W.S. Gilbert and Frederic Clay’s Ages Ago presents substantial editorial challenges for anyone seeking to put on a production, brought on by the fact that the extant sources of text often disagree with each other, but there is no one text which includes the complete dialogue and music, and so if you wish to produce the piece, you must necessarily make some decisions about which versions to combine, and how.

The text of our production of Ages Ago is derived from five sources:

• The Joseph Williams edition of the libretto, published in 1895 (JW)

• The Mallett edition of the libretto, undated, but published during the original run of the opera in 1869 or 1870 (M)

• The Boosey & Co. edition of the vocal score, published ca. 1870 (VS)

• The license copy of the libretto, held by the British Library (LC)

• An undated manuscript held in the Library of Congress, of uncertain provenance. (UM)

One other text source was consulted, but did not materially contribute to the text of this edition:

• An undated typescript held in the Houghton Library at Harvard (UT)

Additional historical documents, such as programs and newspaper reviews, were consulted in cases where they could help clarify what version of the text was actually performed in a given production.

JW is the only libretto containing dialogue published within Gilbert’s lifetime. Joseph Williams acquired the rights to publish editions of several of Gilbert’s early one-act works in the 1890s, and it does not appear that Gilbert substantially supervised the publication. The text is rife with errors. Nevertheless, for many years it was only source of dialogue for Ages Ago that was generally available to the public.

M was the first published text of Ages Ago. Although called a “libretto” here, it would be more specific to say that it is a book of the lyrics, containing no dialogue, but some descriptions of the action between the songs. It was available for audiences to purchase during the original run of Ages Ago, but it is not certain whether it was available on opening night itself.

VS was published around the time of the original run, although again the exact timing is uncertain. It is a vocal score, containing no dialogue. VS and M contain all the same songs, although they sometimes differ in the details of the lyrics.

LC is the license copy of the libretto in the British Library. In England at the time, plays had to be submitted to the Lord Chamberlain for approval in order to receive a license to be performed. Because the license copy had to be submitted far enough in advance of opening night in order for the license to be granted in time for opening night, the license copies of Gilbert’s plays typically represent a pre-production version of the text, not reflecting changes made during the rehearsal process. On the other hand, the date of the license being granted for Ages Ago is listed as November 23, 1869, one day after opening night, suggesting that the license copy may have been submitted very close to the wire.

UM is handwritten, not in Gilbert’s hand, and presents a state of the text somewhere in between LC and JW. Notably:

• LC and UM both contain lyrics for a trio for Dame Cherry, Lord Carnaby, and Brown, which is absent from M, JW, and VS.

• UM lacks all of the same songs that are missing from LC, but present in M.

• LC and UM both contain a passage of dialogue for Sir Cecil/Aubrey and Lady Maud about smoking tobacco which is absent from JW.

• UM, M, and VS all give the tenor the name of “Aubrey.” LC names him “Cecil Blount”; UM has “Cecil Blount” in the dramatis personae, but “Aubrey Blount” in the main text; and M has “Aubrey de Beaupré” throughout.

• LC, UM, and M all contain a third verse for Lady Maud’s ballad, which is absent from VS and JW.

Because of this, I believe UM to be a production period libretto representing a state of the text in between LC and M.

The dialogue of UT is largely consistent with JW, save the odd word or typo, but in the songs it more closely resembles UM, omitting several of the same songs, but, like UM and LC, including the third verse of Lady Maud’s ballad, and the trio for Brown, Lord Carnaby, and Dame Cherry. The presence of this cut trio suggests that this is an early version of the text, dating from the original production period.

UM and UT both bear resemblance to manuscript and typescript copies of Gilbert’s Our Island Home – the immediate successor to Ages Ago – which are held in a collection in the Morgan Library, packaged in a box with rehearsal scores. Our Island Home was never published or revived, and so the logical inference is that all of these scripts date to the original production period. The similarity of the Our Island Home typescripts and manuscripts to UM and UT could help corroborate the hypothesis that these are rehearsal period materials from 1869.

Most productions of Ages Ago derive their text from JW and VS. JW was the only source of dialogue, and VS the only source of music, published within the authors’ lifetimes. The challenge for any production lies in deriving a performable text from these two sources which do not agree with each other, yet neither of which is performable without the other.

The most obvious difficulty is that VS contains three songs which do not appear in JW. In fact, the first three songs of the show: No. 2, 3 and 4[1], which are, respectively, the opening trio, the duet for Rosa and MacMotherly, and MacMotherly’s solo.  JW begins with Rosa’s line “Mrs. MacMotherly, I’m a dreadful little goose,” and no lyrics appear until Hebblethwaite’s entrance (No. 5). If you want to include all of the music in VS, then you have to figure out how to insert them into that first stretch of dialogue.

