Piano Recital: Kit Armstrong
Byrd, Sweelinck, Bull, Bach
Druckerei, Baden/CH, 2016-02-26
2016-02-29 — Original posting
2016-09-29 — Brushed up for better readability
A Personal Note
I had of course heard of Kit Armstrong before; I have not actively followed his achievements prior to this concert, but still, just by the name of the artist, this concert announcement caught my interest. However, when I read the list of compositions, I had instant reservations. I don’t want to deny that I have a strong preference for performing baroque (even more so pre-baroque) compositions on original instruments or replicas. In the case of keyboard compositions, this basically precludes the use of a modern piano for such works.
Pre-Classical Works on a Modern Piano
I don’t want to dwell on this topic in too much detail here, but there are two main aspects to consider:
Articulation, Sound, Dynamics
The sound characteristics of a modern concert grand (articulation, sound, dynamics) have nothing in common with those of a harpsichord or related instrument. It is virtually impossible to “imitate a harpsichord” on a modern concert grand;
The Tuning Aspect
A modern concert grand (all pianos, essentially) uses equal-temperament tuning: if we ignore the rare phenomenon of perfect (absolute) pitch hearing and people with synesthetic hearing, a given chord (say, a major triad) essentially sounds the same in all keys. In medieval times, equal temperament tuning was unknown. Hence, the keys were not “balanced”. A given chord may sound beautiful, shining, with pure thirds and/or fifths, while in other keys, the same chord might sound more dissonant, “impure”, maybe even hardly playable. The issue of having to distribute a certain amount of dissonance among the various keys is inherent with all tuning (see my separate post and earlier posts referred to therein for a brief discussion on this topic):
- in so-called “mean-tone temperament” and related tunings in pre-baroque times, not all keys were playable, while others exhibited extraordinary purity;
- in baroque times, numerous attempts / compromises attempted to retain some of the “mean-tone purity” of certain keys, while making more “distant” keys playable as well (Bach’s “Well-tempered Clavier” is the fruit of such efforts);
- finally, equal-temperament tuning emerged in classic times, making all keys / tonalities equal and accessible to the same degree—at the expense of losing the purity and “spreading the dirt (dissonant keys)” equally across all keys.
The Purist’s Point-of-View
As for the first point: harpsichords aren’t exactly cheap and available to everyone. Of course, one can’t stop pianists from playing and enjoying baroque and pre-baroque music on their instrument. One may argue that a concert also features educational aspects. In that vein, there are certainly musicians—Gustav Leonhardt (1928 – 2012) was a prominent example—who would vehemently defend that compositions by Bach etc. ought, must be performed on instruments that are as close as possible to what the composer had in mind.
I tend to be close to that point-of-view: Bach’s music doesn’t need the dynamic range, the sound features of the piano, but it profits enormously from the articulation richness of the harpsichord, the possibility to express “Klangrede” (phrasing through agogics and articulation) to a degree that the piano simply can’t match.
The Other Extreme
Ever since Glenn Gould (1932 – 1982) broke into the market with his interpretations of Bach’s Goldberg Variations (see also below), at least these Variations have retained their position in the piano concert repertoire. As long as it helps making Bach’s music (or baroque music in general) popular among today’s audiences, one can hardly argue against such performances.
The Relevance of Tuning
As far as the tuning aspect goes: the Goldberg variations center around G major. In other words: they don’t explore “distant” keys, they don’t play with the sound color differences across the various tonalities. So, the use of equal temperament tuning isn’t all that detrimental to this composition. However, in the case of William Byrd, John Bull and contemporaries, the situation is different. These composers were relying on pre-baroque tuning (in the “mean-tone temperament” family). Without the pure thirds in such tuning, I find that (many of) these compositions lose essential components in their expression, or the impression on the listener.
In that vein: sure, one can play it on any keyboard instrument. If such performances can get new audiences interested in early music—why not? But in this case, I would strongly argue that one should declare it as “peek preview”, intended to encourage new listeners to explore that repertoire in a “proper” performance on period instruments, and with historic tuning: I remember only too well how I first listened to music by Byrd on the piano in my youth. What an eye opener, what true joy it was to me, when I then listened to the same music played in a “historically correct” performance, for the first time!
