Dumas, Alexandre

Le Comte de Monte–Christo [The Count of Monte–Cristo]

Brussels: Alph. Lebègue, 1845-1846. First Edition. 15 vols. bound in 7. Contemporary 3/4 cloth, marbled boards. The true first edition of the most popular novel of the nineteenth century — and very probably the rarest first edition of any novel that could rightly be called a classic of fiction. The first fourteen volumes dated 1845 and the last dated 1846, exactly mirroring the serialization in the Journal des Débats. No copy of this edition has sold at auction in over fifty years. It is not recorded precisely by either Reed or Munro in their Dumas bibliographies. No other set like it has been located. Contemporary 3/4 cloth, marbled boards, slight wear in a few places: vols. 1–2 front hinge and rear joint split but holding, vols. 9–10 hinges split but holding, vols. 13–15 rear joint split but holding; otherwise all seven books are near fine, clean, complete, and sound, with all fifteen half-titles, and they have never been repaired in any way.

Monte Cristo is an exemplar of what it means for a nineteenth-century first edition to have wide appeal — for seven generations the most popular novel across nations (le livre des livres) — and this is a once-in-a-generation copy of it, dropped into a twenty-first-century marketplace that has been crop-dusted with the bibliographically wrong, the misdescribed, and the misunderstood. Near fine. Item #999

The Bibliography
Anybody who isn't confused doesn't understand the problem.

Le Comte de Monte–Cristo (misprinted "Christo" in its earlier editions) has been smothered in bibliographical quicksand. All the bibliographies are wrong; what most readers of them fail to realize is how wrong. One lie doesn't cancel one truth — it cancels the truth. For just one example, the first three entries for Monte–Cristo in Munro's generally competent, and often admirable, bibliography give dates that could not be accurate, none of the three have been verified as even existing exactly as Munro describes them, and his fourth entry, while accurately described, has at least one volume misdated by the publisher (see below). Some sources get some things right, but the correct facts have been so diluted with so many mistakes that consulting the references can only lead the inquisitive reader astray. As to sets listed online, you could drive a caravan of motor homes through the messy descriptions of them and never scrape the truth. What we will do here is strip away the misinformation so that what is material can be considered more clearly.

NOTE: All bibliography is given as pending new discovery, but it has been 177 years and only the naïvely buoyant continue to wait when there is nothing left to wait for.

What We Know
We can start with some facts of which there is no doubt. Dumas had Monte–Cristo serialized from August 28, 1844, to January 15, 1846, in the Journal des Débats, the leading literary newspaper of the period, which had evolved from publishing the debates and decrees of the National Assembly during the French Revolution. Dumas published it that way because, in the 1840s, that was where the money was.
Some references correctly note that partway through the Journal des Débats serialization, the newspaper Le Siècle began a second sequential issue on September 28, 1845, issuing one part per week as a supplement for eighteen weeks, then concluding the novel on February 1, 1846, with one enormous final printing of the novel's last two-thirds — two weeks after the Journal des Débats serial ended.

Parallel to the Journal des Débats serial, four Brussels publishers (Alph. Lebègue, Meline, C. Muquardt, and Société belge de librairie Hauman) each published book editions sequentially, in a total of eleven confirmed variations. Two Paris publishers (Pétion and Baudry) did the same. All six publishers created their editions by copying the text from the journal installments as they appeared. All six picked up the novel a few months into its serialization, and all six issued at least one edition that more or less kept pace with the Journal des Débats publication, with each succeeding volume published shortly after enough of the novel had been serialized to fill it.

Most volumes are therefore dated before 1846, when the journal serialization ended in mid-January. The four earliest Brussels editions vary in size from eight, ten, twelve, to fifteen volumes; the two earliest Paris editions were in eighteen volumes. Since the last journal installment was published on January 15, 1846, all accurately dated book editions must have at least their last volume dated 1846. The one confirmed book edition (Pétion) with all its volumes dated 1845 has, at minimum, its last volume misdated by the publisher — but we will analyze Pétion's edition weighing factors other than the precedence of title-page dates (see "Sixth" below). All references to, and reports of, other editions with all their volumes dated before 1846 are unseen, unconfirmed, and either misread by the bibliographer or misreported to the bibliographer, and can be eliminated as spurious phantasms.

