The Thrilling Adventures of Lovelace and Babbage

Sydney Padua, The Thrilling Adventures of Lovelace and Babbage (Pantheon, 2015). Image source: The Beat, the blog of comics culture

It's an uncanny feeling to open a book and have the sense that it was written specifically for you. I had that feeling almost continuously while reading Sydney Padua's The Thrilling Adventures of Lovelace and Babbage (Pantheon, 2015).

Augusta Ada Byron King, Lady Lovelace, was the daughter of Anne Isabella (Annabella) Milbanke and the poet Lord Byron. Ada's mother separated from Byron after a year of marriage and raised her daughter to have an anti-Byronic cast of mind: rational, scientific, mathematical. Before their marriage Byron called Annabella "my Princess of Parallelograms"; afterwards he called her the "Mathematical Medea."

Annabella Milbanke by George Hayter, 1812, three years before her marriage to Byron. Image source: Wikimedia Commons

Lady Byron's plan to suppress her daughter's imagination through rationality, however, was destined to be foiled by how truly strange math is. One of Ada's tutors, William Frend (who had also taught Annabella as a young woman), refused to believe in the existence of negative numbers, because how could anything be less than nothing? Numbers can also be irrational (if they can't be expressed as a ratio of two integers, such as π) and even imaginary (if they involve the square root of a negative number, such as i, where i² = –1). Ada outpaced most of her teachers.

Ada Byron at 16, 1832, artist unknown. Image source: Wikimedia Commons

In June 1833, at age 17, Ada met the mathematician and inventor Charles Babbage, then the Lucasian Professor of Mathematics at Trinity College, Cambridge. [1]

Engraving of Charles Babbage, 1832 (age 40) by John Linnell, published by Paul and Dominic Colnaghi & Co., 1 January 1833. Image source: National Portrait Gallery NPG D16124

In a letter dated Friday 7 June 1833 to Ada's tutor Dr. William King, Lady Byron wrote:

Ada was more pleased with a party she was at on Wednesday [5 June], than with any of the assemblages in the grand monde [Ada had been presented at court on 10 May and had since attended several court balls]—she met there a few scientific people—amongst them Babbage, with whom she was delighted—. . .Babbage was full of animation—and talked of his wonderful machine (which he is to shew us) as a child does of its plaything. [2]

Babbage was working on a mechanical calculator he called the Difference Engine, and had produced a small working prototype which he demonstrated for Ada and her mother. In a letter to King on Friday 21 June, Lady Byron wrote,

We both went to see the thinking machine (for such it seems) last Monday [17 June]. It raised several Nos. to the 2nd & 3rd powers, and extracted the root of a Quadratic Equation.—I had but faint glimpses of the principles by which it worked.

Sophia Frend, William's daughter, also attended the demonstration and later wrote that in contrast to Lady Byron, "Miss Byron, young as she was, understood its working, and saw the great beauty of the invention." [3]

"Portion of Babbage's Difference Engine," Harper's New Monthly Magazine, Vol. XXX, No. CLXXV, December 1864, p. 34, digitized from the collections of the University of California Berkeley. Image source: HathiTrust.org

In 1834, with the Difference Engine still far from completion, Babbage began working on an even more sophisticated device he called the Analytical Engine. He envisioned a punchcard control system that would specify the mathematical operations to be performed. The principle was similar to the way punchcards controlled the mechanical Jacquard looms used to weave complex patterned cloth (and which had been the targets of the Luddite weavers who had been thrown out of work).

Portion of the mill of the Analytical Engine with printing mechanism, designed by Charles Babbage and under construction at the time of his death, London, 1834-1871. (Trial model). Image source: Science Museum Group Collection Online

In 1835 Ada Byron married William King (a different William King than her tutor). Thanks to Ada's family connections—her mother was the cousin of the prime minister, Lord Melbourne—William was created the first Lord Lovelace in 1838, and so Ada became Lady Lovelace. [4]

Ada King by Margaret Carpenter, 1836. Image source: Wikimedia Commons

Motherhood, family concerns and social demands took Ada away from her mathematical studies, but after the birth of her third child in four years of marriage she was eager to resume them. She began a correspondence with the mathematician Augustus De Morgan, husband of Sophie Frend and friend of Charles Babbage. De Morgan later wrote to Lady Byron:

But I feel bound to tell you that the power of thinking on these matters which Lady L. has always shewn from the beginning of my correspondence with her, has been something so utterly out of the common way for any beginner, man or woman. . .Had any young beginner, about to go to Cambridge, shewn the same power, I should have prophesied first that his aptitude at grasping the strong points and the real difficulties of first principles. . .would have certainly made him an original mathematical investigator, perhaps of first-rate eminence. [5]

In 1840 Babbage gave a lecture on the Analytical Engine in Turin; mathematician (and future prime minister of Italy) Luigi Menabrea attended, and two years later published an article describing the machine in the journal Bibliothèque universelle de Genève—the first publication about the Analytical Engine.