Possible Benefits of Performances on the Piano
One possible counter-argument against a “purely historic” performance in concert: the ears of most listeners have so much adopted equal-temperament tuning that they may not (not instantly, at least) “understand” / feel the benefits of the colorful old tuning modes. I could easily imagine that a newcomer would find this “wrong”, “strange”, “dissonant” or “out of tune”. In that sense, performing such music on a modern instrument may serve to “lure” new listeners into that music, especially since it is hardly played in concert at all!
On the Artist
Kit Armstrong was born 1992 in Los Angeles, into a non-musical family. Very early on he developed into a true child prodigy, talking very early, showing interests in sciences (physics, chemistry and mathematics), languages (his German is perfect and virtually accent-free), and music (composition and piano).
Instead of listing all his achievements here, let me just quote from Wikipedia:
At the age of 5, and without access to a piano, he taught himself musical composition by reading an abridged encyclopedia. He subsequently started formal studies in piano with Mark Sullivan and in composition with Michael Martin (1997–2001).
Armstrong has always pursued music and academic education in parallel. He attended Garden Grove Christian School (1997–1998), Anaheim Discovery Christian School (1998–1999), Los Alamitos High School and Orange County School of the Arts (1999–2001). While in high school, he studied physics at California State University, Long Beach, and music composition at Chapman University.
At the age of 9, Armstrong became a full-time undergraduate student at Utah State University studying biology, physics, mathematics as well as music (2001–2002). In 2003, Armstrong enrolled at the Curtis Institute of Music studying piano with Eleanor Sokoloff and Claude Frank, while simultaneously taking courses in chemistry and mathematics at the University of Pennsylvania. 2004, Armstrong moved to London to continue his music education at the Royal Academy of Music studying piano with Benjamin Kaplan and composition with Paul Patterson, Christopher Brown and Gary Carpenter. In parallel, he studied pure mathematics at the Imperial College London (2004–2008).
Armstrong received a Bachelor of Music degree with First Class Honours from the Royal Academy of Music in 2008 and a Master of Science degree with honours from Pierre-and-Marie-Curie University, Paris, in mathematics in 2012.
Since 2005, Kit Armstrong has apparently studied regularly with Alfred Brendel (*1931). Let me just add that his Wikipedia entry features an impressive list of compositions. Numerous famous artists, orchestras and institutions have commissioned compositions with Kit Armstrong. His career as pianist took him to top concert venues throughout the U.S. and Europe, both as soloist, as well as with renowned orchestras and conductors. Kit Armstrong’s repertoire covers a broad range from pre-baroque up to contemporary music.
On this Concert
This was another concert in the context of Piano District, organized in parts by the pianist Oliver Schnyder in Baden / Switzerland, in the “Druckerei” (print shop) in Baden/CH, i.e., a hall that was formerly housing the rotary print press for the local newspaper, see my earlier reviews from that venue. As in previous piano recitals, the piano was a Steinway B-211; a model D concert grand would not fit into the elevator. In this concert, the ca. 240 seats of this venue were completely sold out. After a short announcement by Oliver Schnyder, Kit Armstrong entered the stage — and started with an introduction of several minutes in perfect German. I won’t translate that here, but rather start with his performances / interpretation:
Byrd: Hugh Ashton’s Ground
The piece “Hugh Ashton’s Ground” (also “Hughe Ashtons Grownde”, BK 20), by William Byrd (1540 – 1623) is a Passacaglia. That is a set of Variations upon a fixed bass line / harmonic progression. Here, the theme is a “ground” by Hugh Aston (ca. 1485 – 1558), an English composer of the early Tudor period. The piece—written in 3/2 time—is at a slow, measured pace, starting with heavy ornamentation (frequent trills / mordents). It progresses to shorter note values, up to quaver sextuplets and semiquaver passages (the term quaver refers to modern notation of this music).
