Chapter 11 - Napoleon Pulps His Enemies

Summary

Chapter 11 demonstrates that De l’Allemagne’s surviving 1810 texts are not identical, as had been thought. We have texts from all three proof runs. In Vienna sits a copy of the 1810 edition; the censors’ proof and the copy‑text for 1813 subsist. This makes a mockery of Napoleon’s efforts to obliterate the book, allowing a peek at the “lost” 1810 edition and tracing a remarkable interplay between four conflicting pulls on the author. Her desires to clarify imprecise or obscure passages, and to use key words from elsewhere in De l’Allemagne, confront her desires to be faithful to her sources and to the facts. Exerting its own pull on this interplay is the fierce pressure on Staël to tone down her polemic. These forced revisions fall in with her book’s slide from politics into literary history, which for two centuries now has dimmed the ringing attack on tyranny that caused its pulping.

Habent sua fata libelli.

Terentianus Marus, De litteris, De syllabis, De metris (c. 200 CE), verse 1286

This chapter focuses on Staël’s thousand-page chef d’œuvre, De l’Allemagne. A founding text in European and American Romanticism, it has the rare distinction of being a work of ostensible literary criticism whose entire first edition was pulped by government troops. On October 11, 1810, year of the Habsburg wedding, Napoleon sent his men to destroy it. He had never trusted Staël, exiled from Paris since 1803, but this time she had gone too far, ignoring his achievements, praising the Germans he had crushed at Jena and Austerlitz, and comparing him, the hope of France, to Attila the Hun. Printers and publishers were already under guard: The time had come to pulp the book. Its 5,000 copies – a vast print run – were therefore briskly transformed, as Staël puts it, into “un carton parfaitement blanc sur lequel aucune trace de la raison humaine n’est restée” (DA I 4/11). But the Emperor wanted more than to prevent the book’s release; he wanted to erase all trace that this book had ever existed. At the château where Staël was packing for further exile, more police arrived, demanding her proofs and manuscripts. Pen had met sword, it seemed, and lost.¹

During the next three years, rumor spread that something – le brûlé – had survived this eradication. In Germany, Johann Wolfgang von Goethe saw extracts from the book; Staël gave readings in Vienna and in Moscow, weeks before the city burned, on her circuitous flight to England in 1812. Rumors were confirmed on November 4, 1813: De l’Allemagne was released in London, the day news came of Germany’s triumph at the battle of Leipzig, and it sold out in three days.²

History has been sparing of whole editions lost in such bizarre circumstances, and we may be surprised to wait almost a century before hearing any more news of this infamous 1810 text. In 1907, a set appeared at auction. Since then, various other sets have been spotted; it has become, in fact, rather difficult to catalog them. The 1958 edition mentions four sets that hold a special interest: (1) in the château of Coppet, near Geneva; (2) in Paris; (3) in Vienna; and (4) in the château of Broglie, in Normandy. But in listing proof variants, the edition refers exclusively to the Broglie text. This means the loss of a whole series of delicate revisions, in 1810, as Staël played a subtle and dangerous game with the imperial censors. Indeed, as we look at these texts in more detail, it will become increasingly clear just how much did escape the Emperor’s police.³

Coppet: The Early Proofs

De l’Allemagne’s 1810 text was in three volumes. The château of Coppet, where Staël wrote her text, still has five volumes dated 1810: four copies of volume I (A/A’, B/B’), being double copies of two near‑identical printings – just sixty‑odd changes over 350 pages – and one copy of volume III, labeled C. These are the early proofs, while Staël, now at Chaumont and Fossé on the Loire, plans her departure for America: Their story is of an author’s last touches to her art, of dealing simply with friends and editors. Copy B’ is untouched except for one corrected misprint. A contains no Staël hand: A proofreader used this copy to correct misreadings of the final manuscript for the printers’ “B” run. A’ and C, however, contain over 2,000 unpublished revisions in Staël’s hand, along with over a hundred interventions and suggestions by three or four other readers. These revisions are identical in format, suggesting that the volumes are from the “same” first printing, though likely printed months apart. Here is proof of the sort of collaboration Staël encouraged on her greatest works.⁴