L.-F. Menabrea, "Notions sur la machine analytique de M. Charles Babbage." Bibliothèque universelle de Genève, Nouvelle Série, tome XLI (Sept.-Oct. 1842), pp. 352-376. Image source: HathiTrust.org

Shortly after it appeared Ada began working on a translation of Menabrea's article; her translation was published the following year in Richard Taylor's Scientific Memoirs, Vol. III, Part XII. The 25-page-long "Sketch of the Analytical Engine invented by Charles Babbage, Esq." by Menabrea was accompanied by 41 pages of "Notes by the Translator," signed A.A.L. (for Augusta Ada Lovelace). [6]

Ada shared the notes with Babbage as she was writing them, but was understandably exasperated if he made revisions without her agreement. As her drafts were traveling back and forth between their houses she wrote him, "I am much annoyed at your having altered my Note. You know I am always willing to make any alterations myself, but that I cannot endure another person to meddle with my sentences." [7]

A.A.L., "Notes by the Translator" [to accompany L. F. Menabrea, "Sketch of the Analytical Engine invented by Charles Babbage, Esq." (pp. 666-690)] in Richard Taylor, ed. Scientific Memoirs, Selected from the Transactions of Foreign Academies of Science and Learned Societies, and Foreign Journals, Vol. III, Part XII (Aug. 1843), pp. 691-731. Image source: HathiTrust.org

Ada's notes to Menabrea's article include instructions for the calculation of Bernoulli numbers by the Analytical Engine, an algorithm that has been called the first computer program.

Sydney Padua, The Thrilling Adventures of Lovelace and Babbage (Pantheon, 2015), p. 25 (detail). Image source: Goodreads

Ada's copious Notes, which often themselves have footnotes, were clearly an inspiration for the format of Thrilling Adventures. Marketed as a graphic novel, it's a hybrid mix of graphics and text: virtually every page features extensive footnotes, and, in a droll touch, the footnotes have even more extensive endnotes (which, like Ada's "Notes by the Translator," also have footnotes). The fascinating dual subjects of the book deserve that extensive scholarly apparatus, even if its purpose is partially parodic. As someone who writes blog posts whose footnotes are sometimes longer than the post itself, I felt an immediate sympathy. [8]

Sydney Padua, The Thrilling Adventures of Lovelace and Babbage (Pantheon, 2015), p. 22; right-click to enlarge. Image source: Science Museum Group Journal

In her Notes Ada had made a huge leap; she'd recognized that the Analytical Engine could perform operations on any data to which an algorithm could be applied:

The operating mechanism. . .might act upon other things besides number, were objects found whose mutual fundamental relations could be expressed by those of the abstract science of operations, and which should be also susceptible of adaptations to the action of the operating notation and mechanism of the engine. [9]

With this insight Ada anticipated the development of the modern computer a century later.

Ada Lovelace, daguerreotype by Antonie Claudel, ca. 1843. Image source: Wikimedia Commons

Babbage teased Ada (who was slight of build) by calling her a fairy; in a letter to him as she was working on the Notes she embraced that supernatural identity:

That brain of mine is something more than merely mortal; as time will show; (if only my breathing & some other etceteras do not make too rapid a progress towards instead of from mortality). Before ten years are over, the Devil’s in it if I haven’t sucked out some of the life-blood from the mysteries of this universe, in a way that no purely mortal lips or brains could do. [10]

It was not to be. Instead of plumbing the mysteries of the universe, Ada used her mathematical talents to develop a betting system for horse racing. Initially both her husband and Babbage joined her in placing bets using the system. However, when it became clear that the system didn't work, the men stopped using it. Ada continued, disastrously, and fell deeply in debt; she secretly pawned her jewels, twice, and had to ask her mother to redeem them.

As the reference to her breathing in the letter above suggests, Ada's health was also not robust. Victorian doctors prescribed opium in various forms (pills or dissolved in alcohol as laudanum) for a host of maladies, including to ease the breathing difficulties whose actual cause was likely the thick pall of London coal and wood smoke. The inhibition-lowering properties of opium may have played a role in her reckless gambling, although high-stakes betting seems to have been endemic among the British upper classes.

