Language and the National Past in Napoleonic France: Reassessing the Académie celtique, 1805–1813

Abstract

This article reassesses the Académie celtique (1805–13), an antiquarian society concerned with investigating the origins of the French nation, and the progenitor of the current Société nationale des antiquaires de France. While most prior scholarship has focused on the Académie as an innovator in the field of French folklore studies, it is argued here that the Académie’s main scholarly concern was with language, specifically the use of historical linguistic research for tracing the origins of European nations. Underpinning the Académie’s research programme was a belief that the ancient Celts—based in Gaul—once dominated Europe and spoke a language from which most European tongues descended, and was perhaps preserved in modern Breton. The article therefore also presents a coherent view of the importance of Celtic ancestry for the developing French nation in the early nineteenth century.

I

On his election to the Institut national in 1797, Napoléon Bonaparte (1769–1821) declared that the most honourable occupation was the expansion of human knowledge: ‘true conquests … are those made over ignorance’.¹ Several years later, the Napoleonic state’s support of various learned bodies seemed to reflect this assertion, retouching Paris’ portrait as the intellectual capital of Europe after it had been tarnished by the suppression of scholarly societies during the Revolution. A wave of scholarship has recently mapped this intellectual terrain, with much focus falling on the foundation of the Institut national (1795) and particularly its second class, driven by the idéologues and concerned with establishing a ‘Science of Man’ based on the principles of sensationist epistemology derived from John Locke (1632–1704) via Étienne Bonnot de Condillac (1714–1780).² The pioneering anthropological Société des observateurs de l’homme (1799–1804) has also garnered considerable attention and is the subject of an important monograph by Jean-Luc Chappey.³ By comparison, the Académie celtique (1805–13), an antiquarian society concerned with investigating the history of France through linguistic, ethnographic and folkloric approaches, might seem old-fashioned.⁴ Yet a detailed examination shows its members to have been engaged in a new scholarly enterprise seeking to establish the historical provenance of the ascendant French nation. Most prior attention has focused on the folkloric interests of the Académie celtique, but it is argued here that the group’s primary concern was with language and its uses for determining the origins of nations and their genealogical relationships to one another (what would later be called ethnology). Inspired in part by Gottfried Wilhelm Leibniz (1646–1716), the Académie’s linguistic programme centred around the belief that the primordial language of Europe was ‘Celtic’, that most modern European languages descended from it and that Breton best preserved this ancient Celtic tongue. These insights helped to structure a revitalized conception of the French nation as rooted in a Celtic past, a view whose origins, political implications and the ways in which they were reflected in and bolstered by the Académie celtique’s scholarly practices—primarily linguistic—are the subjects with which this article is concerned.

According to its own records, the Académie celtique was dreamt up in the summer of 1804 by the philologist Éloi Johanneau (1770–1851), who enlisted the support of the diplomat Michel de Mangourit (1752–1829) and the administrator Jacques Cambry (1749–1807), érudit and author of the famous Breton travelogue Voyage dans le Finisterre (1799).⁵ The group quickly gained official sanction and, after initially meeting at the Louvre, settled in just across the Seine at Alexandre Lenoir’s (1761–1839) Musée des Monuments français.⁶ Members were elected and ambitious plans laid out: the Académie would concern itself not only with historical investigations into France’s Celtic, Gaulish and ‘French’ (français) pasts, but with that of ancient Europe more generally. In a memorable passage delivered at the first official session (30 March 1805), Johanneau linked the ancient Celts to the wider history of Europe and the newly constituted French Empire: ‘Almost all the peoples of Europe are descendants of the Celts, almost all are the children of the Celtic: newly reunited, they almost all form, still today, a single and great family under the same federal government.’⁷ But despite such grand rhetoric and the interest of a few prominent philosophes, scholars and politicians, the Académie failed to attract the attention for which it hoped and in 1813 overhauled its goals and expanded its field of enquiry into the medieval era, reconstituting itself during the 1814 restoration as the Société royale des antiquaires de France.⁸

The Académie has generated historical interest largely for two different reasons: as what some have seen as the institutional founder of French folklore studies, and as a barometer for measuring French nationalism, the early Bretonist movement and how the latter grew out of the former.⁹ As discussed below, the late nineteenth-century nationalist context played a crucial role in rehabilitating the memory of the Académie as a pioneering folkloric institution, and the historian of French folklore Nicole Belmont has further cemented this interpretation with a series of articles and essays published since 1975.¹⁰ In 1981 the distinguished historian of the Revolution Mona Ozouf better situated the Académie within its intellectual context in the first article to make use of the group’s minute book, but her investigation was again mostly interested in the Académie as institutional founder of French folklore and ethnography. Ozouf did, however, recognize the Académie as a manifestation of French nationalism and the multifaceted power of the adjective ‘Celtic’, vividly described as combining a ‘declaration of war, a faith, a programme’.¹¹ The huge amount of meaning invested in ‘Celtic’ has been more thoroughly prospected by historians of Brittany and French nationalism, namely Bernard Tanguy and Jean-Yves Guiomar, who have both contextualized the Académie celtique within the development of Breton national identity and its complex relations with France.¹²

Largely absent from the range of historiography concerning the Académie celtique is a serious consideration of its interest in language in both the scholarly and nationalist contexts, and how it served as a bridge between them.¹³ The Académie certainly has claims to importance beyond the histories of folklore and ethnography, especially in the histories of archaeology and art history; however, on its own terms it is perhaps best understood in the context of the development of the language sciences. One of the two primary goals the Académie set itself was to ‘study and publish the etymologies of all the languages of Europe’, which would shed light on the history of the Celts and their languages, the ‘Celto-Breton, Welsh, and Erse’ still spoken in the corners of north-western Europe.¹⁴ Its most eminent member, Constantin François de Chasseboeuf, comte de Volney (1757–1820), suggested that the Académie could complete Leibniz’s great encyclopaedic linguistic project, while contemporary critics also noted the Académie’s concern with language and alleged that its ‘guiding principle’ was the belief that Celtic was the first language spoken by humans.¹⁵ This linguistic focus—long-established in early modern antiquarianism—cuts against the popular narrative that interest in the Celts was down to a recent post-Ossianic Romanticism and Rousseauvian idealization of the Gauls as noble savages.¹⁶ On the contrary, the enlightened scholars of the Académie considered their Celtic ancestors to have been as civilized as their Roman conquerors, as proven by their possession of a sophisticated language and, supposedly, an alphabet.

