As I did with the early symphonies, I would like to start here with putting some time and date context to the early keyboard sonatas. It’s nice to have a dream, and I will cherish this one forever. Fulfillment is not right around the corner though, I fear.
As I have alluded several times in previous essays, Haydn was composing sonatas as teaching aids for his students. In addition, he was composing them on a more advanced level as composition practice for himself, and for the use of Kenner, which is to say, people who really knew and understood music, the scholars and professionals of the day. There are currently three accepted listings of the sonatas. The first, dating from 1918 and which serves as the basis for Hoboken, was compiled by Karl Päsler. It consists of 52 sonatas, and thus is the basis for the total which has been used for nearly a century to define Haydn’s sonata oeuvre. The second list is contained in the massive chronicle of the Joseph Haydn Society called Josef Haydn Werke (or JHW in the musicology world). The sonata section of JHW was completed in 1958 by Georg Feder, and it later goes on to become the chronology listing in The New Grove Haydn. The third and possibly best-known is the list of Christa Landon, completed in the mid-1960’s, which has given us, among other things, the set of numbers used to enumerate the works today. When you see “Sonata #32”, for example, it is a Christa Landon number, not a Hoboken one.
Now, this would all be plain as day if the three lists showed the same (or similar) results. I wouldn’t even need to write this, there it would be, c’est tout simple! But of course, this isn’t the case. It is generally accepted now that from before Haydn’s employment by the Esterházy’s, there are only eighteen extant sonatas. Päsler/Hoboken show fifteen, but Landon and Feder had more material to work with 40 years on. In addition, Feder approaches things just a bit differently than the others. He realizes that there are two different classes of works in play, and so divides them up into two groups of nine. He ends up with what he calls ‘Nine Early Sonatas (for experts)’ and ‘Nine Small Early Sonatas (for students)’. These he ranks in order of increasing complexity to create an earliest --> latest list. The stumbling block comes when one realizes the two lists must be integrated since the groups of sonatas weren’t composed one group after the other, but concurrently. In most other ways though, it is a very nice idea which tells us much about the type of work it is and its level of sophistication. I will call these two groups SES (small early sonata) and ES (early sonata) in the list, and so the JHW numbers will appear twice each. So here are how those lists break down;
Päsler/Hob. JHW/Feder C. Landon
1 SES 1 10
7 SES 2 2
8 SES 3 1
9 SES 4 3
10 SES 5 6
deest* SES 6 4
deest SES 7 7
3 SES 8 14
4 SES 9 9
16 ES 1 deest
5 ES 2 8
12 ES 3 12
13 ES 4 15
14 ES 5 16
6 ES 6 13
2 ES 7 11
deest ES 8 17
deest ES 9 18
* deest = doesn’t appear here
See how straightforward it all is? In truth, this barely scratches the surface of the complexity involved. For example, Landon’s ‘missing’ #5 is Hoboken 11. Since it doesn’t appear to be among Feder’s first eighteen, he inferentially doesn’t consider it to be actually ‘#5’. In the latest revision of the New Grove Haydn, it shows up in the Anhang (appendix) of unsure works. One of many which require more work to figure out.
Does all this mean that you might as well just give up? Not at all! Not only is work going on in this field at all times, in some cases awaiting nothing more than publication to resolve the mystery, but the real key lies in being flexible in one’s thought process, enough so as to be able accept that nothing is written in stone. So while this page doesn’t offer the easy answers, I will certainly present you with the most up to date information I can find, and hopefully we will come to grips with this thorniest of subjects.
Thanks for reading!