Sydney Padua, The Thrilling Adventures of Lovelace and Babbage (Pantheon, 2015), p. 28 (detail). Image source: Pinterest

Ada was also frequently in pain, and in mid-1852 she was diagnosed with terminal cancer. Perhaps as a memento, her husband commissioned a final portrait of her by the artist Henry Wyndham Phillips; despite her condition, she sat for the painting throughout August. On 13 August Lord Lovelace recorded in his diary that "the suffering was so great that she could scarce avoid crying out"; although she was in agonizing pain, "she sat at the piano some little time so that the artist could portray her hands." [11]

Daguerreotype by an unknown photographer of a portrait of Ada Lovelace by Henry Wyndham Phillips, 1852 (detail). Image source: Bodleian Library

Babbage was forbidden by Lady Byron from visiting Ada after mid-August. After months of suffering Ada died on Saturday 27 November 1852, two weeks before her 37th birthday.

In the decades following the publication of Ada's article, Babbage made few advances toward building the Analytical Engine. Progress remained stalled pretty much as described by an anonymous article entitled "Addition to the memoir of M. Menabrea on the analytical engine" published in The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science in September 1843 (immediately after the publication of Ada's translation and Notes in Scientific Memoirs). The article, almost certainly written by Babbage, states:

The present state of the Analytical Engine is as follows:—
All the great principles on which the discovery rests have been explained, and drawings of mechanical structures have been made, by which each may be carried into operation. . .
Mechanical Notations have been made both of the actions of detached parts and of the general action of the whole, which cover about four or five hundred large folio sheets of paper.
The original rough sketches are contained in about five volumes.
There are upwards of one hundred large drawings. [12]

The plans for the Analytical Engine were clearly of daunting complexity, and it's no wonder that they were not realized. Babbage was a spiky personality and an eccentric polymath, and spent a good deal of time and energy not only on other projects, but pursuing cantankerous campaigns against street musicians, hoop-rolling, and other public nuisances. At the time of his death in 1871 a partial model of the Analytical Engine was still incomplete (see above).

In 2009 over a beer at a pub animator Sydney Padua was commissioned by Suw Charman-Anderson, who was planning the first annual Ada Lovelace Day, to produce a short web comic on Ada's life. In the preface to Thrilling Adventures Padua writes, "Lovelace died young. Babbage died a miserable old man. There never was a gigantic steam-powered computer. This seemed an awfully grim ending for my little comic. And so I threw in a couple of drawings at the end, imagining for them another, better, more thrilling comic-book universe to live on in" (p. 7).

Sydney Padua, The Thrilling Adventures of Lovelace and Babbage (Pantheon, 2015), p. 32. Image source: The Beat, the blog of comics culture

Those "couple of drawings" grew into the 300-page Thrilling Adventures. Each adventure takes place in a steampunk alternate universe in which a full-scale Analytical Engine has been built. Over the course of Thrilling Adventures Lovelace and Babbage encounter the poet Samuel Taylor Coleridge, Queen Victoria, the novelist George Eliot, the logician George Boole, and the world of the Alice stories written by Lewis Carroll. [13]

Padua's Thrilling Adventures of Lovelace and Babbage is an exhilaratingly witty and imaginative journey through computer history and the Victorian scientific, political, and literary worlds. It is also an illuminating portrayal of Ada Lovelace, whose pioneering work, while still underappreciated, has begun to receive the attention it deserves. The annual Ada Lovelace Day, on the second Tuesday in October, is now a global grassroots celebration of the achievements of women in science, technology, engineering and mathematics; in 2023 the 15th annual Ada Lovelace Day will take place on 10 October. [14]

And in the alternate universe of Thrilling Adventures, there is a happy ending for Lovelace and Babbage, as the Analytical Engine is completed and they can spend many contented hours in conversation and calculation:

Sydney Padua, The Thrilling Adventures of Lovelace and Babbage (Pantheon, 2015), pp. 286-287. Image source: Mark Riedl, Medium.com

---

1. Lucasian Professors have included Isaac Newton (1669–1702) and Stephen Hawking (1979–2009). ^ Return
2. Quoted on p. 299-300 in Velma R. Huskey and Harry D. Huskey, "Lady Lovelace and Charles Babbage," Annals of the History of Computing, Vol. 2, No. 4, pp. 299-329, Oct.-Dec. 1980, doi: https://doi.org/10.1109/MAHC.1980.10042 (subscription or institutional access required). ^ Return
3. Quoted in Huskey, p. 300. ^ Return
4. William Lamb, Lord Melbourne, was the widower of Caroline Lamb, who had been the lover of Ada's father, Lord Byron, before his marriage to Annabella. For details, see An Elegant Madness: High Society in Regency England. ^ Return
5. Quoted in Huskey, p. 326. ^ Return
6. Notes A - F are signed A.A.L.; Note G is signed A.L.L., likely a typographical error for A.A.L. The suggestion for signing the notes with her initials came from Ada's husband. ^ Return
7. Quoted in Huskey, p. 316. ^ Return
8. In Thrilling Adventures Sydney Padua offers the following footnote: "Lovelace added seven footnotes to her translation of Menabrea's Sketch of the Analytical Engine; they are a little over two and a half times longer than the original paper. . .Together they take up 65 pages of the September 1843 edition of Taylor's Scientific Memoirs, a journal dedicated to publishing English translations of work from continental Europe" (p. 25).
   