Considering the Académie celtique in relation to the diachronic development of French linguistic scholarship adds a vector of understanding unavailable in a synchronic politico-cultural approach, such as that taken by Stuart Woolf, who overlooked the Académie’s methodological concerns and described it as a group of pedants with only a ‘cranky interest in whatever could be regarded as relics of an ancient Celtic society’ for the ‘propagandistic purposes’ of French imperialism.¹⁷ To be sure, empire provided a new justification for research into French and European antiquity, but as will be seen, the relationship between scholarly objectivity and nationalist bias actually operated in a complicated dialectic, in which scientific developments forced scholars to adapt their national apologetics to higher critical standards over time. Furthermore, nationally minded interest in the Celts had a centuries-long provenance, and as the best historical examination of the Académie has stressed, the Napoleonic interest in the Celts only makes sense with this history in mind.¹⁸ Section II therefore briefly outlines the development of French linguistic interest in Celtic subjects, before section III analyses the aims and practices of the Académie celtique itself. Section IV concludes with the Académie’s fall and the vicissitudes of its nineteenth-century afterlife.

II

An obvious starting point is to understand what ‘celtique’ meant around 1800. Today ‘Celtic’ as a label pertains to a group of ‘nations’ on the Atlantic edges of western Europe who self-identify as being descended from a distinct ancient people, but in the early modern era most of Europe claimed Celtic ancestry in some measure.¹⁹ Just as belonging to a linguistic group identified as ‘Celtic’ is central to our contemporary definition, linguistic understandings also dominated early modern ‘Celtic’ interpretations, as scholars sought to prove their national language was, or was descended from, the ancient ‘Celtic’ language of Europe (which suggested that its speakers also descended directly from the ancient Celts). In other words, it was believed that all or most European languages stemmed from a single parent language, often called ‘Celtic’, and if one could demonstrate that one’s language was, or descended from, this ancient language one could claim the historical prestige of the Celts for one’s own nation.²⁰ French, German, Dutch, English, Irish, Welsh and Scottish scholars (among others) at various points have claimed that their respective nation descended directly from the Celts and therefore preserved the cultural prestige of this people, including the ancient Celtic language thought to have once been spoken across the length and breadth of Europe. Only from around 1700 did ‘Celtic’ start to become associated more definitively with the British Isles and Brittany, and it was not until the nineteenth-century acceptance of the Indo-European linguistic paradigm that the idea of a European ‘Celtic’ parent language finally became extinct. For virtually all of the early modern period then, linguistic instability and scholarly uncertainty left the prestigious Celtic label to be claimed by any scholar with enough erudition, linguistic skill and, especially, hermeneutic ingenuity.²¹ This quest for linguistic and cultural prestige was a central preoccupation of the Académie celtique, which should be seen as a new manifestation of the long-running competition over the Celtic legacy.

As ‘Keltoi/Celtae’ was a label given to ‘barbarous’ tribes of northern and western Europe by the ancient Greeks and Romans, most early modern European nations could claim Celtic ancestry, but because Julius Caesar (100–44 bce) and Pliny the Elder (23/4 –79 ce) had recorded that one of Gaul’s three parts was called ‘Celtica’ (after its inhabitants who called themselves ‘Celtae’), France had particularly strong claims to Celtic heritage.²² Jean Bodin (1530–1596) made the important step of enlisting linguistic evidence as proof of the Celtic identity of the Gauls, the large extent of their European conquests and, controversially, their creation of the Germans: ‘the German language in great part is derived from the purest Celtic sources … it also follows that they owe their origin, arms, laws, and finally their culture to the Gauls’.²³ Throughout the seventeenth century, German scholars such as the Danzig-born geographer Philippus Cluverius (1580–1622), German vernacular champion Justus Georg Schottel (1612–1676) and Leibniz all retorted that in fact the Germans best preserved the ancient ‘Celtic’ tongue of Europe.²⁴

Yet in the quest for Celtic ancestral primacy French scholars had a linguistic trick up their sleeve: the Breton language, with which they became increasingly familiar through missionary activity. It was (mistakenly) theorized that Breton preserved the original Gaulish language, an idea reinforced in the first substantial Franco-Breton grammar, composed by the Jesuit missionary Julien Maunoir (1606–1683) and published in 1659. Ancient Gaulish, Breton and Welsh had become more firmly associated over the seventeenth century, but the strongest identification of Breton with the ancient Celtic language came with the work of the Breton Cistercian monk and chronologist Paul-Yves Pezron (1639–1706), who forced the consideration of Breton, and by extension Celtic, linguistic antiquity into wider contemporary debates over the age and nature of languages and nations.²⁵ Building on the work of Maunoir and Pezron, the Capucin Grégoire de Rostrenen (1667?–1750) and the Maurist Dom Louis Le Pelletier (1663–1773) compiled dictionaries and grammars that identified the Celtic language of Brittany as the root of modern French.²⁶

These seemingly arcane linguistic theories were of immediate political relevance for the eighteenth-century debates over France’s ancient constitution, as language was thought to best indicate national origins. The attention to Breton underscores the increasing interest in the Gaulish/Celtic heritage of the French nation at the same time as a debate over the Latinity of the French language coincided with the famous ‘Germanist’ versus ‘Romanist’ dispute that took place in and around the Académie des inscriptions in the first half of the eighteenth century. ²⁷ After the theories of Henri, comte de Boulainvilliers (1658–1722), presenting the idea of a violent Frankish conquest—in support of the idea of limiting monarchical power—were posthumously launched in the late 1720s and early 1730s, the Abbé Jean-Baptiste Dubos (1670–1742) responded with a large history arguing that the Gallo-Romans had allowed the Franks into France, where the two peoples merged relatively peacefully.²⁸ To take one example that mixed the Romanist-Germanist debate into a moderate argument for the Latinity of the French language, Charles Pinot Duclos (1704–1772) argued that the Gauls and Germans were both of Celtic descent, the upshot being that it did not matter where the Franks came from as they spoke a ‘Celtic’ language and therefore did not have to force it on the people—there was therefore more cultural assimilation than conquest.²⁹ Montesquieu (1689–1755)—who critiqued both Dubos and Boulainvilliers in the final books of De l’esprit des loix (1748)—influentially supported the Frankish origins of a limited monarchy, but the republican Abbé Gabriel Bonnot de Mably (1709–1785) probably best reflects the prevailing pre-revolutionary consensus in accepting a Frankish entrance and mixture with the Gauls, rather than a violent conquest and enslavement of the Gallo-Romans.³⁰