What We Are Looking For
Once the non-existent ghost books have been removed from our study, what we are seeking is a multi-volume edition that follows the progression of the Journal des Débats serial and mirrors its dates most closely to its conclusion. That means all of its volumes would be dated before 1846 except for the last, which would be dated 1846 and published in January. If we trust the dates on the title pages when they are possible — there is no need to trust dates that are impossible — we can begin to determine who actually got the entire novel published first. And in the real world, publication of the entire novel to its conclusion defines publication of the first edition.

First: We can dispense with the misprinted "Christo" for "Cristo" as a point of issue, because it was not corrected for the first time until late in 1846, after numerous book editions had been published, most of them reprints, including many translations.

Second: We can eliminate all editions dated 1846 throughout as having not been progressively published alongside the serialization — they were published later, sometimes months later, and the wise collector, librarian, or bookseller would never buy one.
Third: We can reject all editions that have some volumes dated 1845 but with more than just the last volume dated 1846, knowing they would trail any edition that parallels serial publication more precisely.

Fourth: That leaves us looking for a multi-volume edition with only its last volume dated 1846, exactly matching the timing of the Journal des Débats serial. The only confirmed edition matching this criterion is this one from Lebègue, offered here. Lebègue's other editions of Monte Cristo, in fewer volumes, are later reprints.

It is in fifteen volumes, with just the last volume dated 1846. Lebègue's edition is ranked as the first of the Brussels editions by Munro in his bibliography (Alexandre Dumas Père: A Bibliography of Works Published in French), though he gets one date wrong. It is rare because the print run was small, and the individual volumes were only for sale while the corresponding part of the serialization was ongoing, each volume lateral to it. As each volume's first printing sold out, Lebègue followed the fifteen-volume edition — beginning in mid-1845 — with reprinted sets also issued progressively but condensed into twelve, ten, or eight volumes, with more text per volume, renumbered and redated title pages, and more than just the last volume dated 1846.

So, for example, when the first six volumes in the fifteen-volume set had been issued and the first three had sold out, Lebègue continued issuing the later volumes for customers who had begun with it, but also started a reprinted edition that combined volume 1 and part of volume 2 into a new volume 1, with subsequent volumes similarly condensed. For a short while the book was for sale both ways. Lebègue's editions of fewer than fifteen volumes were the same size as the other 1845–1846 Brussels editions, with two exceptions: (1) A Brussels edition from Muquardt, in fifteen volumes, not examined, but which Munro lists twelfth — so assuming he actually saw a set, or even if it was only reported to him, it must have characteristics marking it as not only later but much later. (2) An auction at Christie's (September 14, 2021) of another Brussels edition in fifteen volumes whose description was miscataloged by Christie's. The other Brussels editions, in twelve, ten, or eight volumes, had the earlier among them with some volumes dated 1845, but every one had more than just the last volume dated 1846 — usually the last two or three — and all are later.

Fifth: One reason our Lebègue edition is so rare is that the original purchaser would have had to buy it in fifteen volumes, one at a time, over a year, and with dedicated promptness — buying each volume while it was still available, because if they missed one and it had sold out, they could only buy that part of the novel in a new compacted edition, with different title pages and later dates. The other three earliest Brussels publishers listed by Munro (Meline, Société belge de librairie Hauman, and C. Muquardt) are the only other Brussels publishers recorded to have issued Monte Cristo sequentially, and all of their known editions had more than one volume dated 1846. The editions from all other publishers outside France, regardless of language, also had more than one volume dated 1846, and all surely follow our Lebègue edition.

The Paris edition published by Baudry had its last two volumes dated 1846 and thus, if the dates are correct (and Baudry's usually were), it also follows our edition. This is not at all unexpected.

Sixth: The Paris edition published by Pétion is trickier. What has emerged as the real first printing of it has all eighteen volumes dated 1845. Its supporters are stubborn and difficult to persuade, but we are going to provide evidence for placing it after our Lebègue edition.

Two facts we know: Pétion registered his first two volumes in the Bibliographie de la France on November 23, 1844, which may be why some references wrongly say that volumes I and II of his published edition are dated 1844. But in three sets examined (Paris, Edinburgh, London), all eighteen volumes are dated 1845. Pétion either skipped registering volumes III and IV, or their registration was mislaid. He registered volumes V through XII between July 19, 1845, and December 20, 1845, and then never registered the last six volumes.
The registration of volumes I and II in 1844 says either that those registered copies had an 1844 title-page date, or that the two registered volumes were postdated to 1845 — but the registered copies no longer exist, and no published copy of either volume dated 1844 has been proven to exist. The confusion traces to a misreporting to Munro of the Parran copy, sold at the Hôtel Drouot in 1921, claiming volumes I and II were dated 1844. But Drouot's 1921 catalog states clearly that all eighteen volumes were dated 1845. Munro's citation of a Pétion edition with volumes I and II dated 1844 — his first edition — can be rejected.