One reader, who uses pencil, may well be August Wilhelm Schlegel, who lived with Staël for fourteen years. She often simply crosses out this reader’s many ?s and !s in the margin, but four direct comments also remain. When p. 192 mentions “écrivains de génie (et quelle nation en possède plus que la France),” this provokes a brusque reply: toutes. Schlegel’s 1808 Vienna lectures spring to mind. On p. 209, a comparison between Friedrich Gottlieb Klopstock “et presqu’au même moment, Winckelmann dans les arts, Lessing dans la critique, et Goethe dans la poésie” receives a one‑word commentary: ensuite. Finally on p. 223 a translation of Klopstock, “la magie des plaisirs terrestres ne m’a point trompé,” is corrected thus: assoupi. Staël crisply deletes this remark, and none of these corrections caused a change in her final text; she encouraged friends’ advice without being its slave. In C, when p. 329 says that “plusieurs poètes lyriques, depuis Klopstock jusqu’à nos jours, ont tous dans leurs écrits une teinte de mysticisme,” the margin again adds one word: Novalis.

B contains a neat copy of Staël’s revisions in A’. The unknown copyist’s hand is none of her usual helpers – her children Auguste and Albertine, the governess Fanny Randall. Staël then reviewed these corrections, so minutely that her interventions would often be impossible to distinguish without the luck that made her choose a paler ink. She also made eleven changes of over one line in length; but when compared with the massive revisions on A’, it seems clear that De l’Allemagne was by now taking its final shape. Indeed, the “A” run closely follows Staël’s last manuscript, while B’s revisions are equally close to the censors’ proof. It seems that “A” and “B” are De l’Allemagne’s first and second proof runs, with the censors’ proof the third: There is no obvious reason to posit any lost printings interrupting this series.⁵

Bibliothèque nationale: The Censors’ Proof (BN)

It also seems certain that the Bibliothèque nationale text is the censors’ proof. We know that a copy was still with the censors when the police went in. This was illegal, and Napoleon later denied all knowledge of it, though Simone Balayé has proven his personal involvement. Joseph-Marie Portalis, who ran the department, was allowed to keep his copy, and the Paris text has a telltale P stamped in gold on the spine of its three volumes. When Portalis had the set bound, the book’s uneven pages were trimmed, and several marginal revisions were cropped and lost; a kind fate has left echoes of these lost phrases in the contraband sets the police pursued – in all a touching footnote to the book’s checkered history in imperial France.⁶

With no typesetters’ marks and nothing in Staël’s hand, this text was visibly not used by printer or by author; yet it has three sets of marginalia. Its title pages have these phrases, in red ink: “7 mai 1810 3898,” “E 8 août 1810 8272,” and “E 15 7bre 10,49[5 – p. 289].” In black, on III p. 369/gathering 24, are the words “3eme 20 7bre” in Gabriel-Henri Nicolle’s hand. Other ink is black. Volume I’s title page states, “Certifié conforme à la dernière épreuve de Mme de Staël. Signé: H. Nicolle, libraire.” – a sentence repeated after each gathering with two initials, HN and M for the printer Mame. Each page is also initialed HN. Volume II’s title page says “3eme corrigée” in Nicolle’s hand, with his signature, and his initials are on each page. Volume III repeats this format, reading “3eme 4 7bre” on its title page. Volume I is full of marginalia; in II and III, all changes but proofreaders’ corrections cease. All marginalia are in a similar hand, but for eleven black crosses in volume I – the mark of the censor.

These facts suggest various conclusions:

1. 1. Phrases in red give reception and registration dates for text. E may stand for the censor Joseph-Alphonse Esménard.⁷

2. 2. Dates in black indicate completion dates for text: Hence, the time lag September 4/15 for volume III is between completion and submission.

3. 3. Consequently, BN’s last forty pages finished printing on September 20.

4. 4. Also, if red ink marks registration, volume I’s marginalia were written before May 7, further antedating its printing. The Coppet C text probably postdates all this.

Here, then, is a rough timetable for De l’Allemagne’s composition and completion: “ten lines” of draft A composed, July 8, 1808; draft B begun, May 13, 1809; printing begins before April 11, 1810; volume I submitted to censors, May 7; final pages submitted to censors, September 20; printing ceases, September 25. Clearly Staël made notes before 1808; but two years of hard work, 1808–1810, turned those notes into literature.⁸

Bibliothèque nationale Marginalia

Among this text’s several hundred marginalia, whose vast majority are proofreaders’ corrections, stand Staël’s last‑minute revisions to her doomed book, after three manuscripts and two other sets of proofs; some of them run to several lines. The BN text therefore has two types of lost variants. Here, we review BN typeface deleted in its margin before censorship – just another author’s draft. The next section reviews phrases shown to the censors – the “official” 1810 text of De l’Allemagne. The Vienna text will then offer a third stratum, the “unofficial” 1810 text. “Last‑minute” is also a relative term; these revisions appear in volume I alone, thus dating from before May 7, 1810, a summer away from September. We must therefore judge with discretion a small but fascinating group of topical revisions. De l’Allemagne always was political, a sweeping attack on Bonapartist France; now it was leaving the printing house for the Emperor’s corridors. Did Nicolle wake up to his risks and ask Staël for these changes? Or did she read her work with a new mind as it left the private for the public sphere? One can only guess.

