temples, altars, statues, cavalry, pack-animal personnel, frontier units, personal vows, and protective dedications from Gaul to Germania, Pannonia, Britain, Moesia, and Raetia. Her cult travels through the Roman west, especially in relation to horses, transport, military life, and protection. The inscriptions do not tell a story about Epona's birth or deeds. They show people making vows.

The Matronae are even more local and plural. The Matronae of Cologne and Germania Inferior appear under many bynames: Audrinehae, Autriahenae, Aufaniae, Afliabus, Andrustehiabus, Boudunneihis, and others. A reader looking for "the Mother Goddess" may miss the point. These are local Mothers, often regionally named, receiving household vows, military dedications, women's dedications, civic-space gifts, and commands attributed to the goddesses themselves. The plural matters. The epithets matter. The place matters.

Rosmerta appears frequently beside Mercury. The Mercury and Rosmerta dossier shows paired divine names across Gaul, Belgica, Germania, and beyond, with shrines, altars, village dedications, family safety, and local epithets. This pairing should not be flattened into "Mercury and his consort" without attention to the inscriptional evidence. Rosmerta has her own name, distribution, and cult force; Mercury's Roman name may be carrying an indigenous god, a Roman god, a fused provincial god, or a local religious relation that cannot be reduced to a family diagram.

The same is true for Sucellus and Nantosuelta, Apollo Grannus and Sirona, Sequana at the Seine springs, Belenus at Aquileia and elsewhere, Mars under Celtic epithets, and regional divine figures attached to springs, healing sites, military life, households, and civic dedications. Interpretatio Romana is not a disguise to be peeled away until a pure Celtic god appears underneath. It is itself a religious process. A local power receives a Roman name, or a Roman god receives a local epithet, or a provincial cult uses the empire's public language to preserve a regional relationship.

The reader should therefore ask four questions of each Latin dedication.

Who dedicates? A soldier, freedperson, woman, family member, village, guild, official, priest, or unit.

Where? A spring, sanctuary, frontier fort, city, rural shrine, workshop, bridge, or temple.

To whom? A Roman god, local god, paired deity, plural Mothers, regional epithet, imperial divine house, or mixed formula.

What act? Vow, temple, altar, statue, shrine, bridge, command, thanksgiving, safety prayer, household protection, public offering.

The answer will rarely be mythology. It will be cult practice. That is the treasure.

V. Galatia: Celtic Speech In Anatolia, Greek And Roman Witnesses

The Galatian room reminds the reader that Celtic history did not stop at Gaul or the western fringe.

In the third century BCE, Celtic groups moved through the Balkans and into Anatolia, where they became known as Galatians. Greek and Roman sources name the Trocmi, Tolistobogii, and Tectosages; describe their settlement, warfare, tribute, and political structures; and locate them in relation to Bithynia, Phrygia, Cappadocia, Pessinus, the Halys, and later Roman provincial order. Strabo's chapter on Galatia is especially important because it describes the three peoples, their shared language, their tetrarchies, their council at Drynemetum, and their territorial centers. Livy gives a Roman narrative of migration and military campaign, including Manlius's hostile speech before war.

These narratives are not Galatian self-writing. They are Greek and Roman sources about Galatians. Yet they preserve a field otherwise difficult to see. Galatian was a Celtic language spoken far from the Atlantic and western European zones later associated with Celtic identity. Its public record is thin, but personal names, tribal names, place relations, and occasional religious traces keep it visible.

The Galatian divine-epithet dossiers are small but important. Greek inscriptions naming Zeus Bussurigios, Zeus Souolibrogenos, Zeus Sarnendenos, Zeus Narenos, and related local forms are not written in Galatian, but they preserve local cult names in a Galatian-Anatolian world. A name such as Deiotaros in a votive setting does not by itself create a Celtic theology, but it anchors a personal and political name-world beside local divine forms. The result is a borderland record: Greek language, Anatolian religious geography, Galatian personal and tribal memory, Roman chronology, and local Zeus cults.