In the final decades of the eighteenth century there emerged a number of linguistic antiquarians who have become known as ‘Celtomaniacs’ for the grand claims they made for Breton as the repository of the ancient Celtic language. The prominent Huguenot philosophe Antoine Court de Gébelin (c.1725–1784) attempted to recover the primitive natural language as a sort of linguistic key to all knowledge in his nine-volume (unfinished) Le Monde primitif (1773–82).³¹ He posited that this original natural language was Celtic, from which descended ‘Gaulish, Runic, Teuton, Greek, and Latin’ along with the other languages of Europe.³² The most notorious ‘Celtomaniac’ was a Breton parlementaire named Jacques Le Brigant (1720–1804), who became a source of problematic inspiration for the Académie celtique. Bent on claiming the European prestige of the Celts specifically for his native Brittany, Le Brigant’s earliest works focused on proving that Celtic/Breton had once been spoken from ‘le Cap Finisterre … jusqu’à l’Hellespont’—an assertion not hugely out of the ordinary for the time—but eight years later he expanded this idea to argue that all world languages derived from Celtic, and included comparisons to the Semitic languages, Chinese, Sanskrit, Galibi (Carib) and Tahitian.³³ Treated as a curiosity by many and mocked by some, Le Brigant and his ideas were nevertheless well known in learned circles and he corresponded with cosmopolitan philosophes and provincial érudits alike. The marquis de Condorcet (1743–1794) saw the historical potential in his project, and if a contemporary English biography is to be believed, then Le Brigant met Louis XVI at Versailles, who bestowed him with a pension and joked ‘Voilà le plus savant homme de mon royaume’; more certain is that he was awarded 3,000 livres by the Convention at the end of 1794.³⁴

If, as Guiomar suggests, theories of ethnic origin were not a factor in the outbreak of the Revolution, we can nevertheless detect ways in which they were useful for revolutionary polemicists, who occasionally cited the Celtic linguistic theories to bolster their historical frameworks.³⁵ For example, the founder of the Girondist Cercle social, Nicolas de Bonneville (1760–1828), described the ancient Celts as a warlike, free and democratic people with weak kings—an obvious example for a constitutional monarchy. He praised the contributions of Le Brigant and also recommended the project of Court de Gébelin.³⁶ Another Girondin-associate, Jacques-Antoine Dulaure (1755–1835), wrote several scathing anti-nobility tracts, depicting the Germans as a race of barbaric thieves who subdued the Gauls through treachery.³⁷ As a future member of the Académie celtique, he would publish a paper reviving Dubos’ old contention that a Gaulish senate had existed independently of Rome—the obvious implication being that Gaul’s native Celtic government was republican.³⁸ The Abbé Grégoire (1750–1831) demonstrated a grasp of the Celtic theories and cited Le Brigant in his famous 1794 linguistic report to the Convention.³⁹ Michel de Certeau, Dominique Julia and Jacques Revel found that the concept of Celtic ‘obsessed’ the respondents to Grégoire’s 1790 linguistic survey, on which the 1794 report was built, and that the theories of Le Brigant and other Celtic theorists can be found sprinkled throughout the responses Grégoire received.⁴⁰

One of Le Brigant’s students, Théophile Malo-Corret de La Tour d’Auvergne (1743–1800), wrote a history of Gaul focusing on the Breton language, which went through three editions by 1801. La Tour d’Auvergne cautiously distanced himself from some of Le Brigant’s excesses and dispensed with the biblical descent from Noah’s family—making only one oblique historicist reference to the Ark in Armenia—but was nevertheless bullish on Breton as the mother language of Europe.⁴¹ There was one key linguistic contribution: the realization that Basque was neither related to the Celtic languages nor to any of the other European languages (this is today’s scholarly consensus).⁴² La Tour d’Auvergne published a new edition of his major work in 1796, upon returning to France after a year as a prisoner of war in Bodmin, Cornwall.⁴³ Reworked for the increasing nationalism of the French Republic, the book more than doubled in length and featured a more inclusive title, Origines gauloises. It struck the right tone for the insecure republic, and in its review the Moniteur declared that ‘Latour d’Auvergne no longer belongs to himself; he belongs to the Patrie.’⁴⁴ La Tour d’Auvergne received congratulatory letters from the Directors Louis Marie Réveillère-Lépeaux (1753–1824) and Lazare Carnot (1753–1823) for his military and scholarly service to the patrie.⁴⁵ Following La Tour d’Auvergne’s death in 1800 at the Battle of Oberhausen, an 1801 edition of Origines gauloises was released with a eulogy by Mangourit, who linked La Tour d’Auvergne’s career as a soldier to his scholarship—depicting national antiquarianism as heroic in itself was particularly expedient for this future co-founder of the Académie celtique, who made La Tour d’Auvergne its symbolic first member. More prosaically, La Tour d’Auvergne made an indirect contribution to the Académie celtique by bequeathing his library to the group’s future permanent secretary Éloi Johanneau, whom he had also taught to read Breton.

Although this discussion has been necessarily impressionistic, several points are hopefully evident. The first is that a long-established tradition of historical enquiry into the Celtic national origins of France and Europe more generally existed at the dawn of the nineteenth century. The second is that much of this research was carried out in a linguistic mode thought to be the best method for proving the historical relationships of different nations. Finally, the intertwined debates on language and history factored into wider political debates about France’s constitution and underpinned a strengthening French national consciousness, as well as a related Breton consciousness, a relationship that strengthened under the empire.