Some directly polemical remarks, in italics here, are simply deleted on the submitted page, an act made to please any censor’s eye: “Frédéric II … avoit su se faire aimer sincèrement dans son pays, et depuis qu’il n’est plus on le chérit autant que pendant sa vie: rare éloge pour un despote!” (I 229/15). Staël was probably wise to cut this veiled reference to Napoleon – though a sad loss. It was charming to see her thus menace her persecutor with his posthumous reputation. The Vienna chapter also lost a topical remark, which had survived to this late stage: “[C]omme on ne compte pas beaucoup d’écrivains distingués à Vienne, qu’on y lit assez peu, et que l’émulation manque dans la carrière des lettres comme dans toutes les autres, ceux qui la suivent ne se soucient guère du grand monde” (I 130/4).

This remark was probably deleted because of recent political events, which might in turn explain Staël’s odd 1813 footnote, stressing that her Austria chapter was written “dans l’année 1808.” The year 1805 had seen the Peace of Pressburg and 1806 Francis II’s abdication as Holy Roman Emperor. In 1809, Austria rose and was beaten once more; in March 1810, Napoleon married the Archduchess Marie‑Louise. Napoleon in 1808 had no reason to object to Staël’s words; in 1810, they would rub salt into Austrian wounds, and Napoleon might object strongly.⁹

A famous chapter in De l’Allemagne concerns the fête d’Interlaken of 1808. Ironically, this symbol of Staël’s “idyllic” view of Germany – a myth that earned her two centuries of abuse – was studded with digs at Napoleon, which Staël prudently removed when facing the censors. Thus, a passage comparing Germany and Switzerland: “Néanmoins la nature des gouvernements influe tellement sur l’esprit des peuples qu’une nation libre s’élève toujours à de certains égards au-dessus de la nation, son égale d’ailleurs, mais qui n’auroit pas les mêmes avantages politiques” (I 280/2). That is among the longest cuts made in 1810. Other deletions seem at first less clearly political: “Des ténèbres profondes environnoient ce point lumineux, et l’ombre de Berthold brilloit encore au milieu de la nuit comme un beau souvenir à travers l’obscurité du présent et de l’avenir” (I 287/14). Even before deletion, Staël’s remark was discreet, but she wore no gloves when describing the fête to Maurice O’Donnell on August 17, 1808: “[L]e temps était sombre et le nuage ressemblait au malheur qui pèse sur la tête de cette pauvre nation qui aime tant sa patrie.” Napoleon effectively controlled the country after 1803. Talk of vicissitude also had to go: “La simplicité des mœurs et l’attachement aux anciennes coutumes … rapprochent de nous le passé et nous rendent l’avenir présent. C’est un seul moment qu’une histoire toujours la même, et toutes les époques semblent encore présentes, quand nul événement, nulle vicissitude, nul changement ne servent à compter les années” (I 294/11).¹⁰

One last reference to recent Swiss affairs was removed from the chapter, not in 1810, oddly, but very late, before publication in 1813 – perhaps Staël felt it too obscure: “[L]’on chanta des vers sur le bonheur présent de la Suisse et sur la reconnoissance qu’elle devoit à la médiation qui le lui a conservé” (I 292/9). The word médiation carries a specific emotional charge: “The Act of Mediation (1803) brought Switzerland completely under the control of Bonaparte.”¹¹ Genevan by birth, Staël was doubtless aware of her wording. It seems clear, in short, that this famously idyllic description began as yet another of the skirmishes with Napoleon that recur throughout Staël’s 1810 text.

Two early additions (shown in square brackets) turn Staël from polemicist to historian: “le noble orgueil des Espagnols les a rendus [jadis] souverains d’une portion du monde” (I 40/3); “l’on craignoit à tort [dans le dernier siècle], en Autriche, que la culture des lettres n’affoiblît l’esprit militaire” (I 108/9). As Staël wrote in 1813, “je me supposois à cinquante années du temps présent” (I 3/4); that crude retreat into the past is precisely what her revisions concede to Napoleon. From the outset, outside force has pushed her book away from its chosen political arena.

