50. That ain’t the way I heer’d it

I grew up listening to the radio a lot.  South Carolina Public Radio in the 1980s played what you would expect from a listener-supported channel:  classical music and jazz; NPR news; and shows like Prairie Home Companion, but also a good bit of programming picked up from the BBC World Service (like the delightful panel games My Word! and My Music!) 

Sometimes American shows from half a century earlier would be re-broadcast.  Of these, I remember particularly Fibber McGee and Molly.  While not exactly a show about nothing, it did manage to succeed without much of a plot.  Instead, it consisted of a continuous series of running gags—most famously the avalanche from a precariously-stacked hall closet, which I'm sure was a perennial challenge for the sound effects team.  Many of the gags involved phone calls or visits with recurring characters.  One of these was known simply as “Mister Old Timer.”  The set-up was the same every time:  Fibber would tell a joke, which Mister Old Timer would then improve, starting off with “That’s pretty good, Sonny; but that ain’t the way I heer’d it.”

Another show I grew up hearing—or really overhearing, since it was usually my dad who tuned in—had the astonishingly pretentious title Adventures in Good Music.  (Of course, “good” meant dead, white, and male.  It was a massive dose of mainly nineteenth-century German repertoire—piano, chamber, and orchestral—the canon of instrumental music.)  It was hosted by the avuncular Karl Haas, who had a spectacular radio voice; more than that, he always had something to say.  (Bootleg recordings of a number of his broadcasts are available on the Internet Archive.)  He opened every episode—and it was broadcast every weekday—by performing anew the opening sixteen bars of the second movement of Beethoven's “Pathétique” Sonata, op. 13.  So I heard that opening many, many, many times.  What I heard—what my brain perceived—was this:

But, of course, that is not what Beethoven wrote. I was astonished when I later saw a score and realized I had been mishearing it for many years.

SOURCE:  cropped scan of Bärenreiter BA10851, ed. Jonathan Del Mar (2012), p. 11

Well, that ain’t the way I heer’d it.  The b-flat I heard on beat two of the second bar was actually just in the accompaniment.  If had paid more attention—if, that is, my mind had not already decided what I had heard—I would have noticed an oddity when the accompaniment becomes more complicated: the b-flat isn’t there on the beat (in bar 10).  But I didn't.  And even now it is hard for me not to hear it “my” way, unless I’m actually playing it.  My college roommate had an expression for this sort of thing:  it puts a crease on your brain.  No matter how you may unfold it and try to flatten out, that initial impression is indelible, like it or not.

In a graduate school ethnomusicology seminar we were assigned to read Bruno Nettl’s magisterial text The Study of Ethnomusicology, which has appeared in ever-more expanded editions.  The copy I read at the time had the subtitle Twenty-Nine Issues and Concepts; the second edition on my desk at the moment says Thirty-one Issues and Concepts; and I see that before his death in 2020 the subtitle of the third edition had become Thirty-Three Discussions.  It is a superb book, welcoming newcomers into a very dynamic and complex field of study.

One of Nettl’s early chapters deals with transcription—writing down music in some sort of notation—and I recognized myself in the title of his chapter:  “I Can’t Say a Thing until I’ve Seen the Score.”  He gives a brief history of transcription as a component of ethnomusicological study—recognized from the start to be problematic, but in some sense unavoidable.  “We have an excessive association of music and notation, equating conventional Western notation with musical sense and musical reality.  We can’t quite remove ourselves from this hallmark of urban Western academic music culture....  To the Western ethnomusicologist, Western notation makes the music seem like ‘real music’.”  [2nd ed. pp. 87f.]  Sure, music exists—thrives even—without notation; but the academic study of it has relied on notation to reify as a visual entity some salient detail of the (sounding) music.  And it doesn't matter what music you’re talking about:  the Beethoven above and my misperception of its performance are both presented to you here as transcriptions.  One is what Beethoven presumably “heard” as he composed (“audiated” is the musical jargon); the other is what I perceived in Karl Haas’s nightly broadcast.  The musical notation isn’t the music, but I use it by default because it helps me to communicate it to you.