III

The members of the Académie celtique willingly took up the project of further establishing the Celtic origins of the French nation, a view given new historical significance after the rejection of Frankish ancestry during the Revolution, and the establishment of the empire. It was natural, according to the Académie’s 1807 dedication to its patron Empress Josephine (1763–1814), that Napoleon’s decade of victories ignited the desire to recover the glories of the nation’s ancestors, summarized by its motto ‘Gloriae majorum’ (Glory of the Ancestors).⁴⁶ Joseph Lavallée (1747–1816)—author of a travelogue depicting a republican Brittany—made the connections between imperial past and present clearer:

The idea to found a Celtic Academy must naturally owe its birth to the epoch where French glory draws all attention to it … The country in which we live was the metropole of [the Celts], whose surplus population colonized many distant lands … If the Celts were the ancestors of Bellovesus and Sigovesus, are we not still today these same Gauls whose forefathers followed these illustrious voyagers?⁴⁷

The former colonies of this ancient ‘Celtica’ were now being reunified ‘under the same federal government’, according to Éloi Johanneau.⁴⁸ Within France, the Celtic past also provided a historical thread with which to sew together the patchwork of territories into the modern French nation: appealing to a shared past allowed for plurality in the present. Given that civil war between Paris and much of provincial France—especially vicious in Brittany and the west—was a recent memory, invoking a common Celtic heritage that seemed to be especially accessible through the Breton language made the historical argument for cohesion particularly significant. La Tour d’Auvergne’s selection as a sort of immortal member was a shrewd choice of talisman, simultaneously aligning the Académie with the military glories of the empire while flattering Breton pride. Considering its membership underlines the point.

When it began meeting in 1805, the Académie celtique drew a range of figures under its broad umbrella of patriotic interest in France’s past, and by 1807 it listed 316 Parisian, national and foreign members. Of the founding trio, Johanneau was a professor and érudit who became an imperial censor in 1811, Mangourit had travelled widely as a diplomat under the Convention and the Directory, while Cambry was a bestselling author, man of letters and governmental administrator.⁴⁹ Alexandre Lenoir, the group’s second president, had almost single-handedly saved many of France’s important cultural monuments from revolutionary vandalism and displayed them in his Musée des Monuments français—the meeting place of the Académie celtique—which presented a chronological national history through objects and architecture.⁵⁰ There were several former Girondins, including Mangourit, Volney, Dulaure and the former Director La Révellière-Lépeaux. Another Girondin associate was the Académie’s third president, the jurist and orientalist Jean-Denis Lanjuinais (1753–1827), who had been president of the Convention just before Thermidor and was an imperial senator. The Académie also accumulated some impressive foreign members, including Paul-Henri Mallet (1730–1807), Christian Gottlob Heyne (1729–1812), both Humboldts (Wilhelm [1767–1835] and Alexander [1769–1859]), and eventually Jacob Grimm (1785–1863) and Friedrich Carl von Savigny (1779–1861), both of whom joined in 1811. The cosmopolitan Italian scholar Carlo Denina (1731–1831), former librarian to Frederick the Great (1712–1786) also joined the Académie in 1805, having relocated to Paris the previous year at Napoleon’s behest. The Académie was therefore not an assemblage of professional pedants or parochial Casaubons, though some of their more provincial members were in this mould. Scholarly interest in the French past sat alongside and supported strong beliefs about the present, and the large number of administrators involved underlines Stewart McCain’s point that the Académie’s interest in collecting the language and customs of the French peasantry was part of a broader attempt by the empire to harness statistics for state-building purposes.⁵¹

Religion was not generally one of the strong beliefs discussed; it was explicitly forbidden to debate this by 1809, though the subject was often touched upon obliquely in studies of mythology.⁵² Most of the members were deists, while Mangourit and La Révellière-Lépeaux were resolutely anticlerical. It is worth noting the complete absence of the trellis of Christian sacred history, which had supported virtually all prior, and especially Breton, research into Celtic history and language. The abandonment of this framework was an issue for the royalist journalist Mathieu Guillaume-Thérèse Villenave (1762–1846), who rebuked Johanneau (a proponent of solar mythological theories)—‘we cannot be anti-diluvians … Genesis is our rule of chronology and of faith’—but joined the Académie and briefly acted as its secretary nonetheless.⁵³ Not all Catholics were able to reconcile their religious beliefs with the Académie’s approach. The nonjuring priest and Breton grammarian Alain Dumoulin (1748–1811) rejected his election in a searing letter to the ‘académie prétendue Celtique’ noting that he ‘trembled with horror’ reading their Mémoires: ‘I will never be part of your society … I am a Catholic priest, and I will die a Catholic priest. I detest your principles …’⁵⁴ However, Jean-François le Gonidec (1775–1838), a devout Catholic, possible Chouan, and the foremost Breton grammarian of the early nineteenth century had no qualms about taking a leading role in the Académie’s final years, while the revolutionary Archbishop of Besançon (and La Tour d’Auvergne’s former mentor), Claude Le Coz (1740–1815), warmly welcomed his election to the Académie, proudly declaring ‘I am a Celt, I confirm it.’⁵⁵ Generally, the group was a homogenous collection of elites: its members were all educated men, most worked for the state in some capacity, many were Freemasons, and the society was disproportionately populated by those who identified as Breton.⁵⁶ But there were differences, and a focus on shared Celtic heritage could actually cut across social, religious and geographical divides.

What then was the Académie’s idea of the ancient Celtic past, and how was it reflected in the group’s practices? Of course there were differences of opinion, but reading through its minute book—updated at its bimonthly meetings—and its published Mémoires, it becomes evident that the Académie celtique’s collective historical vision essentially reprised the early modern idea of an ancient Celtic Europe and gave it a new nationalist gloss.⁵⁷ That vision maintained that the Celts had once ruled nearly the entire continent (including much of Asia) and that Gaul was the centre of their territory; it also posited that the modern French nation descended from these Celtic Gauls, who spoke one of the oldest European languages, which was now preserved most completely in Breton and its linguistic relatives in the British Isles. Cambry noted with alarm in 1805 that ‘After centuries of confusion, modern peoples want to usurp the name of the Celts, and to appropriate for themselves all the glory of these ancient conquerors of Greece, Rome, of all Europe, and almost all of Asia.’⁵⁸ It was part of the Académie’s goal to restore France’s exclusive claim to the Celtic legacy and to rehabilitate the view of the ancient Celts as a powerful and refined civilization, a fact obscured by a disproportionate scholarly focus on Greece, Rome and, more recently, Egypt.⁵⁹ The research programme undertaken by the Académie would ‘prove that the Gauls were more civilized and more advanced in the arts and sciences than has been thought’, announced Johanneau.⁶⁰ But there was one problem: according to Caesar, ancient Celtic culture had been predominantly oral—the Druids (the Celtic priestly/intellectual class) were supposedly forbidden from recording their doctrines in writing—so as far as anyone knew there were no surviving Celtic or Gaulish written records, which meant that it had been easy for Greek and Roman authors to efface the history of Celtic Europe and slander Celtic society as barbarous.⁶¹ How, then, to retrieve this lost Celtic past? It was here that the Académie’s investigations into language, folklore and antiquities came into their own. In the absence of recorded textual evidence, the academicians employed a set of innovative techniques devised to conjecturally reconstruct the history of ancient Celtic society. While posterity has identified the group’s pioneering folkloric practices as of particular significance, contemporaries and the scholars of the Académie itself—following the long tradition of Celtic linguistic antiquarianism already outlined—placed most emphasis on language.