Kit Armstrong played this at a perfectly adequate pace. I found his ornamentation to be well-adapted to that music, even though on a harpsichord mordents will probably be played slightly slower, will also receive a different weight in the melody. I quite liked the overall dramaturgy, the phrasing / arches in each variations.
From a historic perspective, I have some reservations towards the tempo variations. The more virtuosic variations in the center were substantially faster than the beginning, the ending featured a fairly pronounced rallentando.
Dynamic variations: the piece started p / mf, in the center, f / ff was reached, the ending faded away in pp. I would not go as far as prohibiting dynamics: there were indeed keyboard instruments that did allow for limited dynamic variation (clavichord). However, my preference would be limited dynamics.
As indicated above, the modern piano has restrictions: not just in articulation, but primarily in that it lacks the richness in colors of the pre-baroque tuning.
In a historic setting, a silent ending is not just inappropriate because early instruments could not do this: in pre-baroque tuning, the ending is associated with brightening up from more dissonant tonalities into chords with lucid, bright, pure thirds. In a way, there is a big relief at the end, which one certainly would not want to hide in softness. But the approach selected by Kit Armstrong is understandable in the sense that he needs to substitute dynamics and some degree of rubato for features which the concert grand cannot offer.
Even with these limitations / alterations, the music remained impressive—actually, very likely more understandable to the modern, inexperienced ear than a “proper, historic” performance.
Sweelinck: Variations on “Mein junges Leben hat ein End”
The second composition consisted of the 6 Variations on “Mein junges Leben hat ein End“ (my young life is coming to an end) by Jan Pieterszoon Sweelinck (1562 – 1621). This is a melancholic piece that may originally have been written for the organ.
Kit Armstrong played this with good, careful articulation, with an excellent feel for the overall dramaturgy, and with limited dynamic excursions. Sweelinck doesn’t play with colors of chords / tonalities, or at least, to a lesser degree in these variations than Byrd and Bull in their keyboard music. Also, the piece is devoid of annotated ornaments. I suspect it is not known what type of ornamentation (and how much) would have been used around 1600. The artist added very few extra ornaments (typically at the end of a variation). I found his choice of ornamentation and playing style in general perfectly adequate for the modern piano. To my “historic ear”, the most acceptable / adequate performance in the first half of this concert: well done!
Bull: 30 Variations on “Walsingham”
John Bull (1563 – 1628), an almost exact contemporary of William Byrd, write 30 Variations on the (then) popular 12th century folk song “Walsingham” (“Have with yow to Walsingame”). While even on a modern instrument this remains a fascinating piece of music, I think we are back “on shaky grounds”, as in the case of William Byrd’s variations above (or see the note on the first encore below): in my view, these 30 variations on a simple, 8-bar folk tune not only rely upon the “language” (the “syntax”, the ornamentation, articulation and sound characteristics) of a plucked instrument, but they also clearly make extensive use of the extra colors (both towards cleaner, as well as more dissonant intervals) obtained with a typical, historic tuning (e.g., quarter-comma mean-tone temperament tuning)—so, major losses were to expect!
As anticipated, the performance on the modern piano had something of a black-and-white photograph in comparison to the colorful, historic original. Yes, it still is pleasant, even interesting music—but it is so far from what the composer must have envisaged! Apart from the lacking colors, I noted various issues:
- Many of the ornaments were very elegant, but soft (relatively speaking), if not occasionally somewhat superficial. Clearly, this is historically unrealistic. On a harpsichord, the “small notes” in ornaments such as mordents, trills, etc. sound at exactly the same volume as the main notes. This makes them stand out more than in Kit Armstrong’s interpretation.
- The pianist tended to use “small-scale dramatization” through rubato and dynamic variation (crescendo / decrescendo). Certainly, the latter is a “pure piano feature”. The amount of rubato is debatable, at the very least.