   This isn't quite right. Lovelace's "Notes by the Translator," lettered A. to G., are endnotes, not footnotes (Ada also added footnotes on 10 of the original article's 25 pages); the notes are two-thirds again as long as the original article (41 pages of notes to 25 pages of article), not over two and a half times longer; together with the original paper they take up 66, not 65, pages of Vol. III of Taylor's Scientific Memoirs, each volume of which was published in parts, not "editions"; and finally, Part XII of Scientific Memoirs, according to the Proceedings of the American Philosophical Society, Vol. 4, No. 28 (Jun. - Dec., 1843), p. 19, and Ada's correspondence as quoted in Huskey, was published in August, not September, 1843.
   
   There are some other minor errors as well. In the first footnote on p. 22 (see above) Padua writes "Babbage was forty-two and Lovelace eighteen when they met." Babbage was born on 26 December 1791, and Lovelace on 10 December 1815; in June 1833 his 42nd and her 18th birthdays would still have been six months away. ^ Return
9.  A.L.L., "Notes by the Translator," p. 694. ^ Return
10. Quoted in Huskey, p. 315. ^ Return
11. Ursula Martin, "Only known photographs of Ada Lovelace in Bodleian Display," Ada Lovelace: Celebrating 200 years of a computer visionary, https://blogs.bodleian.ox.ac.uk/adalovelace/2015/10/14/only-known-photographs-of-ada-lovelace-in-bodleian-display/ ^ Return
12. [Charles Babbage,] "Addition to the memoir of M. Menabrea on the analytical engine. Scientific memoirs, vol. III. Part XII. p. 666," The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science, Vol. XXIII, No. CLI, September 1843, pp. 235-239. https://babel.hathitrust.org/cgi/pt?id=wu.89048308431&view=1up&seq=249 ^ Return
13. Thrilling Adventures is also a corrective to the sexualized and misogynistic treatment of Ada in the 1990 William Gibson and Bruce Sterling novel The Difference Engine, in which a steam-powered computer revolution based on Babbage's Engines takes place in Britain in the mid-19th century. In that novel "persistent sexual slurs" are used to describe Ada; "the insistence on misogynistic imagery. . .perpetuates rather than interrogates nineteenth-century stereotypes" (Jay Clayton, Charles Dickens in Cyberspace: The Afterlife of the Nineteenth Century in Postmodern Culture, Oxford University Press, 2003, p. 112).  ^ Return
14. For information about Ada Lovelace Days, see the website Finding Ada. ^ Return

Dido & Aeneas: A brief survey

La Morte di Didone (The Death of Dido) by Guercino, 1631. Image source: Classical Inquiries

Henry Purcell's Dido and Aeneas (1688?), the greatest opera in English, was written for parents' night at a girls' school. Possibly as a result, the original performances—there is evidence suggesting that there were at least two, one in 1688 or earlier and another in 1689—were only about 75 minutes long, including a prologue and the recital of a poem after the end of the opera (for details, please see The Mysteries of Dido and Aeneas). The music to the prologue has been lost, and we no longer recite poems after the curtain has come down; the surviving music can be performed in about 50 minutes.

The opera's tunefulness and brevity (it can fit on a single LP, and easily on a CD) has made it an irresistible recording project for several decades, and the role of Queen Dido of Carthage has been sung by a long list of renowned singers, including:

• Kirsten Flagstad (conducted by Geraint Jones, 1951),
• Teresa Berganza (Pierre Dervaux, 1960),
• Janet Baker (Anthony Lewis, 1962),
• Victoria de los Angeles (John Barbirolli, 1965),
• Tatiana Troyanos (Charles Mackerras, 1967, and Raymond Leppard, 1977),
• Emma Kirkby (Andrew Parrott, 1981),
• Lynne Dawson (René Jacobs, 1998),
• Susan Graham (Emmanuelle Haïm, 2003),
• Julianne Baird (Valentin Radu, 2007),
• Sarah Connolly (Steven Devine & Elizabeth Kenny, 2008).

On MusicWeb International you can read Ralph Moore's survey of 30 Dido recordings (his tastes aren't quite mine); the Dido discography on Wikipedia lists nearly 50 versions.