The Académie set itself two broad goals—to write the history of the ancient Celts, and to do so primarily through a comparative linguistic project—which remained stable though their wording occasionally changed. Johanneau provided the most complete formulation:

1^o. to recover the Celtic language in ancient authors and monuments; in the two dialects of this language that still exist, Breton and Welsh, and even in all the popular dialects, the patois and jargons of the French Empire, as well as the origins of languages and place names, of monuments and customs which derive from them, and to give dictionaries and grammars of all these dialects, whose inventorisation must be hastened before their total destruction

2^o. to collect, write, compare, and explicate all the antiquities, all the monuments, all the usages, all the traditions; in a word, to make a statistical study of ancient Gaul (faire la statistique antique des Gaules), and to explain ancient times by modern times.⁶²

As Nicole Belmont has noted, for the Académie ‘monuments’ stood for much more than just physical remains of the past.⁶³ Cambry and Johanneau outlined this in their Monumens celtiques (1805), which used the menhirs (standing stones) of Carnac as a jumping-off point for the investigation of ancient Celtic history more generally. Although Cambry’s focus was ostensibly on the standing stones, these physical monuments were only the raw material to be examined using ‘the true science of etymologies’, which would see ‘a new geography born in France’ through the comparison of place names.⁶⁴ Johanneau made the point more clearly: ‘words are precious monuments when we can understand them, because through them we can recover the moeurs, habits, religion (le culte), and finally the history of a people, even if they have left no other monument than the names of places, people, and things’.⁶⁵ Whether in physical monuments, folklore or mythological remnants, ‘words are the most durable and most numerous monuments of the most distant times’.⁶⁶

Académie scholars constructed an illustrious linguistic genealogy, identifying themselves as continuing the tradition of Cluverius, Leibniz and Pezron, among others, and taking William Jones’ (1746–1794) Asiatic Society of Calcutta (f. 1784)—the birthplace of the Indo-European idea—as a model for scholarly practice.⁶⁷ The Académie’s approach to language was shaped chiefly by Johanneau, who published by far the largest number of linguistic articles in the Mémoires and also, crucially, corrected the linguistic mistakes of others in reviews and edits. His approach to language study can be summarized as follows: etymology—long derided but little understood—is, if approached correctly, a ‘science’; the correct scientific methodology was not to search for the origins of all words in one particular language, but to place words within their own linguistic contexts.⁶⁸ Where many previous, and especially Celtic, scholars had gone wrong was to trace all words to a pre-selected favoured language, rather than comparing the vocabularies of different languages:

Those who have compared a great number of languages, know that there is in all of them a common base held with the first needs of man, with onomatopoeias, and which constitutes a truly primitive, common, and universal language. The folly of etymologists has been to search for it in a single language, such as in Hebrew, Celtic, Flemish, etc. etc.⁶⁹

The former Director La Révellière-Lépeaux, who contributed fairly regularly to the Académie’s journal, described a similar understanding of language in an 1806 letter to Cambry:

In the infancy of societies, there is a language of signs as there is an oral language. The same sign, as well as the same word, is applied to objects that are very different from one another … It is only gradually that ideas become understood and language perfects itself to designate each material, or intellectual, object by a monument, or, by an expression which isparticularly appropriate to it.⁷⁰

Although Johanneau and La Révellière-Lépeaux differed on the arbitrariness and conventionality of early linguistic signs, the point here is that both passages show a naturalistic understanding of the origin of language, and therefore a distinct break with the Franco-Breton tradition of Pezron, Le Brigant and, to a lesser extent, La Tour d’Auvergne, which had relied on a divine origin and the biblical framework of descent from Noah. Johanneau’s paragraph shows that once this shibboleth had been dismissed Breton became particularly valuable as one of the earliest known languages, rather than for anything intrinsically sacred about it. Indeed, in a review criticizing the linguistic missteps of the Ragusan ambassador, Antun Sorkočević, comte de Sorgo (1775–1841), Johanneau claimed not to have a predilection for Celtic, or any other language, merely ‘a love for truth’.⁷¹

Linguistic research had to be corroborated with the widespread study of material antiquities and mythology. Johanneau’s investigations into alphabets provide an example of the way these different approaches supported each other. The search for the origin of alphabets was a common antiquarian pursuit, but was particularly important in the Celtic nations, where the discovery of a native alphabet would be the smoking gun that would prove the sophistication of Celtic civilization before Roman incursion. Spanish research into Basque therefore became of interest to Johanneau, who published a French translation of the antiquarian Juan Bautisto Erro’s (1773–1854) Alfabeto de la lengua primitiva de España (1806), which sought to prove that ancient inscriptions on medals and coins were written in an early Basque alphabet and therefore that Basque was the universal language of the ancient Celtiberians (the inhabitants of ancient Iberia named in classical sources). Despite taking issue with many of Erro’s etymologies, Johanneau agreed with his alphabetical thesis but sought to contextualize it within the historical development of ancient European letters more generally. For Johanneau, Greek was the parent alphabet of Celtiberian, Celtic, Runic and Etruscan (or ancient Latin), rather than Phoenician. This seemingly mundane conclusion had a powerful upshot for the initiated, which requires some explanation.