- In variations 7 – 9, 11 – 17, and 22 & later ones, the artist chose a very fast, virtuosic tempo. This probably needs to be seen as an attempt to use virtuosity as substitute for the lacking articulation richness of a harpsichord. To me, this was virtuosic, maybe even fascinating and very artful, but pure piano art, far away from Bull’s original; the extreme in that sense was variation 27, which reminded of a Boogie-Woogie.
- The last variation (30) brought a relatively abrupt return to the initial, slow / measured pace (or even less). Again, Kit Armstrong ended with a strong rallentando and pp—at a point where the composition ends in brightest, lucid major chords that a harpsichordist is unlikely to “hide” in pp.
As stated, it is fascinating music even if “filtered” through a modern piano, but…
Bach: Goldberg Variations, BWV 988
Johann Sebastian Bach (1685 – 1759) first published his “Aria mit 30 Veränderungen“, i.e., the Aria with 30 Variations, BWV 988, known as “Goldberg Variations”, in 1741. As mentioned above, this extensive set of variations doesn’t suffer from modern tuning as much as the earlier music played before the intermission. However, the restrictions in the area of articulation and dynamics of course remain the same. Articulation may be even more of an issue, given that Bach was such a master on all keyboard instruments. On the other hand, the Goldberg Variations are extremely popular among pianists (both in concert and in the CD repertoire). Invariably, any pianist will be facing comparison to famous performances, such as Glenn Gould‘s several recordings, or of course any of the many, more recent ones by artists such as Grigory Sokolov, etc.
Bach’s Aria is a piece of 16 + 16 bars, both parts with repeat signs. That structure is maintained through all variations. However, as a performance with all repeats lasts well over an hour, Kit Armstrong played the variations without any repeats.
The composition is organized in groups of three variations. In each group, there are first two variations exploring various keyboard techniques: most are for a single keyboard, some explicitly for a harpsichord with two manuals. The latter are a bit more tricky to play on a piano with its single keyboard. Some of the variations also explore musical forms, such as a fugue, or a French Overture. In each group, these two variations are followed by a canon “a 2” (two-voice canon). The first canon is “in unison” (both voices at the same level), the second one has the two voices separated by an interval of a second, etc., up to variation 27, which is a “canon in the ninth”. Most canons also feature an accompanying third voice.
Variation 30 is not a canon, but a “Quodlibet“. That is a polyphonic piece, where multiple popular folk songs are mixed / intertwined, often even as an improvisation. The composition concludes with a return of the initial Aria.
There is also a “super-structure” of 15 + 15 variations, in that variation 16 is a (“French”) overture, in a way starting anew.
The Performance — Dealing with Heritage
Let me start with some general comments on Kit Armstrong’s interpretation here: the artist must have heard other interpretations. I assume he is familiar with what I consider the two extremes: for one, Glenn Gould’s “sporty”, virtuosic-artistic 1954 recording, and at the other end the highly intellectual (and emotional), final 1982 performance by the same artist. In the latter, Gould wanted to highlight / expose the continuity between the variations, their linkage not just to a common theme, but to a common pace.
In listening to the concert performance, I felt that Kit Armstrong didn’t try following any of these two. He freely formed each variation to what he felt is best for that segment. The transitions between the variations felt natural, not constructed, purely intellectual or artificial in any way. It’s his personal interpretation, for sure, definitely valid as a general approach. As already in the earlier pieces, I found his ornamentation well-adapted to what is today considered “proper, baroque style ornaments”. In fact, he was mostly close to what a harpsichordist or maybe organist might do on historic instruments.
The Performance — Comments
Obviously, it’s a very multi-faceted composition. I won’t comment on each of the variations, but for reasons of clarity, let me still organize my comments by variation:
See the general comments above
1st Variation (a 1 Clav.)
Not sporty, for my taste dynamically too expansive in the second part
2nd Variation (a 1 Clav.)
Simple, clear — good!
3rd Variation (Canone all’ Unisono, a 1 Clav.)
Fairly slow, not aiming for tempo continuity. Interestingly, the second canon voice was kept softer than the first one. I find this questionable, as a canon is usually seen as a set of equal / competing voices.