Thus far, we have stressed “political” revisions, but nonpolitical changes (also presented in square brackets) appear on the censors’ proof as well: “Le genre lyrique se prête au vague, à l’infini dans les pensées et dans les sentiments: aussi la Messiade doit-elle être plutôt considérée comme un hymne majestueux que comme un véritable poème épique [Les païens, dans leurs poèmes, comme sur les bas-reliefs des sépulcres, représentoient toujours des tableaux variés, et faisoient ainsi de la mort une action de la vie; mais les pensées vagues et profondes dont les derniers instants des chrétiens sont environnés prêtent plus à l’attendrissement qu’aux vives couleurs de l’imagination]” (II 154/5). BN’s original argument was new to Empire France: Because lyric poetry tends to the infinite, Klopstock’s Der Messias is a hymn, not an epic. Genre labels are misleading for Staël, who here treats epic as a branch of lyric. Her margin’s argument about bas-reliefs and eschatology instead echoes Gotthold Ephraim Lessing’s 1769 essay Wie die Alten den Tod gebildet, equally new to Staël’s European public.

In another instance, Staël struggles with the place of sentiment in her philosophy: “les vérités primitives, celles que le sentiment et le génie peuvent seules saisir [saisissent], ne sont pas susceptibles de démonstration” (I 252/4). Either way, this is hardly a Cartesian statement; and if the use of reason was here banished from metaphysics, so a remark removed twelve lines earlier had cast sweeping doubts on reason’s role in almost all other areas of existence: “Une proposition en fait de chiffres est décidément fausse ou vraie; sous tous les autres rapports le vrai se mêle avec le faux d’une telle manière, que le sentiment [souvent l’instinct] peut seul nous décider” (I 251/9). Evidently Staël decided that her original polemical claim, that all life’s problems outside mathematics must be decided by sentiment, was after all overstating her sentimentalist case, at least for her Empire audience.

Vienna: The 1810 Edition

De l’Allemagne’s 1810 edition was thought destroyed. In reality, a copy survives, in Vienna’s University Library, brought there by the two Schlegels. Its most striking political statement is its ending: “Chaque fois qu’une nouvelle génération entre en possession de son domaine, ne croit‑elle pas que tous les malheurs de ses devanciers sont venus de leur foiblesse? ne se persuade‑t‑elle pas qu’ils sont” (IV 377/13). Here, at the foot of its volume III, p. 240, the text abruptly and forever ceases to speak – that is the hand of Napoleon, on September 25, 1810.¹²

Surprisingly, this final 1810 printing has yet another set of stop-press revisions, and it is odd to find this mysterious hand – no manuscript is known – still revising remarks the censors had never objected to. Who made these revisions, and are they a sop to the Emperor? One mild passage thus gains a concessive: “Ce seroit une vanité bien ridicule que de motiver [dans tous les cas] l’activité politique par le prétexte de l’utilité dont on peut être à son pays” (IV 311/5). Another change may evoke Napoleon’s recent divorce: “parce que les femmes brisent aussi souvent que les hommes les nœuds qui leur pèsent, et que les idées romanesques introduisent [les plus saints. La facilité du divorce introduit dans les rapports de famille] une sorte d’anarchie” (IV 374/13). Muddling its variants, the 1958 edition comments that “toute allusion au divorce pouvait attirer la censure impériale”; but Staël in 1810 did not replace the word “divorce,” she added it.

Yet most revisions here are not clearly political. In one instance, “la tour de Pise qui penche vers la mer [sur sa base]” (V 37/12), Staël may have heard that Pisa is six miles east of the coast, and its tower leans north. In this section, italics show variants absent from Vienna’s typeface, whose new revisions stand in square brackets.

Philosophy and Sentiment

Staël’s unremitting attack on received categories of thought is nowhere revealed in clearer outline than in the changes she felt obliged to make to her discussion of philosophy. Unlike 1810’s earlier, broadly political revisions, the massive changes here date from the Vienna copy, after Portalis had already approved the text: a bizarre series, and for the reader to judge.