In his Gallic Wars, Caesar recorded that the Druids had no alphabet of their own but could write in Greek characters when occasion required. However, Caesar specified that these characters were not the modern Greek letters that he could read, but an older form—perhaps, antiquaries surmised, the characters he had seen were an adapted Celtic alphabet? For Johanneau, the exciting conclusion was that if Gaulish monuments were compared alongside the ancient Greek-derived Celtiberian letters in Erro’s book, it would be possible, with the aid of the Breton and Welsh languages, to finally ‘read and to decipher all the inscriptions of Gaul in characters regarded as unreadable and unknowable until now … this discovery, so desired, so often hoped for by the best scholars of Europe, opens a very vast field for the reform of history’.⁷² However, for Erro, and undoubtedly for other antiquaries interested in the question of a Celtic alphabet, the most important conclusion to be drawn was that ‘Spain was a lettered nation many centuries before the era of its monuments’ and before it was invaded by ‘foreign nations’; in other words, the native Celtiberians—and by extension their Celtic relatives in Gaul, Britain, Ireland and elsewhere—were just as enlightened in this arena as the Romans and other ancient civilizations. In addition to underlining the hunt for evidence of native pre-Roman refinement, the interest in alphabets also reinforces the holistic approach to antiquities that characterized research by Académie scholars, in which the importance of physical antiquities was multiplied in a comparative linguistic matrix. Johanneau made the point elsewhere:

[T]he origins of words make known to us the origins of thoughts and habits of peoples … A good etymology of an altered word of obscure and difficult origin is worth more than a badly conserved and well explained coin. It’s under this relationship that the science of words has analogy with that of antiquities, and must not be foreign to those who occupy themselves with them.⁷³

The Académie’s linguistic project was also advanced by Volney, the group’s most high-profile contributor, who was apparently briefly offered the presidency in March 1806 but turned it down.⁷⁴ Although he is primarily remembered as an orientalist, Volney had demonstrated an interest in the Celts as early as his 1795 lecture series at the École normale, Leçons d’histoire, where he indicated that he understood ‘Celts’ and ‘Gauls’ to belong to a relatively homogenous northern European culture, and that research into these ancient Europeans had stagnated due to a fixation on ancient Greece and Rome. Scholars had only recently started to move beyond them with methodologies—like linguistic comparisons—that would bring about a ‘revolution’ in understanding ancient history. However, ‘work of this kind cannot be executed by one individual … There is wanting one numerous society, which … would methodically follow every branch of one identical plan of research.’⁷⁵

A learned group such as the Académie celtique was therefore ideal for such a project, and in his address to it ten years later Volney essentially regurgitated these points while elaborating on a work that could serve as a model, Peter Simon Pallas’ (1741–1811) Linguarum totius orbis vocabularia comparativa (1786–9).⁷⁶ Compiled at the behest of Catherine the Great (1729–1796), Pallas’ dictionary featured wordlists of two hundred languages collected by questionnaire.⁷⁷ The Celtic dictionaries of Rostrenen, Le Pelletier and Gébelin featured prominently.⁷⁸ While overall Volney took a favourable view of the work, he criticized its use of the Russian alphabet—rather than the ‘European or Roman’—as lacking the phonetic flexibility needed to record the true sound of the words, and suggested that the Académie celtique sponsor a prize for the transliteration of the work into Roman script. But, in order to compile a true polyglot dictionary worthy of the name, it would be necessary for a learned academy to establish a polyglot commission and begin collecting data from an expansive network of correspondents; of course the Académie celtique, Volney wheedled, seemed ‘destined to fulfil this mission, honourable to the nation’.⁷⁹ Although a committee was immediately established to fulfil this goal, little came of it, but Volney’s address also brought some notice to the newly founded Académie; it caught Thomas Jefferson’s (1743–1826) attention, for example, and was reprinted in the Moniteur.⁸⁰ There was also some discussion around this time, led by Denina, of changing the remit of the Académie to a broader ‘académie philologique’.⁸¹

Beyond Johanneau and Volney, the most prominent active linguistic member of the Académie was Lanjuinais, who served as president from 1809–10. A committed comparatist, Lanjuinais translated and reported on relevant sections of the German J. C. Adelung’s (1732–1806) huge comparative dictionary, Mithridates (1806–17).⁸² The Académie also featured the leading Breton scholar of the first half of the nineteenth century, Le Gonidec, who in 1807 published the most important French–Breton grammar for a century.⁸³ He dedicated it to the Académie celtique, eventually becoming a vice-president of the society, and placed several of his research papers in the Académie’s Mémoires. Published between 1807 and 1812, the more than two thousand pages in the six volumes of the Mémoires de l’Académie celtique testify to Johanneau’s industriousness as both editor and the author of many articles, in which language featured prominently. Of the five classes involved in editing the journal, two were directly concerned with language, while approximately one-quarter of the articles were concerned directly with linguistic questions; but this far undersells the issue, as most pieces drew on etymology as a corroborative technique in some measure.⁸⁴ The contemporary press also noted the Académie’s preoccupation with language. The Magasin encyclopédique—edited by A. L. Millin (1759–1818), a member of the group—described its primary goal as researching ‘primitive languages’ and determining the antiquity of the Celtic language, and when the Moniteur chose to report on the Académie’s sessions they were primarily those that touched on linguistic issues, such as the presentation by the musician Anne-Pierre-Jacques Devismes (1745–1819) on his ideas on music as a universal language.⁸⁵

Because they have been amply discussed elsewhere, the Académie’s pioneering folklore researches need only a brief comment. Almost as soon as the Académie was founded, the need for innovative research methods was identified as necessary to ‘supplement historical gaps and shine new light on the shadows which cover the birthplace of the Gauls’.⁸⁶ A committee comprising Cambry, Johanneau, Denina, Dulaure and the geographer Edme Mentelle (1730–1816) directed a questionnaire composed of fifty-one questions to be sent out to the different communes of France. It appears from the minute book that this committee arose around the same time as Volney’s ‘polyglot’ commission, which never came to fruition but might help to explain why there is little space devoted to language in the famous Académie ‘folklore’ questionnaire.⁸⁷ A number of local dignitaries responded with details of customs preserved in their communes, which were printed in the Mémoires and usually accompanied by Johanneau’s endorsement that such traditions were of Druidic origin.⁸⁸ Just as Lenoir’s museum preserved the physical monuments of the French past, the questionnaire was designed to record and therefore preserve knowledge of the more instable ‘monuments’ of language and thought retained by the French peasantry. That most of these traditions were believed to extend from Gaulish times, and might preserve Druidic rituals, added to their significance and underscores the point that, despite posterity’s heavy focus on the folkloric research of the Académie, it was one component of an integrated approach to antique studies that stressed linguistic research above all.