I'm going to do nothing so ambitious as Moore or Wikipedia. Instead, I'm going to offer notes on six recordings involving period-instrument ensembles and their HIP conductors (that's "historically-informed performance," although not everything HIP conductors do is historically justified, as we will see).

My mini-survey was inspired by my previous post on Venetia Murray's An Elegant Madness: High Society in Regency England. That post included a caricature by James Gillray in which the Prince of Wales is cast as Aeneas and his mistress Mrs. Fitzherbert as Dido:

"DIDO FORSAKEN. Sic transit gloria Regina [Thus passes the glory of the Queen]." George, Prince of Wales: "I never saw her in my Life." Steersman Charles Fox: "No, never in all his Life, Damme." Frederick North: "No, never." Edmund Burke: "Never." Caricature by James Gillray, published by Samuel William Fores, 21 May 1787. Image source: National Portrait Gallery NPG D12994

(For a fuller explanation of the historical circumstances of the caricature, please see An Elegant Madness; for a discussion of its imagery, please see that post's footnote #4.)

The story being referenced in the caricature is from Virgil's Aeneid. The Trojan hero Aeneas, escaping from the destruction of Troy with a few ships and a few dozen men, finds shelter at Carthage, ruled by the widowed Queen Dido. At a court banquet Aeneas tells the harrowing story of the fall of Troy and his escape from the burning city. Dido, to her dismay, finds herself falling in love with the hero. Aeneas' fate, though, is to sail to Italy and found Rome, and when the gods send Mercury to remind him, he abandons Dido. She kills herself with a sword Aeneas gave her as a gift; as his ships sail away, Aeneas sees the smoke from Dido's funeral pyre rising behind them.

Nahum Tate's libretto for Purcell's opera alters Virgil's epic by removing the gods as agents of Dido's destruction and substituting a sorceress whose hate for Dido inspires her to send a spirit ("in the form of Mercury") to command Aeneas to leave Carthage. Dido does not commit suicide but dies of heartbreak. Schoolgirls could hardly depict self-murder, although the implication that Dido and Aeneas consummate their love at some point between the Palace and the post-coital Grove scenes (Aeneas sings "one night enjoy'd/the next forsook") is certainly eyebrow-raising—perhaps it was acceptable for unmarried young women to enact Dido's story as a cautionary tale. For more on the opera, please see Opera Guide 1: Dido and Aeneas.

In Gillray's caricature, the dialogue of the men in the boat and its repetition of "never" is likely a reference to the First Sailor's song from Dido and Aeneas:

Come away, fellow sailors, your anchors be weighing,
Time and tide will admit no delaying.
Take a boozy short leave of your nymphs on the shore,
And silence their mourning with vows of returning,
But never intending to visit them more,
No never intending to visit them more,
No never, no never intending to visit them more.

Discovering the caricature and its apparent quote from Dido and Aeneas sent me to listen to opera once again, and in choosing the link to the First Sailor's song I began to make comparisons among several performances. I limited myself to period-instrument versions of the opera, and ultimately chose six to survey, given in the table below:

Role	Arkiv Produktion, 1988	L'Oiseau-Lyre, 1992	Harmonia Mundi, 1993	Erato, 1994	Chandos, 1995	Rhombus Media, 1995
Dido	Anne Sofie von Otter	Catherine Bott	Lorraine Hunt	Véronique Gens	Maria Ewing	Jennifer Lane
Aeneas	Stephen Varcoe	John Mark Ainsley	Michael Dean	Nathan Berg	Karl Daymond	Russell Braun
Sorceress	Nigel Rogers	David Thomas	Ellen Rabiner	Claire Brua	Sally Burgess	Jennifer Lane
Belinda	Lynne Dawson	Emma Kirkby	Lisa Saffer	Sophie Marin-Degor	Rebecca Evans	Ann Monoyios
Second Woman	Sarah Leonard	Julianne Baird	Donna Deam	Sophie Daneman	Patricia Rozario	Shari Saunders
Witches	Elizabeth Priday Carol Hall	Elizabeth Priday Sara Stowe	Christine Brandes Ruth Rainero	Sophie Daneman Gaëlle Mechaly	Mary Plazas Pamela Helen Stephen	Shari Saunders Meredith Hall
Spirit	Kym Amps	Michael Chance	Christine Brandes	Jean-Paul Fouchécourt	James Bowman	Meredith Hall
First Sailor	Nigel Rogers	Daniel Lochmann	Paul Elliott	Jean-Paul Fouchécourt	Jamie McDougall	Benjamin Butterfield
Conductor	Trevor Pinnock	Christopher Hogwood	Nicholas McGegan	William Christie	Richard Hickox	Jeanne Lamon
Orchestra	The English Concert	The Academy of Ancient Music	Philharmonia Baroque Orchestra	Les Arts Florissants	Collegium Musicum 90	Tafelmusik Orchestra
Chorus	Choir of The English Concert	Chorus of The Academy of Ancient Music	Choir of Clare College, Cambridge	Soloists, plus Steve Dugardin (alto) and Jonathan Arnold (bass)	Collegium Musicum 90 Choir	Tafelmusik Chamber Choir