Some forceful claims concerning philosophy and religion are simply removed: “Une philosophie superficielle rend incrédule; mais l’étude approfondie de l’univers fait avancer dans la connoissance de la divinité” (V 40/3). This claim denied final truth to every nonreligious philosophy, leading us to ask what Staël in fact understood by philosophie when she first wrote her book. Elsewhere, two similar remarks are cut within two pages: “aussi les Français ont‑ils eu dans le dix‑septième siècle des penseurs plus sévères que dans le dix‑huitième; des sentiments plus purs et plus religieux les animoient”; and a moment later, “ce mot d’utilité est‑il assez noble pour s’appliquer aux besoins de l’âme? Se replier sur soi‑même et concentrer sa force dans le sentiment et la conscience, est ce qui constitue la véritable grandeur de l’homme; à cet égard le dix‑septième siècle est plus philosophe que le dix‑huitième” (IV 59/17; 61/6). A third such remark two pages earlier escapes this pruning, to remain in the 1813 text: “la philosophie consiste à trouver l’interprétation raisonnée des vérités divines” (IV 57/3).

These variants had condemned empiricist philosophie by the yardstick of its objective truth. Yet Staël is always anchored in the real: “Les nouveaux philosophes … ont habilement flatté l’amour-propre de leurs adeptes, dans un noble but [et l’on doit les louer de cet art innocent]; car les Allemands ont besoin de dédaigner pour devenir les plus forts” (IV 390/10). Another variant thus calls scientism not only wrong but dangerous:

Les sciences présentées isolément … n’attirent pas les esprits exaltés, la plupart des hommes qui s’y sont voués … ont donné à notre siècle cette tendance vers le calcul qui sert si bien à connoître dans tous les cas quel est le plus fort; car les sciences ainsi conçues accoutument à tout réduire à des idées mathématiques qui ôtent la vie à la nature et le libre arbitre à l’âme.

(IV 271/1)

Politics clearly looms behind these debates.

The claim that mathematical ideas remove man’s free will needs explaining. If we are born with free will, our choice of career is not going to deprive us of it. Mathematics cannot remove our free will; at most, it might assert the thing’s nonexistence. Staël here claimed that the practice of science risked making us slaves, leaving critics to explain why she did so. And once again the problem seems in part linguistic; Staël is consciously expanding the normal context of given terms and ideas, in her bid to transform eclectic raw material into Romantic propaganda. A lost remark thus discussed the free will of sentiment, not a routine context for a precise philosophical concept: “Jacobi … proclame qu’il faut tout abandonner au libre arbitre du sentiment” (IV 343/23). The concept in fact recurs throughout De l’Allemagne, concerned as it is with contrasting the spontaneous upwelling of the soul from within – a sentimental phenomenon – and the rigid, quasi‑mathematical patterns imposed from outside: “il peut arriver qu’une conduite morale [parfaite selon le monde] vienne d’un mauvais principe” (IV 345/8).

Similarly, the BN margin had twice removed the word extérieur. Exterior is not a well‑delimited concept, but these examples may suggest some parameters: “on doit trouver même dans tout ce qui tient à l’extérieur [aux manières] plus d’usage du monde à Vienne qu’à Berlin” (I 240/2). Staël had used the same terms of France: “les expressions prêtent bien plus à la plaisanterie que les pensées, et dans tout ce qui tient aux mots l’on rit avant d’avoir réfléchi. L’extérieur de tout est mieux connu par les Français: cependant la beauté du style n’est point, on doit en convenir, un avantage purement extérieur” (II 16/9). It is fascinating to watch topical words, upon which Staël grounds sweeping metaphorical arguments, thus inserted or deleted from edition to edition – and to see what words or phrases she treats in her revisions as synonymous. For these references to exteriors echo De l’Allemagne’s North–South, classical–Romantic division of European character; different characters, traditional in themselves, are here being associated with broad geographical and cultural divides. Thus, a stop-press revision denies Gottfried Wilhelm Leibniz calcul, a French trait: “Leibniz avoit une supériorité aussi incontestable dans la connoissance du calcul et du monde réel que dans celle des opérations de l’âme [à cet égard]” (IV 94/16).

In Faust, Staël cut two words when Gretchen prays in the cathedral, tracing a delicate play between the dévote and the sentimentale: “elle prie avec ardeur; elle pleure; [,] et lorsqu’elle commence à se flatter de la miséricorde divine, le mauvais esprit lui parle” (III 108/17). Not only the number but the scope of these revisions may begin to give the impression that the text as written was concerned neither with philosophy nor with religion, nor even with art as such, but rather with all these things as seen through the prism of emotion. Compare this stop‑press revision to Staël’s famous chapter on classical and Romantic poetry: “ces dispositions de l’âme ne peuvent se faire voir que par … l’intérêt romantique enfin, qui varie sans cesse les tableaux: les sources de l’émotion [des effets de l’art] sont donc différentes à beaucoup d’égards dans la poésie classique et dans la poésie romantique” (II 133/21). This remarkable equation of emotion and artistic effect seems fuel for anyone wanting to call Staël an emotionalist literary critic. Elsewhere, beauty and virtue are treated as synonyms: “nous appelons utile ce que vous appelez beau [vertueux]” (IV 290/17).

