This performance was given by the English Concert, conducted by Harry Bicket, as part of the biennial meeting of the American Handel Society which was held this year on the campus of Princeton University. The first half consisted of the suite from Purcell’s King Arthur, Handel’s concerto Op 3 no 2, HWV 313, and Telemann’s viola concerto in G, TWV 51:G9; in the second half we had Bach’s A minor violin concerto, BWV 1041, and a group of about half the movements of Handel’s Water Music, HWV 348-350, described in the programme as “selections from suites 1, 2 and 3”.
Harry Bicket led the band from the harpsichord in the approved modern manner; the performance was magnificent, with well-judged tempi (some early music bands do tend the rush the allegros!), and superb playing by all concerned, including one of the horn players who did the Water Music one-handed. The Purcell was the well-known selection of instrumental movements from the opera, ending with a fine chaconne, an art-form in which he was such a master. The three concertos were equally good, especially pleasing as they are among my most favourite Baroque works.
The Water Music selection was just as impressive in performance, but for me there was a problem. We are all grateful for the remarkable development over recent decades in “authentic performance” (horrible term!) of Baroque music, and the English Concert is high on the list of groups which have mastered the style and techniques of the period; but it is also essential that they keep up to date with the parallel development of sound critical editions which are the basis of what is performed; and this was, unfortunately not the case here. Everyone knows the Water Music, a suite of 22 movements performed on the River Thames on that memorable evening in July 1717; but the absence of a surviving autograph has led to much confusion and speculation about the text of these movements, and especially about the order in which they were played. Until recently the earliest known manuscript copies of the score were several which date from the late 1730s, and these give conflicting accounts of the order of the movements: the copy in the Barrett Lennard Collection in the Fitzwilliam Museum (c.1738) has 10 movements in F major/d minor scored for strings, oboes, horns and bassoons, followed by five in D major, with trumpets added, then 7 in G major/minor, with a gentler scoring of strings, flutes, recorders and bassoons. It is now clear that this arrangement was made for a patron for whom it was convenient to have the movements arranged thus so that they could be played in separate groups each of which had the same instrumentation. About 1950 a number of scholars in Cambridge, among whom Thurston Dart was prominent, and for whom this manuscript was the earliest known copy of the work and was readily accessible, concluded that this order suggested that there were not one but three suites, which may not all have been played at the same time (Dart suggested that the third group could have been played at the supper in Chiswick, or on another occasion altogether). So was born the “three suite theory”, hence the three HWV numbers and the first edition in the Hallische Händel-Ausgabe, published in 1962, where the music appears in this form. Others such as Roger Fiske joined in, and published editions using the same order, and the heading in the programme note of the Princeton concert followed this tradition.
However no-one had ever questioned why the earlier published scores by Arnold in 1788 and Chrysander in 1886 had a different order, with the D major movements and those in G major/minor interspersed with each other, and ending with the brilliant D major “Trumpet Minuet”. As more manuscript sources came to light in the late 20th century, so doubts increased, and I was commissioned to do a revised edition for the HHA. As I worked on it, I had in 2004 the good fortune to make a remarkable discovery in the archive of the Royal Society of Musicians: a complete manuscript copy of the work, written by two of the copyists who were working for Handel about 1717-18, RM1 and Linike; and the watermarks and rastra of the paper give a date of c.1718,only a year after the performance on the river. The MS gives the music in the order used by Arnold and Chrysander; close analysis of this score shows beyond doubt that it was Arnold’s source, and Chrysander followed Arnold, explaining in a contemporary article (1887) that it seemed the most satisfactory sequence of movements. The new HHA score, which also has an edition of the Fireworks Music, edited by Christopher Hogwood, was published in 2007.
As Harry sat at the harpsichord I could see that he was using the 1962 HHA score, which has a number of textual errors as well as the wrong order of movements: of course he gave only a selection of the movements, but those were in the 1962 order, except for the famous D-major Hornpipe, which he gave as a conclusion – quite a good idea, but the genuine finale, which he had performed in its place as no. 13 in the 1962 score, should have been the Trumpet Minuet. It is curious that the writer of the excellent programme note mentioned the recent research, but this had not been adopted!
Postscript: as a follow-up to Peter Schofield’s excellent account of Harry Bicket’s performance of Radamisto in Birmingham (accessible here) I have to report that it was performed by the same team equally brilliantly on 24 February 2013 at the Carnegie Hall in New York, and of course we all went from Princeton to hear it. I have nothing to add to Peter’s review, except to comment on the edition used, as a reinforcement of my argument above: this time Harry was bang on, using the HHA volume (also edited by me in 2000), and giving what I believe to be the first performance of the opera in modern times in the version which I have suggested is the best, that of November 1721, in which the somewhat redundant role of Fraarte was omitted. So press on with using up-to-date editions!