Janáček's "nápěvky mluvy"

« previous post | next post »

Jonathan Secora Pearl, "Eavesdropping with a Master: Leoš Janáček and the Music of Speech", Empirical Musicology Review 2006:

The composer Leoš Janáček (1854-1928) has been noted for his interest in speech melodies. Little discussion has focused however on the field methods that he used in gathering them, nor on the products themselves. Janáček spent more than three decades, transcribing thousands of what he termed nápěvky mluvy [tunelets of speech] in standard musical notation. The record that remains of these efforts is impressive both for its volume and its quality, as well as for its potential to reveal aspects of the perceptual overlap between music and language.

Here's an edited and typeset example of one of his "tunelets": According to Pearl, this one

is dated 18 November 1897, presenting us the scene of an old woman at the butcher’s. Já ty paznehty nemám ráda [I don’t like those hooves] she says. Janáček provides the following background: Žena stará – nechtěla asi koupit si “nožičky” u řezniku – mluví k druhé ženě – zvolna – klidně [An old woman – it seems she didn’t want to purchase “little feet” at the butcher’s – she speaks to a second woman – slowly – calmly].

Pearl provides this audio file of (a piano version of) the tune:

Audio clip: Adobe Flash Player (version 9 or above) is required to play this audio clip. Download the latest version here. You also need to have JavaScript enabled in your browser.

Apparently Janáček heard the intonation of Czech phrases as a sequence of musical notes, or could persuade himself that he did so. I've never shared that impression — my experience is much more like that of Joshua Steele, who wrote (Prosodia Rationalis: Or, An Essay towards Establishing the Melody and Measure of Speech, to be Expressed and Perpetuated by Peculiar Symbols, 1775):

[T]he melody of speech moves rapidly up or down by slides, wherein no graduated distinction of tones or semitones can be measured by the ear; nor does the voice (in our language) ever dwell distinctly, for any perceptible space of time, on any certain level or uniform tone, except the last tone on which the speaker ends or makes a pause. For proof of which definition we refer to experiment, as hereafter directed.

Whilst almost every one perceives and admits singing to be performed by the ascent and descent of the voice through a variety of notes, as palpably and formally different from each other as the steps of a ladder; it seems, at first sight, somewhat extraordinary, that even men of science should not perceive the slides of the voice, upwards and downwards, in common speech. […]

Steele was responding specifically to the theory put forward in Of The Origin and Progress of Language by James Burnett, Lord Monboddo (1774):

We have accents in English, and syllabic accents too ; but there is no change of the tone in them ; the voice is only raised more, so as to be louder upon one syllable than another. — That there is no other difference is a matter of fact, that must be determined by musicians. Now I appeal to them, whether they can perceive any difference of tone betwixt the accented and unaccented syllable of any word? And if there be none, then is the music of our language, in this respect, nothing better than the music of a drum, in which we perceive no difference except that of louder or softer.   [as quoted by Steele — the original seems to be slightly different.]

It's certainly true that most people are surprisingly bad at characterizing the "tunes" of speech as a time-function of pitch, even in such simple matters as whether a syllable is rising or falling, though Burnett seems to have been unusually imperceptive.

Steele trained himself to transcribe such pitch contours precisely by removing the frets from his viol, imitating a remembered speech-melody by slides of his finger on the string, and then notating the movements in terms of a modified form of traditional musical notation, in which

instead of using round or square heads for the notes to be marked on this scale (as in the ordinary music) let us substitute sloping or curving lines, such as expression may require; […] which lines, when drawn one the foregoing scale, will easily shew through how many quarter tones the voice is to slide; and these I call the accents or notes of melody.

Here's an example of what the results look like: Thanks to Google Books, you can now read Steele's 1775 work from beginning to end, an experience that I heartily recommend. And while there are probably benefits to doing your intonational analysis on the viol, it's easier these days to use free software like Praat or Wavesurfer.

So why did Janáček perceive a "graduated distinction of tones or semitones" where Steele (and I, and modern pitch-tracking software) hear only "slides"? I don't think that Czech was sung back in 1900, "through a variety of notes, as palpably and formally different from each other as the steps of a ladder" — it's certainly now no less slide-y than English is.

Perhaps Janáček had a form of synesthesia within a single sense.  Or maybe he just forced himself to answer the question, "If I had to notate this as a song, what would be the best way to do it?"