No. 2 is simple enough. An opening number, clearly musically intended to directly follow the prelude. No revisions to the dialogue are necessary, although there is then awkwardness in Rosa’s first spoken line, which does not seem to follow naturally from the trio that precedes it.

No. 3, the duet for Rosa and MacMotherly, is the most challenging song to place, as there is no clear indication as to where in the dialogue it is meant to be inserted. For this production, we have opted to include it directly after No. 2 with no intervening dialogue. We have devised a staging which we believe works with having the two songs back to back.

No. 4, Mrs. MacMotherly’s solo, is only mildly challenging. Its placement in the text can be roughly inferred; it seems clear that it belongs just before MacMotherly’s exit from the first scene. But the dialogue in JW provides no clear cue line for it. If it is simply inserted after MacMotherly’s final line in the scene, the transition is awkward, and the lyric has MacMotherly responding to things that Rosa never said in the dialogue.

UM, helpfully, contains one line that seems unique to that version of the text: An explicit cue line for No. 4. The dialogue leading into the song is slightly different from any other version, and while the lyrics of the song are not written out in UM, it explicitly indicates where the song is meant to go. We have therefore opted to use the text from UM for this scene.

With the first three songs positioned, the remainder of JW and VS can largely be woven together with no glaring conflicts. However, we have opted to include some additional text from LC where we felt it might contribute something.

A minor editorial decision had to be made regarding the ensemble repeats of the text in No. 6, “It does perplex, annoy and vex.” The song is in two verses, each led by a soloist, and then repeated by the quintet. For the quintet, VS only prints out the lyrics for the first verse. Some productions have taken this at face value, having the ensemble repeat the exact same text both times, even when the soloist’s text changes. Others have decided to have the ensemble repeat some or all of the soloist’s text in the second verse, as is typical for songs of this structure. We have opted for the latter approach.

There were two other significant editorial decisions which had to be made, both pertaining to the character of Sir Cecil/Aubrey.

The first is the name. “Sir Cecil” is what is most commonly seen in modern performance, because that is what is present in the dialogue of JW. LC also names him Cecil. However, UM, M, and VS, all have the name “Aubrey.” All three of those sources I believe to date to around the original production period, and later than LC. UT has Cecil. There is not a clear directionality of differences between UM and UT, and I believe they may have been prepared around the same time, and not necessarily in direct sequence, allowing for mistakes and inconsistent communication about changes. M and VS, being printed and published for general purchase, most likely represent a later state of the text than UM or UT. JW, though published later, seems to be based primarily on UT.

A review published in The Era published a couple of weeks into the show’s original run in 1869 names the character as Sir Aubrey de Beaupré. Reviews of revivals in 1874 and 1881, as well as of the opera’s 1880 New York premiere also give the name Aubrey. [2, 3]

The other major editorial decision regarding this character is the identity of the painter who painted him. Nearly every libretto source has “Michael Angelo”[4] except for M, which has “Titian.” At least two reviews of the original production in 1869 mention Titian, although it is possible that “Michael Angelo” may be what was spoken, but that the reviewers in question copied the names of the painters from M. I have found no reviews from 1869 mentioning Michael Angelo. An 1880 review of the New York premiere mentions Michael Angelo[5], as does a review of an 1881 London revival,  suggesting that the change had been reverted, possibly because Michelangelo was, and remains, the more recognizable name to most audiences. For our production, we have opted to go with Titian for the following reasons:

1. I believe it to be what was performed in the original production based on M and the reviews.

2. The dates of Titian’s career and death line up slightly better with the years given in the libretto, though still not perfectly.

3. For our staging, we opted to use poster prints of paintings by the actual artists mentioned, and this was more viable with Titian than with Michelangelo, who did not paint very many portraits.

Another circumstance in which we have taken some editorial liberties with this text pertains to the years and ages of the various portraits. Gilbert gives specific years for when each person represented was born, painted, came into possession of the castle, and died. Between the frontmatter, the dialogue, and the lyrics, Gilbert often gives several conflicting numbers, which also change from revision to revision. In addition, the portraits are all stated to have been painted by real painters, and the years given don’t always line up with the careers of the painters as we know them historically. While we have not rectified every inconsistency – there remain some glaring contradictions which I don’t know what to do with – in cases where small inconsistencies can be smoothed over across the various versions of the text, we have done so.

There is one final liberty we have taken with the text for our production. As we are using for our design copies of real portraits by the real painters mentioned in the text, we have necessarily altered two words relating to a visual description of one of the paintings in order to better match the portrait being used. We hope that this will not offend.