4th Variation (a 1 Clav.)
Measured, not sporty, clear
5th Variation (a 1 ovvero 2 Clav.)
6th Variation (Canone alla Seconda, a 1 Clav.)
Too much dynamic variation / build-up for a variation “a 1 Clav.”
7th Variation (a 1 ovvero 2 Clav.)
Too fast in the ornaments, somewhat superficial in the demisemiquaver motifs (a disadvantage of the modern piano!)
8th Variation (a 2 Clav.)
Maybe a bit on the soft side in general, with the ascending or descending quaver chords standing out rather strongly
9th Variation (Canone alla Terza, a 1 Clav.)
10th Variation (Fughetta, a 1 Clav.)
Very “pianistic” (as opposed to derived from a harpsichord version), to me, too fast & smooth in the trills and other ornaments
11th Variation (a 2 Clav.)
12th Variation (Canone alla Quarta)
Again, a “pianistic” solution, in that the pianist used dynamic highlighting to indicate the start of a (potentially hidden) canon voice — Bach’s music should need such help, i.e., I regard this as unnecessary, though certainly not a “deadly sin”.
13th Variation (a 2 Clav.)
Kit Armstrong took this very slow: too slow in my opinion, given that the piece is written in 3/4 time. The fast figures turn into melodies, rather than being taken as (sort of) ornaments. However, that’s an approach that most pianists take in this variation.
14th Variation (a 2 Clav.)
Here, I find the trills (and related ornaments) and staccati a bit too flashy, ostentative, sporty. But apart from that, I like the clear dynamic distinction between the two hands / keyboards (in such cases, dynamic differentiation is legitimate, desirable, even a necessity!)
15th Variation (Canone alla Quinta, in moto contrario, a 1 Clav., Andante)
16th Variation (Ouverture, a 1 Clav.)
I quite liked the double-punctuated first part of this French Overture: very good in the ornaments, etc.; the polyphonic second part was slightly stiff / rigid?
17th Variation (a 2 Clav.)
A very pianistic, almost murmuringly soft variation—but too pianistic for my taste.
18th Variatio (Canone alla Sesta, a 1 Clav.)
That’s a canon that I enjoyed for its clarity and simplicity.
19th Variation (a 1 Clav.)
I liked that, too, maybe except for the crescendo towards the end.
20th Variation (a 2 Clav.)
21st Variation (Canone alla Settima)
22nd Variation (a 1 Clav., Alla breve)
23rd Variation (a 2 Clav.)
A bit too sporty / pianistic, partly too flashy
24th Variation (Canone all’ Ottava, a 1 Clav.)
Very pianistic, (too) smooth in ornaments and fast figures
25th Variation (a 2 Clav.)
For the interpretation on the piano, this is a tough variation. Such slow, heavily ornamented pieces are inevitably very far from the interpretation on a harpsichord, and hence far from the composer’s intent. But within this restriction, this was a good interpretation: not too overblown / loaded, etc.
26th Variation (a 2 Clav.)
The semiquaver triplets were very (too?) smooth & soft, but more importantly, to me, the two punctuated melody voices were too mellow, the punctuations often softened, almost to triplet figures.
27th Variation (Canone alla Nona, a 1 Clav.)
In general, I found this variation better than the preceding one, though I did not like the tied quavers: Kit Armstrong turned these into (fairly strong) syncopes. That’s maybe a nice idea, but not in Bach’s notation, for sure!
28th Variation (a 2 Clav.)
Another, very pianistic interpretation, with its smooth trills, the martial, jumping melody voices, and the pronounced crescendo towards the end.
29th Variation (a 1 ovvero 2 Clav.)
This was maybe the weakest, most “piano-distorted” variation in this interpretation, with some superficial motifs for one, but also some over-articulated figures.
30th Variation (Quodlibet, a 1 Clav.)
Another bit that I disliked in this interpretation. The biggest flaw was that the artist focused on one single folk song among the several that are mingled into this piece. He strongly highlighted every single occurrence of “Kraut und Rüben haben mich vertrieben“. This neglected all the other melody components. It also made the piece sound simplistic, even boring, repetitive (even without repeats!!). There’s much more in this than what we heard.