Galatia also guards against a narrow idea of Celtic tradition as western, insular, misty, and medieval. The Galatians lived amid Phrygian, Greek, Anatolian, Hellenistic, and Roman worlds. Pessinus and the Mother of the Gods belong to the same regional conversation as Galatian political order. When the New Testament later addresses Galatians, it is speaking into a region shaped by these older movements, though the Christian text is not itself evidence for ancient Galatian religion in a simple way.

Read the Galatian sources as migration evidence, political geography, and provincial religious trace. Do not force them into Irish mythic parallels. Do not erase their Celtic identity because the texts are Greek and Latin. The room is a map of entanglement.

VI. Celtiberian And Hispano-Celtic Sources: Public Texts At The Edge Of Translation

The Celtiberian and Hispano-Celtic room is one of the strongest warnings against pretending that direct evidence is automatically easy evidence.

Celtiberian is a Celtic language of ancient Iberia, preserved in inscriptions written mostly in Palaeohispanic scripts and, in some later cases, the Latin alphabet. The corpus includes coin legends, graffiti, stone inscriptions, hospitality tesserae, bronze documents, sanctuary inscriptions, and public or legal texts. It is direct Celtic-language evidence, but much of it remains difficult. Scholars can identify names, formulae, cases, institutional words, some legal or public contexts, and certain divine anchors, but many complete translations remain impossible.

The Novallas Bronze is a perfect Good Works test case. It is a late Republican public document in Celtiberian, written in the Latin alphabet, and one of the most important recent Celtiberian finds. It probably concerns public space, legal measurement, or regulated open distance; the Latin-looking public vocabulary and numerical expressions are important. But the controlling publication is explicit that a wholly satisfactory translation cannot be given. A responsible page therefore says what is secure: public document, Celtiberian language, Latin alphabet, late Republican context, display character, likely legal or spatial regulation. It does not pretend to provide a smooth law.

The Luzaga Bronze is a lost object known through the source record, usually treated as a hospitality or agreement document. Words such as karuo, gortika, and gortikam stand at the center of the interpretation, but the syntax remains debated. The source's lost status also matters. Good Works preserves the dossier as a record of a record: line text, apparatus, institutional vocabulary, place and community anchors, and caution.

The Penalba de Villastar Luguei inscription gives a stronger religious anchor: Luguei / Lugus appears in a sanctuary context. That is significant. It does not authorize importing all later Irish Lugh material into the Celtiberian inscription. Lugus, Luguei, Lugh, Lleu, and related name-worlds invite comparison, but comparison is not translation. The Penalba source must be read first as a Celtiberian sanctuary inscription, with its own damaged lines and apparatus.

Hospitality tesserae are another crucial category. Small objects, often bronze, can preserve formal relations of hospitality, pact, recognition, or institutional belonging. They should not be romanticized as tribal passports without evidence. They show that social bonds could be materialized in inscribed objects, that legal and ritual relation could be portable, and that Celtiberian language belongs to public and formal life.

The deeper lesson is that Celtiberian sources are not marginal because they are difficult. They are central precisely because they show Celtic language in institutions: law, hospitality, sanctuary, public display, and local community. If Irish saga gives the reader mythic narrative, Celtiberian bronze gives the reader civic grammar at the edge of decipherment.

VII. Divine Names Are Not A Pantheon By Themselves

This room contains many divine names: Epona, Sequana, Sirona, Rosmerta, Sucellus, Nantosuelta, Belenus, Grannus, the Matronae, Taranus or Taranis-related forms, Cernunnos, Esus, Tarvos Trigaranus, Smertrios, Luguei, local Mars epithets, local Mercury epithets, local Zeus epithets, and many others.

It is tempting to gather them into a pantheon.

That temptation must be resisted.

A pantheon is not just a list of names. It is a structured religious system: relationships, myths, cult hierarchies, ritual calendars, priesthoods, images, places, narratives, and communities. Continental Celtic evidence gives us many names, but rarely the complete relationships among them. A god known from one inscription may have been locally important and regionally unknown. A goddess known from many inscriptions may still lack a surviving myth. A Roman divine pairing may reflect local theology, Roman habit, military movement, or epigraphic convention. A name may be a title, a place-linked epithet, a local form, or a deity known only through the act of one dedicator.