IV

The Académie celtique advanced a complex view of the ancient Celtic past and developed sophisticated methods to reconstruct and analyse it. Ultimately, however, this focus proved unsustainable and the group metamorphosed into the Société royale des antiquaires de France between 1813 and 1814. Why? One prominent explanation blames an excessive exuberance for things Celtic: ‘Celtomanie’, or a fanatical belief that the ancient Celts were the first inhabitants of Europe, spoke its first language and that all European nations descended from them.⁸⁹ This term was coined at the beginning of the nineteenth century and has remained pejorative ever since.⁹⁰ However, the word was birthed by Académie scholars and first appears in the group’s Mémoires; the perpetual secretary Johanneau himself used it, indicating a level of self-awareness denied by the group’s detractors.⁹¹ As these final paragraphs will suggest, ‘Celtomania’ is an unsatisfying answer to the question of why the Académie celtique expired, but identifying how this explanation arose sheds light on the combination of hostility and misperception, which, set against the background of political change and the newly recontested historical interpretation of the French past, seem to have been the real reasons for the Académie’s demise.

A fatal conflation of the Académie celtique with some of its overenthusiastic ‘Celtomaniac’ predecessors, and most of all with Jacques Le Brigant, plagued it from the beginning. Only a year after the Académie’s founding, the journalist René Tourlet (1756–1836) noted the existence of those who wrongly supposed its goal to be to prove ‘that the Celtic language is the primitive language of the Universe, and that all peoples descend from the Celts’, revealing already a malicious and false rendering of the Académie’s purposes.⁹² Tourlet may have been thinking of the geographer Conrad Malte-Brun (1775–1826), who sneered, ‘if the Académie celtique is to be believed, Adam spoke Breton, and Eden was Quimper-Corentin’.⁹³ Malte-Brun, claiming to be ‘impartial’, went out of his way to kick the Académie several times over the next few years, though in 1810 he added the caveat that Johanneau and Lanjuinais should not be considered among those trying to trace all languages back to Breton.⁹⁴ While Malte-Brun may have been relating his experience of the Académie’s meetings, or of the second-rate scholarship of Agricol-Joseph Fortia d’Urbain (1756–1843)—also frequently criticized by Johanneau—the Académie’s Mémoires reveal a continuous effort to react against the tendencies that Malte-Brun alleged that it promoted. There is in Malte-Brun’s reviews more than a hint of Celtophobia, derived from readings of Tacitus, Montesquieu and the Scottish geographer John Pinkerton (1758–1826)—who considered the Celts a separate race inferior to the Goths—as well as his own Dano-Gothic patriotism.⁹⁵

It is a great irony that Académie scholars coined the epithet ‘Celtomanie’—which almost immediately came to be used against them—in order to critique the excesses of their predecessors, from whom they tried to distance themselves. On the dawn of the Académie’s formation, Johanneau specifically targeted Gébelin, Gilles Déric (1726–1800)—‘estimable as a historian, pitiable as an etymologist’—and the two most over-exuberant Celtic linguistic antiquarians, Le Brigant and Pierre Bacon-Tacon (1738–1817), a Jacobin secret policeman who published two volumes arguing that the south-eastern region of Bugey was the cradle of the Celts and the original language.⁹⁶ Importantly, Johanneau—though he respected Le Brigant’s zeal—did not spare him censure: this ‘célèbre celtomane’ had ‘abused the Celtic language’ by seeking to explain all languages through Breton, and did not produce one etymology admissible by Johanneau.⁹⁷ Johanneau repeated the charge three years later, this time ‘admitting with sorrow’ that his old friend and mentor La Tour d’Auvergne had also fallen into the trap of patriotic zeal for ‘the language of his ancestors and his childhood’, and had ‘the same mania, a mania fateful to science’.⁹⁸ But for all of La Tour d’Auvergne’s enthusiasm, he had also made the crucial determination that Basque was not related to Breton, a conclusion drawn ten years before that of Wilhelm von Humboldt and apparently cited favourably by him.⁹⁹ The Académie’s leaders also accepted this distinction and—at a time when a first-rate thinker like Volney conflated the languages of Wales, Brittany and Gaul with ancient German—developed a theory of European linguistic relationships that not only differentiated between the Germanic and Celtic families, but situated them within the wider European language family, even though Celtic’s membership of the Indo-European language family was not proven until the 1830s.

Whereas eighteenth-century ‘Celtomaniacs’ had argued that not only French but all other European languages descended from Celtic, Johanneau argued that if French bore resemblances to Breton it was because Celtic and Latin had both developed from the same root—perhaps Greek, as suggested by Erro’s Celtiberian alphabet. If French preserved a number of words whose etymology derived from Breton, it was likely down to borrowing as much as filiation. This was certainly a break from the excesses of the eighteenth century, and if a distorted belief in the influence and importance of the ancient Celts and their language commonly characterizes the work of Académie scholars, it is perhaps more the result of misconceptions about ancient and early medieval history—due in part to a lack of evidence—rather than a dogmatic adherence to the primacy of the Celtic language. Embracing the ancient Celts had more to do with French national prestige in the present than some sort of fixation with the Celts for their own sake: reason was not completely warped around an obsession with the ancient Celts, as ‘Celtomanie’ implies.

However, the misperception was evidently commonly held, and ‘Celtomanie’ provided a convenient scapegoat in 1814, when the Académie was renamed and its programme became more archaeologically based in a conscious rejection of the previous dominance of etymology and a general linguistic approach. At the session of 18 October 1814, the new secretary Antoine-Jean Saint-Martin (1791–1832) noted that from its origin the Académie celtique had adopted a ‘singular and bizarre system’, with ‘audacious suppositions’ and ‘absurd etymologies’, complaining of its ‘mania for rendering all parts of the ancient world Celtic’—Saint-Martin was an orientalist.¹⁰⁰ He was also a Christian and a monarchist, and proclaimed the new society’s loyalty to the monarchy; after surveying the destruction of ancient Europe by pagan, barbarian tribes—the Celts included—Saint-Martin celebrated the fall of the French Empire, which had threatened to plunge Europe back into barbarism once more. Similar criticisms were voiced and magnified by the influential German linguist A. W. Schlegel (1767–1845), who reacted so strongly against ‘Celtomania’ that for decades he refused to grant the Celtic languages a place in the Indo-European family.¹⁰¹ More prosaically, money problems also afflicted the Académie, which, despite its semi-official status, relied on private financing. In 1809 Johanneau started sending out threatening circulars reporting the unanimous vote that those unpatriotic enough not to finance the publication of the seventh volume of the Mémoires would be struck from the membership, but in the end, there was to be no seventh volume, which is probably telling.¹⁰² The change of direction may also have had something to do with the fact that Johanneau became an imperial censor in 1811, and was not among those listed when the society agreed to shift focus in 1813.¹⁰³ There also seems to have been waning interest in the ancient Celts yet perpetually strong interest in the ancient classical civilizations—there is no better example than Johanneau himself, who spent the rest of his career becoming a respected classical textual philologist and translator. Interest in the classical civilizations, and especially their archaeological remains, was stressed with the reconstitution of Société royale des antiquaires. With the peace, its leadership also stressed a desire to link up with other antiquarian societies across the continent and to investigate Europe’s ancient history without claiming it as the territory of the ancient Celts, which it had alleged to be the purpose of the Académie celtique.¹⁰⁴ The Académie’s perceived Celtomanic ‘système’ and its political implications were an easy scapegoat in this context; and perhaps such forceful scapegoating was necessary as Lenoir’s genuinely popular Musée des Monuments français was closed by the monarchy in 1816, despite its name change to the Musée royal des Monuments français.