This is not an exhaustive list of period-instrument versions by any means. But just as I prefer the sound of historical instruments in this music, I favor a particular voice type in the role of Dido, which eliminated several possibilities.

Dido is a sorrowful figure from the first moments of the opera: the words of her first aria, addressed to her sister Belinda, relate her torment and grief at the love for Aeneas that as queen she must conceal. A young woman might be joyful at newfound love; Dido, older and wiser, understands that falling in love could be disastrous (as, indeed, it turns out to be).  Singers performing this role must combine both depth of emotion and nobility of expression. The role also spans a wide range, and requires a strong lower register.

So I prefer a mezzo-soprano Dido; high sopranos Emma Kirby for Andrew Parrott, Lynne Dawson for René Jacobs, and Julianne Baird for Valentin Radu have the wrong sound, to my ears, and so those recordings were not included in the survey. (I made an exception for Véronique Gens, one of my favorite singers, with William Christie.) I also prefer a baritone Aeneas and a mezzo-soprano or alto Sorceress, although having another voice type in those roles did not automatically eliminate a recording. (After all, we have evidence that at the original performances at Josias Priest's school Aeneas was sung by a young woman.)

Finally, the surviving music is scored for strings and continuo. Adding winds (as do both Jacobs and Emmanuelle Haïm) has no specific justification. Nor does repeating the first section of the overture (as do both Jacobs and Haïm), which has no repeat sign in the early scores. I far prefer versions that rely on the existing musical evidence over interventionist approaches that attempt to "improve" Purcell.

Taking my selections in chronological order:

The English Concert & Choir conducted by Trevor Pinnock. Image source: Presto Music

For Trevor Pinnock's version on Archiv, Anne Sofie von Otter has the requisite regal gravity. But at the words "Remember me" in her final aria von Otter sings "Remombah mah!," calling attention not to the despair of the character, but to the difficulty of singing the highest notes of the role. In his survey linked above, Ralph Moore says that "Von Otter’s final scene is up with the best." This pronunciation flaw is so obtrusive I can't agree; your judgment may differ.

Stephen Varcoe in the bass-baritone role of Aeneas sounds quite similar to tenor Nigel Rogers' Sorceress. While it could be a dramatic choice to depict both Aeneas and the Sorceress as linked agents of Dido's destruction, this feels more like Pinnock was simply following the tradition established in the 18th and 19th centuries for a male Sorceress. As for Rogers, according to Ellen Harris' Henry Purcell's Dido and Aeneas, 2nd edition (Oxford, 2018), the earliest surviving manuscript of the score, from 1700, sets the Sorceress's part in the mezzo-soprano range; in later 18th-century sources the part is written for bass. I prefer a woman in this role; in any case, here I agree with Moore about Rogers that "there is little of menace or the macabre in his threats—he just sounds like a bloke singing the Sorceress' words."

Pinnock's tempos are stately, and as a result this version feels sluggish and lacks dramatic drive at key points. (Characteristically, Pinnock's overture takes 2:20, versus 2:04 for Hogwood, 1:54 for Christie and Hickox, and 1:51 for McGegan.) Taken together these issues are sufficiently bothersome for me to recommend against this version. However, Matthew Boyden in The Rough Guide to Opera, 3rd edition (2002), selects it as one of his top choices.

Chorus and Orchestra of The Academy of Ancient Music conducted by Christopher Hogwood. Image source: Presto Music

Speaking of blokes singing the Sorceress' words, bass David Thomas on Christopher Hogwood's L'Oiseau-Lyre version is even worse than Rogers. Singing in an exaggerated, affected way, Thomas sounds like a pantomime dame. Here is the Sorceress's first appearance, "Wayward sisters":

https://www.youtube.com/watch?v=m81q1Lp9Xog

The witches have also been directed by Hogwood, unsuccessfully, to camp it up (Stowe in particular just sounds excessively nasal). Hogwood makes another odd choice: the First Sailor, Daniel Lochmann, is a boy treble, making incongruous his admonition to his fellow sailors to "take a boozy short leave of your nymphs on the shore." If David Thomas evokes a panto dame, the First Sailor sounds like he's wandered in from a production of Oliver. John Mark Ainsley is a tenor Aeneas, and although he has a good voice, I prefer a bass-baritone in the role.