Judging the cause of these odd stop-press revisions, two theories linger: Either Staël was herself ambivalent about her wilder claims, or she knew her Empire public and tempered the wind of change to that shorn lamb. Propaganda is fruitless if people won’t read it.

Kant

Staël’s chapter on Immanuel Kant was twenty-eight pages long in 1810. Seven stop-press revisions seem to reflect three conflicting pulls on the author: her search for precision; her dislike of neologism and pedantry; and her desire to link Kant to the great themes of her book. The small vocabulary available in classical French, which uses many words in a metaphorical sense, thus gives her a political metaphor at one point, designedly or not: “Kant démontre que nous ne pouvons rien concevoir que sous la condition de [toutes nos perceptions sont soumises à] ces deux formes” (IV 122/2).¹³

Throughout De l’Allemagne’s 1,200 pages, language remains highly charged. The book’s broad structural supports stand out in sharp silhouette against Kant’s own discourse – both very precise and rather dense, two obstacles for his French reviewer: “ces formes de notre pensée que nous imposons aux objets extérieurs [choses] … Les formes du raisonnement n’ont de résultat que quand on les applique au jugement des phénomènes [objets] extérieurs” (IV 123/7). The word extérieur is not Kantian, though common in De l’Allemagne, since for Kant we cannot hope to distinguish Phainomena inside or outside the skull. Staël often rejected that neologism:

Loin de rejeter l’expérience, Kant considère l’œuvre de la vie comme n’étant autre chose que l’action de nos facultés innées sur les phénomènes extérieurs [connoissances qui nous viennent du dehors] … les lois de l’entendement n’ont pour objet que les phénomènes [éléments] donnés par l’expérience. Il s’en suit que la métaphysique ne peut rien nous apprendre sur les vérités qui ne sont pas du ressort de l’expérience, et que c’est au sentiment que l’on doit en attribuer la prescience et la conviction [qu’au-delà de ses limites la métaphysique elle-même ne peut rien nous apprendre, et que c’est au sentiment que l’on doit attribuer la prescience et la conviction de tout ce qui sort du monde visible].

(IV 126/15–21)

Staël’s changes might indeed be minor, without the pattern linking them to grosser distortions of Kant elsewhere to serve Staël’s own ends. Thus, “Kant essaya de tracer les limites des deux empires, des sens et de l’âme, de la nature extérieure et de la nature intellectuelle” (IV 116), where these deux empires are not Kant’s but the Bible’s; behind Staël’s many “empires” stands Napoleon. Another passage combines systematic distortion with one concrete falsehood: “L’homme, à l’aide du raisonnement, n’a-t-il pas nié la liberté morale [le libre arbitre]? … si l’homme est libre, il faut que ce soit d’après des motifs qu’il se détermine, et ces motifs sont l’intérêt ou la vertu [doit se créer à lui-même des motifs tout-puissants qui combattent l’action des objets extérieurs et dégagent la volonté de l’égoïsme]” (IV 133/15). For Kant, intérêt is brute causality, and no motif. These terms – vertu, égoïsme, devoir – all distort Kant; yet that misses the whole point, since Staël is reworking Kant en bloc to support her own key themes from elsewhere.

In short, Staël reconciles Kant with her own religious and political concerns. In a final revision, she again decided she had stretched her source a little too far (interpolations appear in parentheses): “Les philosophes matérialistes … rejettent [placent] (le beau) ainsi dans l’empire des sensations … nous reconnoissons la beauté quand nous la voyons, parce qu’elle est l’image extérieure de l’idéal, dont le type est dans notre âme [intelligence] … nous échappons par l’énergie de notre âme [la vigueur de notre volonté] au sentiment de notre foiblesse physique” (IV 136/8–137/8). Two pages earlier, the BN proof had mentioned Kant’s “théorie abstraite sur la nature de l’âme [intelligence]” (IV 134/13); the two revisions intelligence and volonté illustrate something of the looseness of the word âme in De l’Allemagne.