Aria da Capo e Fine
The Aria da Capo was OK again, in itself, though Kit Armstrong inserted a lengthy pause before starting it: maybe he did not find a logical / natural transition to this ending?
To summarize Kit Armstrong’s interpretation of the Goldberg Variations: I would characterize this as a “kaleidoscope view”. It aimed at showing / exposing richness and variability, rather than aiming for formal stringency (as, e.g., Glenn Gould in his 1982 recording). It was definitely interesting, entertaining and immediately comprehensible / enjoyable by a general audience. Note that with this statement, by no means I want to belittle Kit Armstrong’s interpretation / achievements. Many of the critical remarks above are due to the choice of (modern) instrument. I don’t mean to blame the artist for not using / playing a harpsichord. That’s not just a matter of choosing an instrument. The harpsichord requires an entirely different approach in terms of finger / playing technique.
Encore 1 — Byrd: Variations on “Walsingham”
As expected, the Goldberg Variations did not fail to impress the audience. There was a (partly) standing ovation; Kit Armstrong took the (obvious) pleasure of following up on John Bull’s 30 “Walsingham” Variations, by playing William Byrd‘s counter part on the same theme, the 22 Variations on “Have with yow to Walsingame”, BK 8.
To me, this is one of Byrd’s nicest, most beautiful compositions. It is so beautiful that it even survived its “translation” to the modern piano, making me forget about most of the inherent limitations on this instrument.
I could listen to this music over and over again, almost forever—especially, where it is played on a harpsichord or virginal. I would like to encourage everyone to give such a historic interpretation a serious try. It may take some active / conscious listening to be able to enjoy it in every detail, but I’m sure you won’t be disappointed!
Encore 2, Conclusion
Kit Armstrong added a second encore, (based on) a ground by Thomas Preston (a predecessor to William Byrd, who died ca. 1563), most likely the Ground “Uppon la mi re”. That is definitely more archaic than Byrd / Bull & Co. as a theme. It was rather (too) jazzy in this encore performance. I’d have preferred to leave the concert with the soothing impression of Byrd’s marvelous variations!
I do have my reservations about most of that music on the modern piano. However, I really liked Kit Armstrong’s unpretentious appearance. He was modest, but swift when giving explanations or responding in interviews. As soon as he started playing, he appeared introverted, totally devoted to / immersed in his playing. He exhibited minimal body language / body motion. The only exception: given his small stature, he sometimes needed to move sideways on his piano bench. He is definitely an artist at the start of a promising career, and somebody to keep an eye on!
For the same concert, I have also written a (shorter) review in German for Bachtrack.com. This posting is not a translation of that German review, the rights of which remain with Bachtrack. I created the German review using a subset of the notes taken during this concert. I wanted to enable my non-German speaking readers to read about my concert experience as well. Therefore, I have taken my original notes as a loose basis for this separate posting. I’m including additional material that is not present in the Bachtrack review.
I usually don’t mention “third parts CD recordings” when discussing a concert. However, for reasons described in the first section above, let me make an exception here. I’m covering the above compositions by William Byrd and by John Bull. I have previously referred to these recordings, e.g., in my Listening Diary from 2014-08-24 and also other posts. I don’t need to list recordings of Bach’s Goldberg Variations: there are plenty (countless) of these, even if we were to look for recordings on harpsichord or related instruments alone.
William Byrd: The Complete Keyboard Music
Davitt Moroney — harpsichord, virginal, chamber organ, clavichord
hyperion CDA66551/7 (7 CDs, stereo); ℗ / © 1999
Booklet: 200 pp. English/French
John Bull: “Dr. Bull’s Jewel” — Keyboard Music
Kathryn Cok — harpsichord and virginal
The Lyrichord Early Music Series, LEMS 8060 (CD, stereo); ℗ / © 2007
Booklet: 8 pp. English/Dutch