This does not make the names less holy or less important. It means we should not force them to answer questions their sources do not answer.

Epona is widely attested, but her surviving record is mostly inscriptional and iconographic, not mythic. The Matronae are abundant in dedications, but they are not one generic Mother Goddess. Sequana's power is tied to the Seine springs and healing offerings. Rosmerta appears often beside Mercury, but the nature of that relation must be read inscription by inscription. Sucellus and Nantosuelta appear in paired and local forms. Belenus, Grannus, Sirona, and other healing or solar-associated figures live in cult settings that may cross Roman and local categories. Cernunnos is famous because of images and labels, but the surviving source base is far thinner than modern popularity suggests.

Modern Pagan and Druidic readers may meet these names devotionally. That is legitimate as modern practice when named honestly. But a source room must distinguish devotional reconstruction from source evidence. A prayer to Epona written today is modern religion. A Latin altar to Epona from a Roman frontier context is ancient evidence. The two may be in relationship. They are not the same object.

Good Works can serve both scholarship and living spirituality only if it does not confuse the two.

VIII. Translation As Ethical Restraint

The most important craft virtue in this room is restraint.

A weak translation of fragmentary evidence often fails by being too generous. It smooths broken syntax, fills lacunae, chooses the most exciting meaning, translates a disputed word as if it were secure, turns a list into a ritual, turns a formula into theology, or turns a divine name into a personality. The result is readable and false.

A better translation may look less satisfying at first. It may say "source guide" instead of "translation." It may preserve brackets, question marks, apparatus notes, alternate readings, line numbers, object notes, and damaged zones. It may translate only the secure parts. It may explain why a continuous English rendering would lie.

This is not a failure of imagination. It is respect for the dead and the source.

The Good Works method for Continental Celtic pages should therefore be explicit:

  • Give a reader-facing summary first, so ordinary readers know why the source matters.
  • Identify object, place, date range, script, language, edition, and source route.
  • Separate direct translation from interpretation guide.
  • Keep divine names, personal names, and technical source words visible when translation is unsafe.
  • Mark damaged readings and major alternate readings.
  • Do not import Irish or Welsh parallels into the translation layer.
  • Use parallels only in commentary, and label them as comparison.
  • Preserve the difference between Latin, Greek, Gaulish, Celtiberian, and modern English.
  • Tell the reader when the source is a vow, curse, account, label, dedication, public document, literary passage, or later summary.

This is how a public library avoids manufacturing false antiquity.

There is also a spiritual reason for restraint. Fragmentary sources ask for humility. They are not ours to complete at will. The missing part belongs to history, and sometimes the holiest thing a translator can do is leave the break visible.

IX. A Reader's Path Through The Room

Begin with the Reader's Guide to Continental Celtic Sources for a short map.

Then read the Classical Witnesses if you need continuous prose. Start with Caesar on Druids, sacrifice, society, and the gods of Gaul; Strabo on Gaul and Galatia; Diodorus on Celtic warriors, poets, and Druids; Pomponius Mela on souls and the Gallizenae; Pliny on priests, plants, and sacred natural history; and Livy on the Gallic sack of Rome or Galatians in Asia. Keep asking: what does the author know, what does the author need, and what genre is this?

Move next to the Gallo-Roman Latin dedications. Read Epona, the Matronae, Mercury and Rosmerta, Sucellus and Nantosuelta, Sequana, Apollo Grannus and Sirona, Belenus, and Mars with Celtic epithets. These pages show the living public grammar of vow and cult. They are the room's best doorway into everyday religious action.

Then read the Gaulish inscriptions. Start with the Pillar of the Boatmen because its monument program is visually and religiously rich. Read Larzac when you are ready for a difficult magical source. Read Reze and La Graufesenque to remember that Celtic language appears in work and account as well as religion. Read the regional Gallo-Greek votive clusters for divine names and sanctuary traces.