Napoleon

There are eleven black crosses in the margin of the BN proof – the marks of the censor. They do not always correspond with the passages Staël lists as censored in 1813. The crosses occur as follows: I 23/3; 25/5; 62/3; 104/3; 122/17; 123/9; 145/1; 226/16; 227/10; 242/10; II 220/5. Not only are there unmarked passages that were evidently censored, there is also one marked passage, I 25/5, unchanged in the Vienna text. In 1813, Staël lists thirteen cuts: All but two of her footnotes occur in volume I. Thus, I 23, 166, 242, IV 307, V 230, with extended notes; I 62, 104, 123, 124, 145, 227, 228, II 220, simply noting censored phrases. I give in square brackets the postcensorship 1810 versions, with precensorship readings in italics. These censors’ versions come from the Vienna text, more evidence that it is not just another proof copy, as had been thought. The term Portalis identifies Vienna’s few departures from suggestions Portalis had made, in the letter Nicolle sent on to the author (I xxvii–xxx).

je vais exposer … des opinions étrangères à celles qui règnent en France … (mais) soit qu’on les adopte ou qu’on les combatte, elles donnent toujours à penser. Car nous n’en sommes pas, j’imagine, à vouloir élever autour de la France littéraire la grande muraille de la Chine, pour empêcher les idées du dehors d’y pénétrer [: parmi les productions du dehors il en est une dont l’importation doit toujours être permise, ce sont les idées].

(I 22/13)

Les hommes éclairés de l’Allemagne … abandonnent assez volontiers aux puissants de la terre tout le réel de la vie. Ce réel, si dédaigné par eux, trouve pourtant des acquéreurs qui portent ensuite le trouble et la gêne dans l’empire même de la pensée.

(I 62/7)

Il réussit momentanément … mais après sa mort il ne resta rien de ce qu’il avoit établi.

(I 104/1)

Les bases de l’édifice social sont bonnes et respectables; mais il y manque un faîte et des colonnes pour que la gloire et le génie puissent y avoir un temple [l’éclat et la splendeur qui pourroient en faire le temple de la gloire et du génie].

(I 122/17; Portalis – pourrait; was Staël’s prose style the problem here?)

la jeune impératrice, élevée dans ces temps cruels [si cruels pour sa famille].

(I 123/9)

L’ascendant des manières des Français [de la mode et de la littérature françaises] a préparé peut‑être les étrangers à les croire [les Français] invincibles.

(I 145/1)

Un homme peut faire marcher ensemble des éléments opposés, mais à sa mort ils se séparent [tôt ou tard ils rentrent dans leur direction naturelle].

(I 226/16)

la Pologne fut une conquête machiavélique, et l’on ne pouvoit jamais espérer que des sujets ainsi dérobés fussent fidèles à l’escamoteur qui se disoit leur souverain.

(I 228/2; Portalis – cut entire phrase)

Ce caractère en Prusse est plus noble et plus exalté que les derniers événements ne pourroient le faire croire, et l’ardent héroïsme du malheureux prince Louis doit jeter encore quelque gloire sur ses compagnons d’armes [; les nations comme les individus ne doivent pas toujours être jugées uniquement par leur histoire].

(I 242/8)

Le bon goût en littérature est, à quelques égards, comme l’ordre sous le despotisme, il importe d’examiner à quel prix il est acheté [Il importe d’examiner à quel prix on achète ce qu’on appelle le bon goût en littérature].

(II 220/5)

Staël also footnotes three passages unmarked in the BN text and unchanged in Vienna. Here they are:

Une femme d’esprit a dit que Paris était le lieu du monde où l’on pourroit le mieux se passer du bonheur.

(I 166/8)

Dès qu’on se met à négocier avec les circonstances, tout est perdu … la chose publique qui se passoit parfaitement d’eux.

(IV 307/6–308/2)

Oh, France! terre de gloire et d’amour! … la trace des torrents de sable, terribles comme les flots, arides comme le désert.

(V 230/5)

And here, Staël’s comments: “Supprimé par la censure sous prétexte qu’il y avoit tant de bonheur à Paris maintenant qu’on n’avoit pas besoin de s’en passer.” “Ce passage excita la plus grande rumeur à la censure. On eût dit que ces observations pouvoient empêcher d’obtenir et surtout de demander des places.” “Cette dernière phrase est celle qui a excité le plus d’indignation contre mon livre; il me semble cependant qu’elle n’auroit pu déplaire aux Français.” Staël had heard that the first and third phrase here might be censored; the second she perhaps invented.¹⁴