More important, and of critical significance: the registration of volume XII on December 20, 1845, implies that Pétion did not have volumes XIII through XVIII in hand to register at the end of 1845, making belief in the accuracy of the 1845 dates on the last six volumes reasonably doubtful and certainly unverified. If Pétion had any of them printed on December 20, he probably would have registered them along with volume XII. Let us accept, for the moment, that volumes XIII through XVI, or XVII, were actually issued in 1845 — there is no positive proof that they were not. We would still have no reason to accept that the 1845 date on volume XVIII reflects its publication date, since the text for it was not available to Pétion until the serialization concluded on January 15, 1846. No rational argument for the authenticity of that date has been posed. And one more peculiarity: volumes XII, XIV, and XV of Pétion's edition carry an advertisement for Dumas' Le Vicomte de Bragelonne on their back wrappers — a book whose serialization did not begin until October 20, 1847.

Now let us approach Pétion's edition from a different direction, to preempt future discovery and assuage any lingering doubts. Suppose we accept that the Pétion sets we have verified — all volumes dated 1845 — were issued paralleling our Lebègue edition, with volume XVIII the only one issued in 1846 and simply misdated. Which of the two editions would have preceded?
No publisher's records beyond the registration dates address this question, but surrounding evidence convincingly separates them. Munro's bibliography lists Lebègue's fifteen-volume edition as the first of the Brussels editions, and this priority has not been competently challenged. The only edition he lists ahead of it is the 1844–1845 Paris ghost from Pétion — which he says he did not see — with two unseen dates and an unproven 1845 date on the last volume. This edition has been confirmed not to exist. Munro further lists as Pétion's second edition (ranked fourth overall) the issue with all volumes dated 1845, and that is what seems to be the actual first Pétion edition, of which three sets have been verified.

Dismissing the ghost edition, our Lebègue edition is the one Munro lists as the real first edition of Monte Cristo. And what if Pétion was setting his edition from Dumas' manuscript rather than waiting on the journal's January conclusion? He was not. Minor typographical errors in the Journal's serialization appear in Pétion's edition (and all the other earliest editions) but would not have been in Dumas' manuscript — proof enough that Pétion, along with everyone else, was copying text from the journal as it appeared. Further, the serious money for Dumas in the mid-1840s was in the serialization, not the book editions, and his contract with Pétion would have prohibited publishing any part of the text as a book before the journal had issued it. So despite the 1845 date on Pétion's volume XVIII, it was certainly issued after January 15, 1846, and because Pétion was in no hurry (see below), the same may be true of volume XVII — and imaginably, even volumes XIII through XVIII, those not registered by December 20, 1845.

It is worth stating clearly here that none of the Paris publishers were ever preoccupied with beating the Brussels editions to market. The customers of the Paris publishers, for their small editions of a few hundred copies, were the wealthy, who were willing to wait two weeks or a month to get their wide-margined, larger-type copies on fine paper, and were not interested in pocket-sized Brussels issues on ordinary paper with smaller type (though the Brussels type was better spaced than the journal type). However, two of the Brussels publishers — Lebègue and Société belge de librairie Hauman — were patently interested in being first to market, and between 1844 and 1847, one or the other successfully published their editions of all of Dumas' major novels — all eight of them — before the other Brussels or Paris publishers, in every case, beginning with Lebègue's issue of the true first edition of Les Trois Mousquetaires in 1844, for which there is no debate about priority.

From Les Trois Mousquetaires in 1844 through Les Quarante-Cinq in 1847, those two Brussels publishers published the real first book edition of every important Dumas novel serialized in newspapers or journals. In fact, from Le Chevalier d'Harmental in 1842 to Le Collier de la Reine in 1849–1850, no Paris edition ever preceded one or another of the Brussels editions. Faith in the priority of Pétion's edition of Monte–Cristo — because all its volumes were dated before 1846, even though at least the last was misdated — is like faith in the light around Tinkerbell, with that light's power source based solely on children's willingness to believe in it. There is no reason, and no precedent, to suppose that either Paris edition of Monte Cristo preceded Lebègue's.

