Ma Mère
l'Oye.

Reading the essential article by Nigel Simeone¹, which presents a chronological survey of the publication of Ravel's works through the Durand plate-number ledger, provides valuable information on the editorial history of the "Ballet in one act" Ma Mère l'Oye.

The orchestration of the pieces for piano four hands, undertaken in the autumn of 1911² following Jacques Rouché's commission for the Ballet, was only completed — according to the composer's own letters — a very few days before the first rehearsals and the premiere of the work, on 29 January 1912³: a prolific and dance-filled year for Ravel, with three Ballets added to his catalogue (alongside Adélaïde in April and Daphnis et Chloé in June).

While the Durand firm nevertheless found time to engrave the orchestral score of the five children's pieces for publication in January of the same year, and the orchestral parts followed in February, it is surprising to note that the additions to the piano cycle — the Prélude, the Danse du Rouet and the "raccords" between numbers, according to the terminology used by the author in the scenario written directly by Ravel⁴ and reworked belatedly, changes imposed by the contingencies of the brand-new production — were not published separately until August 1912. After the success of the first performances, Rouché re-programmed the Ballet in May 1912, then in March 1913, still at the Théâtre des Arts.

Gabriel Grovlez, the conductor who led the premiere with a miniature orchestra from the narrow pit of the Théâtre des Arts — at the head of an ensemble of 32 musicians⁵ (with at most six first violins, two double basses for the strings, winds, keyboards and harp, one timpanist and two percussionists) — formally complained to director Rouché about the "very special" state of the material⁶.

It is easy to imagine orchestral parts both engraved and handwritten — with numerous paste-ins, crossings-out, additions and unworkable page-turns — reordered to match the arrangement of the five-movement cycle that became for orchestras the "Suite" of Ma Mère l'Oye. N. Simeone's precise inventory raises an essential question: what of the publication of the complete orchestral score of the Ballet, and of its performance material?

While the piano-solo reduction made by Jacques Charlot — an exercise sanctioned by an over-busy Ravel, and which remains the single most faithful source for the original stage work — was published by Durand in July 1912 (the BnF holds the composer's precious copy strewn with a few manuscript annotations⁷), the English musicologist found no trace, in the Durand plate-number ledger, of any editorial work on the complete work carried out a posteriori, even several years later.

Yet during the composer's lifetime, monographic Durand catalogues of Ravel's works — for instance those published in 1914 or 1931 (the 1911 one already announces the forthcoming publication of the orchestration of the Cinq Pièces enfantines) — do list, under the heading "Ballet in one act", the title Ma Mère l'Oye with the mention "Material for hire", according to the same commercial scheme reserved, in those same publications, for Adélaïde or for the separate parts of Daphnis.

Consulting the libraries of the Parisian orchestral ensembles known to Ravel — Société des Concerts du Conservatoire (now the Orchestre de Paris), Lamoureux, Pasdeloup and Colonne orchestras, Opéra de Paris — confirms the dense editorial fog that surrounds this title.

Apart from the orchestral score that belonged to Pierre Boulez and which was certainly engraved in the 1970s⁸, no orchestral score of the "complete" Ballet can be located⁹ in the major research libraries. A void in the form of a total absence, which matches the catalogue of the score collections of the great Ravelian conductors such as C. Münch or A. Cluytens¹⁰ — who possessed, at best, only the first printings of the Cinq Pièces enfantines and the supplementary, separately-published volume of the Prélude and the Danse du Rouet.

To our knowledge, only the library of the Opéra de Paris¹¹ has piously preserved a "historical" complete performance material of the Ballet, bearing the date stamp of 26 December 1947 — a date that corresponds to the "Gala Ravel" given by André Cluytens at the Opéra-Comique on the tenth anniversary of the composer's death.

On these parts (the strings called for were 8-6-6-6-3), we clearly see that the matrix used was that of the "Suite" material combined with that of the Prélude and the Danse du Rouet, at the heart of a homemade compilation. It contains — by means of paste-ins, tape and other manual work by the librarian, added in handwritten ink — the missing notes of the transitions between the four interludes (Ravel's "raccords", labelled A, B, C and D by Durand), together with the cuts in Laideronnette and the reprise of the Pavane. Elements overlooked by many conductors, and by the most serious of them, which were nonetheless noted by musicologists such as Roger Nichols¹². A metrical fact confirmed by reading Jacques Charlot's reduction, and which resolves the impossible duplication of stage directions on identical bars as they have always been presented in the Durand edition.

While Arbie Orenstein, in 2012, was able to advance considerably our understanding of Ravel's original text (notably with the identification of historically missing bars in the first bassoon in the Pavane), the great lack of piano or orchestral manuscript sources for the Prélude, the Danse du Rouet and the "raccords" — even within the largest collections of Ravelian manuscripts — left questions open: the use of a drum of foreign timbre in the Danse du Rouet which does not appear in the instrumentarium published by Durand; the difficult reading of the fanfares for the two open or stopped horns — unless, following the example of the fanfares of Daphnis and the first Rosenthal and Cluytens recordings, these are spatialised; etc. We have attempted to resolve these questions in this new revised edition 2021.