Finally, Catherine Bott's mezzo-soprano is not particularly rich or resonant, particularly in the role's lower range, and her manner is strangely detached: when she sings "I am press'd with torment," she hardly sounds it, nor does her final scene sound like a woman facing the end of her hopes, and her life. (Bott does sing a beautifully floated and correctly pronounced "Remember me!"). With these substantial problems, I can't recommend this version either, although it is given an "Outstanding" rating in the Gramophone Classical Good CD Guide 2004.

The Choir of Clare College, Cambridge and the Philharmonia Baroque Orchestra conducted by Nicholas McGegan. Image source: Presto Music

I confess that I can't be fully objective about Nicholas McGegan's version on Harmonia Mundi: McGegan and his Philharmonia Baroque Orchestra were in the pit for our first performances of the opera by the Mark Morris Dance Group, our Baroque opera conversion experience, and his was the first recording of this work we heard.

McGegan does make two missteps, in my view, which I'll address straight away. The first is casting contralto Ellen Rabiner as the Sorceress. Her pronounced vibrato and Wagnerian vocal style set her apart from the rest of the cast. Rabiner is a significant improvement over either Thomas or Rogers, and casting a low female voice in the role of the cave-dwelling Sorceress is a choice that could have worked extremely well. However, I regret that McGegan didn't opt for Jennifer Lane (who sings the dual roles of Dido and the Sorceress for Jeanne Lamon on the DVD of the Mark Morris production) or Judith Malafronte (who sang Dido and the Sorceress for McGegan when we saw the Mark Morris production live).

McGegan's second misstep is to immediately follow the final scene of the opera with Purcell's music for The Gordian Knot Unty'd. This is clearly done to fill out the CD, which would otherwise be only 50 minutes long. The contrast, though, between the moving final chords of Dido's death scene and the dance-filled incidental music to The Gordian Knot Unty'd is jarring. Fortunately this is a fault easily remedied by CD programming, playlist editing, or just hitting the stop button. Dido does not need any fillers.

The performances of the witches and sailors (soloists and chorus) are quite characterful. The witches have a kind of wild, sinister hilarity, and Paul Elliott's First Sailor is sung as an authentic-sounding Jack Tar. This theatrical approach may bother some listeners, but I find that (unlike Hogwood's) McGegan's direction doesn't go over the top, but rather enhances the flavor of the proceedings—this is, after all, a stage work.

But the glory of this set is Lorraine Hunt's Dido, a role perfect for her plangent mezzo-soprano. Hunt is regal and conveys deep feeling; a better embodiment of the tragic queen is impossible to imagine. Helped by McGegan's urgent pacing, Hunt also summons fierce anger in Dido's final confrontation with Aeneas, when she rejects his last-minute change of heart and promise to stay. And her sorrowful closing lament simply stops time. We are fortunate indeed to have this record of Hunt's performance.

Dido's first aria, expressing her foreboding about her developing feelings for Aeneas, "Ah! Belinda, I am press'd with torment":

https://www.youtube.com/watch?v=Tz9aXLHIjzA

Hunt's performance makes this version my first choice. Hunt would go on to powerfully portray other tragic women, including Medea (Charpentier's Médée, recorded with William Christie in 1994), Irene (Handel's Theodora, recorded with William Christie in 1996), Phédre (Rameau's Hippolyte et Aricie, recorded with William Christie in 1997), Empress Octavia (Ottavia in Monteverdi's L'Incoronazione di Poppea, at SF Opera in 1998), and Berlioz's Dido (Didon in Les Troyens, at the Metropolitan Opera in 2003). Her death from breast cancer at age 52 in 2006 was a profound loss.

Les Arts Florissants conducted by William Christie. Image source: Presto Music

Given the French influences on Stuart society in the 1680s—the mother of James II, Henrietta Maria, was French, and as a young man James served in the army of Louis XIV—it is certainly defensible to perform Dido and Aeneas in French style. And William Christie's version on Erato has many virtues. As with McGegan's recording, Christie's benefits from extensive experience in staged opera, and offers dynamism and drama (at 45 minutes it's the shortest version in this survey). It also has the suitably heroic Aeneas of Nathan Berg, and the excellent Sophie Daneman as Second Woman and First Witch. The Second Woman's recounting of the tale of Diana and Actaeon, "Oft she visits this lone mountain":

https://www.youtube.com/watch?v=B4yP26skIZg

However, there are also some noteworthy issues. Most of the cast (the British Daneman and the Canadian Berg aside) sing their English lines with light but detectable French accents. (Of course, French, Italian, and German audiences must frequently listen to non-native opera singers outright massacre their languages, so being bothered by this seems unfair; still.) Claire Brua's Sorceress "is a splendidly insinuating conception, with a slithering melodic style," according to the Gramophone guide; you may feel, as I do, that her sliding around the notes is overdone, as are the witches' sneering voices. As Dido, Véronique Gens has a lovely voice and beautifully expresses the tragic nature of character, but the role lies a bit low for her pure soprano. She isn't quite able to offer the vocal richness of Lorraine Hunt (see above) or Jennifer Lane (see below), particularly in the role's lowest range. 