After that, read Galatia. Strabo gives the cleanest political geography. Livy gives the Roman story of migration and campaign. Pausanias gives Greek memory of Galatian violence and Delphi. The rare Zeus epithet inscriptions give small but valuable religious traces in the Galatian-Anatolian world.

Then read Celtiberian and Hispano-Celtic. Begin with Novallas for a modern open-access control case in how far a public text can be understood and where translation must stop. Read Luzaga and the hospitality tesserae for pact and institutional relation. Read Penalba for the Luguei sanctuary anchor.

Finally return to the broader Introduction to Celtic Traditions. The Continental room should change how you read the whole Celtic shelf. It should make the medieval narratives feel less like isolated fantasies and more like one later layer among many. It should also make the inscriptions feel less dry, because they are often the closest surviving public acts of the ancient Celtic-speaking world.

X. What This Room Should Become

This room is already unusually valuable, but it needs continued growth in a disciplined direction.

The first need is stronger source-linking across every dossier. Each page should clearly state whether its source route is RIIG, RIG, EDCS, Heidelberg, Hesperia, PHI, Perseus, The Latin Library, a modern article, a local archive capture, or another control. The reader should never wonder where a line came from.

The second need is a dedicated Coligny calendar doorway. The calendar is too important to remain only a background reference. It deserves a page that explains what survives, what is reconstructed, what the month names and notations can and cannot prove, how lunisolar reconstruction works, and why modern ritual calendars should not overclaim from it.

The third need is a divine-name index that does not pretend to be a pantheon. Such an index should list name, source type, language, region, associated Roman name if any, source count or dossier path, and caution level. It should distinguish widely attested cults from one-off or damaged names.

The fourth need is a source-method page for interpretatio Romana. Readers need a plain explanation of why "Mars Lenus," "Apollo Grannus," "Mercury and Rosmerta," or "Zeus Bussurigios" cannot be read as simply Roman or simply Celtic.

The fifth need is an inscription apparatus primer. Ordinary readers should learn brackets, restorations, lacunae, uncertain letters, damaged readings, edition numbers, findspots, material, script, and why line display matters.

The sixth need is a tighter boundary with modern Celtic spirituality. Modern Druidic, Pagan, and reconstructionist use of Continental evidence should be discussed elsewhere or in clearly labeled modern-practice notes. It should not be silently folded into ancient translation pages.

This growth would make the room more than a collection. It would make it a school of source literacy.

XI. Standing Before The Broken Stone

Continental Celtic evidence does not give itself easily.

It gives a bull with three cranes and a damaged label. It gives a goddess of horses through Latin vows from soldiers and muleteers. It gives Mothers under names that refuse to become one abstract Mother. It gives a river goddess at a spring. It gives a magical tablet whose women can be partly named but not fully understood. It gives a bronze public law that can be recognized as important before it can be translated. It gives a Celtic language in Spain written with Latin letters. It gives Galatian tribes in Anatolia described by Greek and Roman outsiders. It gives gods under Roman, Greek, local, and Celtic names at once.

The brokenness is not a reason to look away.

It is the condition under which this part of the ancient world survived.

The reader who wants a complete system will be frustrated. The reader who wants only decorative Celtic atmosphere will be corrected. The reader who wants to prove a modern spiritual claim by force will find the evidence too stubborn. But the reader who can slow down will meet something more durable than fantasy: people making vows, keeping accounts, writing curses, marking time, naming gods, honoring springs, building monuments, regulating public space, entering empire, adapting scripts, and leaving traces strong enough to be read two thousand years later.

This is not a smooth mythology.

It is a field of contact between language, stone, empire, god, place, and loss.

Read it with patience. Let the fragment remain a fragment. Let the name remain a name until the evidence gives more. Let Latin be Latin and Gaulish be Gaulish. Let a Roman formula carry a local god without forcing the god to choose one identity for your convenience. Let a damaged line teach restraint.

The broken stone is not silent.

It simply refuses to flatter us.

Selected Sources and Further Reading