One last passage, starting at III 143/5, merits attention. Aimé Martin republished it separately in Paris in 1814, just before De l’Allemagne’s appearance: Le Portrait d’Attila, suivi d’une Epître à M. de Saint-Victor. He comments, “[T]el est le portrait d’Attila dans lequel le tyran de la France crut se reconnoître, et qui fit supprimer tous les exemplaires du bel ouvrage sur l’Allemagne de Mme de Staël Holstein.” But there is no change to this passage in 1810, and not a mark on Portalis’s copy – unless one counts the bookmark, at that very page for the two centuries it took to dent and stain it.¹⁵

Broglie: The 1813 Copy-Text

Safe in London, and facing a transformed Europe, Staël once again revised her text in 1813. Here the Broglie proof, her copy‑text in London, is especially useful. It seems – being locked up – to be a double of the BN proof, complete and without censors’ revisions, but with Staël’s own revisions and the words bon à tirer, dated September 1813, all in her hand. There is no evidence that any of her revisions here date from before 1813.

Despite Staël’s claim in her preface that “je n’ai pas cru devoir y rien changer” (I 3/6), her 1810 text now gained a series of additions – notably the twenty-one‑page preface itself, and a forty-five‑line footnote on Bentham, IV 294, part of Staël’s controversial attack on crude utilitarian laissez-faire. Thirteen footnotes showing censored passages add thirty-one lines, and six further footnotes add thirty-nine lines, as follows: I 63 on censorship; I 98 on Austria; II 31 on recent English poetry; III 340 on A. W. Schlegel; V 60 on the siege of Saragossa; V 217 on England. The vast majority of these changes keep the book topical, by stressing its suppression by the Emperor. Within the body of the text, changes are less striking. Few are longer than sentence length, except a fourteen‑line paragraph praising Joseph Priestley and David Hartley for accepting the will’s electric power over the body, added in 1813, starting, “Parmi les savants anglais …” and ending “… mécanisme” (IV 255/1–256/13). Two more revisions again concern Kant, the first a simple concession: “Kant voulut rétablir … la conscience dans la morale, et l’idéal dans les arts. Examinons maintenant de quelle manière il a rempli ces nobles [différents] buts” (IV 113/19). Later, Staël the sentimentalist replaces sentiment with sensibilité over two pages and adds a note to distinguish them:

[Il faut distinguer, en étudiant la philosophie de Kant, le sentiment de la sensibilité; il admet l’un comme juge des vérités philosophiques; il considère l’autre comme devant être soumise à la conscience. Le sentiment et la conscience sont employés dans ses écrits comme des termes presque synonymes; mais la sensibilité se rapproche davantage de la sphère des émotions et par conséquent des passions qu’elles font naître.]

(IV 324/10–325/18)

At least four of these changes seem directed toward the future and Staël’s English audience. But one last series of changes seems focused on the past. A paragraph on the baron Voght at I 276, altered to praise of “des philanthropes … à Hambourg,” looks at first like a local decision – yet Staël made eleven changes in fifteen lines, and a closer look at contemporary events suggests that Staël wanted to evoke Hamburg’s infamous pillage, after 1810, at the hands of Napoleon’s lieutenants.¹⁶

Conclusion

De l’Allemagne’s surviving 1810 texts are not identical, as had been thought. It seems likely that De l’Allemagne had three proof runs: If so, we have surviving texts from all three, in Coppet and in Paris, including two of the first printing’s three volumes. In Vienna sits a neglected copy of the famous 1810 edition itself. We have the censors’ proof, in Paris, and, in Broglie, the copy‑text for 1813. We have all three manuscripts. All this makes a mockery of Napoleon’s rabid efforts to obliterate the book.

These findings allow a peek at the “lost” 1810 edition, and at Staël’s long series of revisions in 1810 and 1813, including her extensive changes after censorship: Surprisingly, discounting the added preface and footnotes, the 1810 text seems longer, not shorter, than the 1813 text.

Finally, these lost 1810 variants trace a remarkable interplay between four conflicting pulls on the author. On the one side stand her desires to clarify imprecise or obscure passages, and to use key words from elsewhere in De l’Allemagne, rather than the neutral or affectless words her sources offer, as with Kant. On the other stand her desires to be faithful to her sources and to the facts. Separate, and yet exerting its own pull on this interplay, is the fierce – though ultimately fruitless – pressure on Staël to compromise, to tone down her polemic. Ironically, these forced revisions fall in with her book’s inevitable slide from politics and into literary history, a slide that for two centuries now has dimmed the ringing attack on tyranny that caused its pulping.