Finally, Christie's chorus seems to be largely made up of the soloists. While this was a common practice in Italian opera seria, which typically only had one short final ensemble, it would probably not have been the case for this opera, where the chorus has such a prominent role and is onstage simultaneously with the principals. Not only does Christie's small chorus lack the impact of a larger group, but as Ralph Moore points out in his survey, their voices don't blend particularly well. 

I find these drawbacks to be significant; however, my reservations are not shared by the Gramophone guide, which gives it a "Recommended" rating.

Collegium Musicum 90 conducted by Richard Hickox. Image source: Presto Music

Richard Hickox's recording on Chandos served as the soundtrack for a BBC film of the opera directed by Peter Maniura, broadcast on PBS in 1996. In the film the setting was shifted from ancient Carthage to the Renaissance, with Hampton Court picturesquely serving as Dido's palace. Apart from giving the filmmakers the opportunity to raid the BBC's vast store of sumptuous costumes, the film's main virtue is to give Maria Ewing's Dido a visual dimension. She was one of the great singing actresses of her generation, and seeing her performance (despite the occasional lip-synching problems) adds to its impact. 

The visual dimension doesn't hurt the handsome Karl Daymond's portrayal of Aeneas, either. And having an Aeneas that is visibly younger than Dido works well dramatically: it implies that Dido sees in Aeneas her last chance of a great passion.

On CD, as a purely aural experience, Ewing's operatic vocal style, and particularly her prominent vibrato, is more bothersome. Here is "Your counsel all is urg'd in vain," Dido's final confrontation with Aeneas after she learns that he is leaving:

https://www.youtube.com/watch?v=SW1Y18-dzyE

Ewing's dramatic engagement is not in question, but her vibrato is a matter of taste, and I prefer other singers in this role. But what's also apparent from this excerpt is that, perhaps due to his relative inexperience with opera, Hickox takes this dramatic scene at far too slow a pace; McGegan's and Christie's versions are both well under four minutes long, while Hickox takes nearly four and a half minutes. Although I think the film version is worth seeing, the combination of the slack pacing and Ewing's vibrato rule out the CD version of this performance for me. In his survey Ralph Moore disagrees, calling it "a top contender" and "highly recommendable."

Tafelmusik Orchestra and Chamber Choir conducted by Jeanne Lamon. Image source: Mark Morris Dance Group

This video preserves Morris' performance in the double role of Dido and the Sorceress in what is perhaps his most moving work. Jeanne Lamon's conducting emphasizes the opera's tragedy with slow tempos, perhaps to enable the dancers to more easily execute the gestural choreography. But as a purely musical experience it lacks the headlong, inexorable rush towards the final tragedy that McGegan's or Christie's version provides. 

Vocally, Jennifer Lane and Russell Braun are an excellent central couple. And, paralleling Morris' double role, Lane sings both Dido and the Sorceress (she is easily the best Sorceress in this survey). The double roles suggest that, as Joan Acocella writes in her great book on Morris (FSG, 1993), the Sorceress is "an anti-Dido," just as her coven of witches is the evil mirror image of the courtiers who surround the queen. The double casting may also suggest that the Sorceress is a self-destructive aspect of Dido. As Roger Savage has written:

It can be argued that. . .all the [female] characters in the opera are really personified aspects of Dido: Belinda and the Second Woman projections of her yearning towards erotic fulfillment, the Sorceress a formidable anti-self embodying all her insecurities and apprehensions of disaster contingent on her involving herself in any deep personal relationship, and the two solo witches nightmarish shadows of Belinda and the Second Woman. [1]

Barbara Willis Sweete's direction is sometimes distracting—we don't really need to see the singers floating on and off the screen again after their first appearance—but this film is an invaluable record of one of Morris' greatest works.

Highly recommended: McGegan, Harmonia Mundi; Lamon, Rhombus Media

---

1. Roger Savage, Producing Dido and Aeneas: An investigation into sixteen problems. In Michael Burden, ed. The Purcell Companion, Amadeus Press, 1995, pp. 445-468.