To condemn the Académie for some of its ultimately wrong-headed linguistic assumptions misses the point that these were developed for the purposes of history-writing where no alternative evidence existed. While there was a historical-political purpose in proving the enlightenment of the ancient Gauls, this had more to do with French national patriotism than a singular obsession with the Celts. As Guiomar notes, a distinction should be made between Celtomania and Celtophilia, a historical interest with the Celts for nationalistic purposes, akin to other contemporary attempts made by scholars linking their nations to ancient Rome or Greece.¹⁰⁵ Johanneau was quick to criticize the work of correspondents who he believed stepped beyond the boundary of verifiable science, and by disavowing the works of eighteenth-century authors like Le Brigant and La Tour d’Auvergne, he sought to place the Académie celtique at the forefront of a more scientifically sound approach to the study of national origins, a trend that would continue in the more ethnologically focused history that developed over the first half of the nineteenth century. In 1812 he published a particularly scathing éloge of Le Brigant, who, like the Irish ‘celtomane’ Charles Vallancey (c.1725–1812), he alleged had done more linguistic damage than good.¹⁰⁶ It is worth noting that in attacking these figures, Johanneau’s criticisms were not far from those of someone like Rasmus Rask (1787–1832)—indisputably one of the founders of modern ‘scientific’ comparative and historical linguistics—who Alderik Blom has suggested fell victim to ‘Celtophobic’ obstinacy, which may have precluded his ability to recognize the Celtic languages as part of the Indo-European family.¹⁰⁷ The Italian Carlo Denina had also in 1804 published a three-volume Clef des langues, in which he criticized Pezron and Gébelin, and opted for a lost Scythian parent language instead of Breton as the original language of Europe.¹⁰⁸ This might lead us to reflect that, while ‘Celtomanie’ was certainly alive and well in the scholarship of some provincial antiquaries, this was not really the case at the Académie celtique under Johanneau’s stewardship, with support from other high-calibre scholars like Denina, Volney and Lanjuinais.

Even with the nineteenth-century professionalization of disciplines like history and comparative philology, political imperatives continued to shape accounts of the past—arguably, the political motives even strengthened. It was in this ascendant age of nationalism that A. W. Schlegel fought against the inclusion of the Celtic languages within the Indo-European family, and penned a vulgar poem ridiculing French linguistic scholarship and in particular the ‘Celtomanes’ Gébelin, Pezron and ‘Lebrigand’, who lacked ‘method, historical criticism and philological tact’.¹⁰⁹ But here he fell into a similar trap, as his prejudices against Celtic writers ossified into a recalcitrant position against the Celtic languages themselves—an opinion that he had to embarrassingly renounce after the repeated interventions of the Genevan Adolphe Pictet (1799–1875), whose father was a non-resident member of the Académie, and J. C. Prichard (1786–1848), who proved the Indo-European credentials of the Celtic language family in the 1830s.¹¹⁰

However, nineteenth-century French nationalism ultimately brought the Académie celtique’s reputation back full circle. Its eventual rehabilitation came at the hands of a pioneering French folklorist—Henri Gaidoz (1842–1932)—who argued that the Académie celtique deserved to be reconsidered as an integral player in the early development of folklore studies. There were clear nationalist overtones to this argument: Gaidoz aimed to undercut German claims to the establishment of folklore as a scientific discipline. In fact, he maintained that the systematization of folklore began under the Directory and continued during the French Empire in the Académie celtique, as manifested in the 1807 departmental questionnaire on local customs. This questionnaire had supposedly inspired ‘Jacques’ Grimm, who had been elected a member of the Académie in 1811; therefore, according to Gaidoz, the most eminent German folklorist recycled French methods established by the Académie celtique when he undertook his own path-breaking work.¹¹¹ Grimm had attended at least one meeting of the Académie during his tenure in Paris, but Gaidoz’s nationalism is thinly veiled and slotted into several decades of anti-German critique.¹¹²

Birthed during the early stirrings of French nationalism, the Académie celtique sought to connect the modern French nation with an idealized ancient Celtic past, a legacy claimable through the long-established tradition of linguistic research; but the Académie’s nationalist underpinnings were twinned with a genuine attempt to investigate the past scientifically. Ideas of language were therefore the cord splicing together the Académie’s scholarly purposes with its conception of the French nation and its place within European history. Viewing the Académie as the product of eccentric interest with the ancient Celtic past for its own sake or, on the other hand, as driven by craven political necessity is to miss half of the story in each case.¹¹³ National bias ran through the work of all the scholars examined in this article; these scholars nevertheless had to conform to contemporary scholarly standards in order to be accepted as scientifically valid. This complicated relationship means that ‘Celtomania’, while useful in some sense as shorthand, is also damaging by its misdiagnosis of interest in the Celts as a quixotic obsession with the imagined glories of the ancient Celts, rather than as a politicized historical stance. It was the tension between the shifting national and scholarly imperatives that largely determined the shape of the Celtic national past advanced by the members of the Académie celtique and other scholars over the eighteenth and nineteenth centuries.

He thanks both institutions for funding the research for this article. An early version was presented at the Society for the Study of French History annual conference at the University of Strathclyde in 2017. Thanks to the audience there for their comments.

Footnotes