This may not seem, to the most skeptical reader, to be documented proof that Lebègue's Monte–Cristo will forever stand as the true first edition, but the only imaginable doubt would be someone reporting an as yet unseen and unconfirmed multi-volume Brussels edition that also has only its last volume dated 1846, stimulating a reexamination almost certainly followed by a quick dismissal upon physical viewing. That possibility is only what one might imagine, and nobody has reliably held up any such edition for inspection in the 177 years since Monte–Cristo first appeared as a book. What we have here is real evidence, and logical evidence, and there is no evidence whatsoever to the contrary.

Seventh: Who were the buyers of all those later Brussels editions — the ones in fewer than fifteen volumes, with 1846 dates on more than one title page, or on all of them? Among them were customers who preferred reading novels in book form rather than in journals with tiny, crowded type, and who preferred reading all at once rather than with week-long breaks between journal chapters or month-long breaks between sequential book volumes. Others, while content to read it in a journal, wanted to own the book but had a bias against buying incomplete editions progressively. Still others wanted to buy their set complete so they could bind all the volumes at once and uniformly, the way they wanted. And some simply did not want to visit the bookseller repeatedly.

Conclusion
We have used title-page dating to narrow the candidates for Monte–Cristo's priority to the only Brussels edition confirmed to have been published in a time sequence exactly matching the serialization to its finale. We have drawn on history to confirm that among the Brussels publishers, Lebègue was one of the two who cared about — and successfully achieved — getting their editions of Dumas' serialized novels published first of all. Every other fact reinforces the bibliography's stated priority for Lebègue's edition preceding all other Brussels editions.
Taking a wider view, we have additionally separated Lebègue's edition from the Paris editions using historiography — the element of it that asks, what else was going on? — to demonstrate that, in the 1840s, the Brussels book editions of Dumas' great serialized novels always preceded the Paris editions, that the Paris publishers were never aspirants for priority, and that no Paris publisher ever got, or cared to get, their edition of any of Dumas' important 1842–1849 serialized novels published first.

Rarity
We explained under "Fifth" that our Lebègue edition had to be bought one volume at a time, with dedicated promptness, over a year, so it was hard to assemble and few sets were. No other set like it has been located. It is not listed exactly as it is in either Reed's or Munro's bibliographies. No other copy is recorded as having sold at auction. Although OCLC's record of library holdings lists dozens of individual volumes of Monte–Cristo by various publishers, and some sets too, many are misdescribed, and none are described as being both complete and having only their last volume dated 1846.

Of course there will be another set somewhere — hiding in a national library, a university library, in another likely place, an unlikely place, in grandma's attic, or in the backwoods — but it hasn't been held up for inspection yet. And even if another copy does appear, or two, or three, or four, our book would still be of the utmost rarity. So if you want what will almost certainly stand all tests and sustain as the undisputed first edition of Le Comte de Monte–Cristo, you had best weigh up our book here and now, because the doors of opportunity seldom slam open, and you are not liable to see another copy soon. Or maybe again.
Our book is the rarest first edition of any novel, by anybody, that could rightly be called a classic of fiction.

Two Asides
1. The Pétion Paris edition is today a $300,000–$600,000 book, depending on binding, condition, completeness, and quality, because there will always be some collectors — once a majority, now a declining minority — who want the first Paris edition of any Dumas novel, even if a Brussels or other edition is the true first and precedes it. A mixed set (ex–Daniel Sickles) was preempted by the Bibliothèque nationale de France at a Paris auction in November 1989. Another mixed set sold at Christie's Paris in 2010 (aspects of it seem to have been inaccurately cataloged). In 2022 a correct set surfaced in the trade and sold, with Pétion's imprint in all eighteen volumes, all dated 1845.
There is still a reasonable, albeit waning, case for buying either Paris edition if you can afford it and can find one, which you probably cannot. But be advised — despite the opinions of closed-minded booksellers, either deaf to marketplace reckoning or inflexibly tied to the past — that unless the buyer's obstacle is price or language, the growing plurality, in an easily seen trajectory that will soon be a majority, now want the real first edition of any book, by anybody, regardless of where it was published.

2. A set of the Journal des Débats serialization, 1844–1846, would be a killer, but no complete set has appeared for sale. Conversely, sets of the Le Siècle serialization are always available and frequently overpriced.

Price: $90,000.00