I have mentioned in a previous post Malcolm Bilson's engaging video lecture “Knowing the Score,” in which he asks the interesting question “How did he write it down?”  That is, how does a composer (and at the moment Bilson is talking about Prokofiev and Bartók) choose to notate the music?  This is especially fascinating when we can compare the notation with the composer's own recorded performances, but is a worthwhile question of any written compositions.  Notation betrays priorities:  it naturally fails to account for things so obvious as not needing to be made explicit.  A great demonstration of this was performed on Nov. 2, 1963 at the eighth annual meeting of the Society for Ethnomusicology and published nearly a year later in an issue of the journal Ethnomusicology—available via JSTOR, but alas behind a pay-wall.  (There is, however, a 2014 open-access Twentieth Century Music article which opens with a discussion of this famous example.)  This session featured an examination of transcriptions made by four distinguished scholars, each asked to transcribe the same field recording.  It was not a competition in the sense of who had transcribed it best, but rather a consideration of what different transcriptions could reveal about the music and our varying perceptions of it.

  

SOURCE:  reproduced from Jason Stanyek, Forum on Transcription, in Twentieth-Century Music Vol. 11 no. 1 (2014), pp. 101-161; here cropped from p. 102.

What this demonstrated most clearly was that we hear different things.  Each of the four transcriptions emphasized certain nuances of the music while ignoring others.  Hardly a surprise.  Have you ever been at a piano recital when the sustain pedal squeaked?  (I hope not.)  I have, and what I found is that, once I had noticed it, the only thing I could hear was the squeak-squeak-squeak of the pedal.  Or take this moment:

SOURCE:  cropped scan of J. S. Bach, "Der Geist hilft unser Schwachheit auf," BWV 226, bb. 154-161 in the BG edition.

Settling Scores · J. S. Bach: "Der Geist hilft," BWV 226, bb. 154-160

All I can hear in this recording are the almost oppressive S sibilants reverberating around the Chapel of King’s College, Cambridge.  I can't hear the fugue for all the hissing, and if I was transcribing this without any awareness of the idiom in which it is written, I might think there was an essential percussive element, which is actually incidental and of which Bach was probably utterly unaware.

SOURCE: discogs.com

In that same gradate ethnomusicology seminar, we were assigned to make transcriptions of anything that caught our attention, just for the discipline of transcribing.  I had run across a fascinating disc released by Lyrichord featuring the choral songs of the Church of God and Saints of Christ, an African-American denomination with a distinct oral-tradition repertoire.  So I set about transcribing the very first item on the disc, a piece called “O Lord, How Excellent is Thy Name,” the composition of St. Cuetta Connors of Dayton, Ohio (d. 1970).  The 1990 Lyrichord recording is available on iTunes and Spotify, and is worth a listen.  My transcription (which I post on a separate page) is revealing in its naïveté.  It presents this piece as if Connors had written [sic] a conventional [ = W.A.S.P.] four-part anthem, which the recorded performance was somehow not quite living up to.  I note at the beginning “this transcription assumes a constant tonic of C major,” as the pitch changes (mainly downward) as the piece progresses.  In other words, I made no effort to reflect the changing pitch as sung on the recording, but rather presented some idealized (to me) version.  I see that I identified the purpose of my transcription as “part-writing analysis,” which itself betrays my own predilection to understand this music as some how reflecting a classical (even Handelian) choral tradition rather than accepting it on its own terms.  I assumed that the four parts I heard were S/A/T/B, but as I listen to it now, I think it is a three-way division of the women, with the men all on the lowest part.  But I’m not sure.  Every now and then I mark extra notes with “some voices,” as if suggesting “well, they’re singing those pitches, but they’re not supposed to be.”

