I’ve been so happy sitting outside with my computer looking at old romantic German poetry—pianists don’t get to practice outside, and I’ve always been jealous of people who can. And I keep feeling moments of We Live In The Future: I’m here on my back deck in my small town in Pennsylvania, working with texts from libraries across the world. The process is fun, in a hardcore nerdy way: read on if that sounds interesting, and definitely don’t if it doesn’t.
Agnes wrote lots of songs (“lieder”), mostly in German but a couple in English, and it’s tricky to figure out what the lyrics are, so that’s my main focus right now. Agnes was so gorgeously clear in her musical notation, but for words she uses an old German cursive that’s harder to decipher. I asked a Viennese pal to take a look and she said she couldn’t read it either. So it’s much better if I can find a printed version of the text somewhere else, and then map it on to what she wrote in the music. Once I find the lyrics, Kate the Stalwart Student Research Fellow adds them into the song when she enters it into music notation software.
Usually Agnes names the poet; when she doesn’t, I think it probably means she wrote the text herself (in some cases the card in the card catalog says she did), so those songs sadly go to the bottom of the pile for now. But there are lots of songs where she does name the poet, and often I can find that poet’s work; she set a couple of poems by Goethe, for example. Those two songs will be a really nice set, An den Mond (To the Moon) and Die Liebende Schreibt (The Lover Writes). It’s fun to find a poem, then look at the first few words in the printed version and see if the music manuscript looks like it matches. I tend to squeal when I find a match. When I do, I copy the words into a word document and add that document to the folder with the pages of the song, so Kate can add them into the notation. Then that whole folder goes into the folder called “Lieder with text available.” That folder has 13 folders in it now, so I’m hopeful that we’ll be able to publish a substantial body of Agnes’s work.
Sometimes the poets are published but out of print, or really obscure. I spent a while this morning looking at poems by Franz Keim; Agnes has a set of 5 songs with texts by him. Two I found, and I think the rest might be from the same large collection. I’m still looking for the other three, and it’s hard to see what the first words are, so it might be more of a challenge to find the others. But it’s fun to look at the words and try to figure out what they are and what they might match.
Here’s an example of the kind of thing I’m learning and exclaiming about on my back deck. Agnes has a song called “Lieder des Mirza-Schaffy No.1,” with no poet listed. At first I thought it was a set of songs (“Lieder” means “songs”) with lyrics by someone named Mirza-Schaffy. Well, sort of. Mirza-Schaffy was an Azerbaijan poet and teacher, and a German writer Friedrich Bodenstedt traveled to study with him, publishing a book about him including some translated poems. Bodenstedt later claimed the poems were his own original works, but apparently the original Persian versions exist. At any rate, Agnes set text from the famous German version by Bodenstedt.
Once I figured out that I was looking for the Bodenstedt book, I found a public domain text that’s a pdf of the entire book of poems…in an old English font that’s hard to read and thus hard to tell which poem might match with which handwritten lyrics. My heroic local librarian Clint—this whole project is making me appreciate librarians so much—ran the pdf of the book through an AI program that turned the images into readable text, and I was able to find and read the actual poem once it was in a friendlier font. It doesn’t have a title, but it’s the first poem from the book. Now I know why Agnes titled the song “Lieder des Mirza-Schaffy No.1” (I was hoping it meant one of many, but it’s just one). Once I found the poem, it turns out that other composers have set it too. There’s a translation at the incredibly helpful resource liedernet.
I’m excited and hopeful about another AI possibility that HLL Clint let me know about: AI software is learning how to read old handwriting, and I might be able to teach it to read Agnes’s handwriting. If I can, that means I can go back and photograph her letters and read them, which would be incredible.