Googling now, I find a performance from 2016 of the same choral song.  In some respects, I’m amazed how consistently it preserves what I heard in the 1990 recording, given that it has not been transmitted through notation (that device that supposedly “fixes” music as it should be).  But as I listen, I delight also in “new” notes that have crept in, particularly in the men's part.  (Are they new or just absent in the 1990 rendition that was caught on tape?)  Different as well are the abrupt modulations upward in the extended coda—and very effective they are.  A new performance reveals that there is much more to the piece than I had heard in my attempt to transcribe it 20+ years ago, but similarly my transcription tells me more about myself than it does about the recording I was trying to reify on paper.  That ain’t the way I heer’d it.

 

Other examples that come to mind of my misplaced mental creases are metrical:  mishearing the meter.  An example I have not been able to overcome is the opening of the third movement of Gustav Holst's Second Suite for Military Band.  I first heard these chords as on the beat, and I cannot break free of that until b. 7 (after the melody enters at the end of b. 6).  Maybe Holst was trying to play with the ears of his audience by articulating a sort of “shadow metre.”  Maybe, that is, I am supposed to fall for it.  But even when I know what it is supposed to be, my ears will not cooperate.  (Here’s audio, if you want it.)

SOURCE:  Gustav Holst, Second Suite for Military Band, mvt. 3 bb. 1-12; condensed score (pub. Boosey & Hawkes, 1922); scan available as IMLSO #166850.  (I have added the red arrow added in b. 6 to highlight the entrance of the melody.)

There are lots of examples of such false meters.  Brahms does it all the time.  The finale of the Saint-Saëns Septet is another great example, as is the first movement of his fifth piano concerto.  And sometimes I enjoy knowing that I’m being deceived.  But—and this is the thing—I wouldn't have known I was being tricked if I had not seen the music.  That ain’t the way I heer’d it.  Yet, how much of the music is really intended to be experienced with score in hand?  Am I cheating by looking behind the curtain to see the notation?

Every now and then I see something that ain’t the way I heer’d itt, but the problem is not my perception but the notation.  Here (on the middle staff) is an example of a triple meter tune—3/2 with an upbeat—shoe-horned into common time (and with the whole-notes in bb. 4 and 11 representing a lingering on the last note of a phrase).  

SOURCE:  cropped scan from The Southern Harmony (1854 edition), from Hymnary.org

You see that sort of thing every now and then in the shaped-note tradition, and occasionally the error has lingered on into modern hymnbooks.  In this instance, many other books of its time got it “right.”  Then again, such an assessment assumes that the composer and I share a like musical understanding, but that the music is just notated wrongly.  But what if it’s not?  How can I know?  I came across a passage in the autobiography of Broadway arranger Robert Russell Bennett (1894–1981) where he notates a bugle call he heard regularly in France during World War I:

SOURCE:  cropped scan of George J. Ferencz, ed, "The Broadway Sound":  The Autobiography and Selected Essays of Robert Russell Bennett (Univ. of Rochester Press, 1999), p. 89

What puzzles me is Bennett’s metrical notation.  What was he hearing?  Was the player thinking in terms of these uneven bars, or was it something more like this?

SOURCE:  my own hypothetical re-barring of Bennett's example

Of course I could not hear the source performance, but it seems to me that Bennett’s transcription is needlessly complicated.  And yet I don’t doubt that he wrote what he heard.  But would I have heard the same thing?  And are we just back to Beauty is in the eye of the beholder?

My concern in this blog—and my professional preoccupation—has been the written music, the notated text.  Now I am confronted by that humbling truism.  Just as we will never be nourished by the delicious odors in the kitchen, still less perusing a cookbook, but only in the eating of meal, the only music that truly matters is not what is written down nor what is realized in performance but rather what is heard.  Of course, it may be heard and appreciated differently by each listener in turn (and it may be heard aright or misheard in various ways).   Neither the composer nor performer(s) may have much power to rectify what any individual listener perceives.  Yet I am inclined to think of that space between the sounding of the notes and the interpretation by the listener as holy ground.