Some LL posts on related topics:

"Poem in the key of what", 10/9/2006
"More on pitch and time intervals in speech", 10/15/2006
"Puzzle of the day: The constitution in B flat?", 10/20/2007


  1. Espen Sommer Eide said,

    September 9, 2011 @ 10:04 am

    Thanks for an interesting read as always. I have as an artist recently been working with a similar but more technological approach to extract the sound of dead and dying languages. Well maybe not so much extract, as to highlight the pure sound characteristics through electronic manipulation. You can listen to some of the results here:

  2. Charles Wells said,

    September 9, 2011 @ 10:08 am

    We lived in northern Ohio for forty years, and have lived in Minneapolis for four years. My impressions (I am only an amateur linguist) of northern Ohio speech is that in many cases it almost fits Burnett's description. It is exceedingly monotone. Minnesota speech has more tune. Native Californians have a lot more tune. Even Richard Nixon did.

  3. bfwebster said,

    September 9, 2011 @ 10:48 am

    Started reading this and suddenly flashed on Professor Higgins trying to teach Eliza to speak 'proper English' in "My Fair Lady" — IIRC, he walks over to a piano and plucks out notes to emphasize intonation and rhythm ("How KIND of you to let me come.").

  4. Joseph said,

    September 9, 2011 @ 11:10 am

    If I'm listening to the same recorded speech, I'll begin to perceive more clearly the 'notes' of speech. As a form of a counterexample, check out the 'talking' piano here: http://www.youtube.com/watch?v=muCPjK4nGY4

  5. a George said,

    September 9, 2011 @ 12:29 pm

    one of the early transcribers was the Norwegian Johan Storm who brings a series of examples in: Johan Storm: "Englische Philologie" (Zweite Auflage), O.R. Reisland, Leipzig 1892, pp. 205-208.

  6. Theodore said,

    September 9, 2011 @ 12:39 pm

    Having Janáček's melody replayed on a synthetic piano seems like an effective satire of the concept of notating speech.

    I would bet on the suggested explanation, "If I had to notate this as a song, what would be the best way to do it?" Especially since Janáček did collect folk melodies in the field. I find the voices on field recordings of untrained singers can be hard to define in terms of "graduated distinction of tones or semitones" yet standard musical notation has been applied to those kinds of performances as well.

  7. Brian said,

    September 9, 2011 @ 1:40 pm

    Nor was Janáček all that unique in this endeavor. Off the top of my head, Steve Reich ("Different Trains") and Harry Partch ("Bitter Music", and arguably much of "The Wayward") both did this sort of thing as well.

  8. Tom Recht said,

    September 9, 2011 @ 2:14 pm

    @Joseph "If I'm listening to the same recorded speech, I'll begin to perceive more clearly the 'notes' of speech."

    There was a Radio Lab segment on this phenomenon, interviewing Diana Deutsch of UCSD: http://www.radiolab.org/2007/sep/24/behaves-so-strangely/

    Do you find that this can happen with any phrase, or just occasional ones that are more music-like?

  9. Xmun said,

    September 9, 2011 @ 3:07 pm

    I once heard a concert in which a singer had difficulty singing the "glides" (as she called them) in a song by the New Zealand composer David Farquhar. "Why was the singer so nervous?" I heard someone say as we all walked out. "She had no-o-o-o-o reason to be." There was a splendid descending glide on that "no". The singer overheard that remark and looked decidedly peeved.

  10. Adam Gehr said,

    September 9, 2011 @ 3:14 pm

    Might he have been thinking of Moravian? I've heard native Czech speakers talking about Moravians as singing their speech.

  11. Luke Dahn said,

    September 9, 2011 @ 3:16 pm

    As a composer and musician, two things came to mind when I read this post. The first was the American composer Steve Reich's technique of writing melodies based on speech patterns, though he often chooses for his material exclamations that are more tuneful that regular speech, such as recorded train stop calls:

    The second thing I thought of was Arnold Schoenberg's attempt to more closely approximate speech in music by using "sprechstimme," a technique in which the melodic line is half-sung and half-spoken. Ironically, the effect is even more other-worldly than one might expect. Or perhaps this effect is the result of his harmonic language?

  12. Barbara Partee said,

    September 9, 2011 @ 7:39 pm

    Neither Mark himself nor any of the commentators have mentioned Mark's early memorable work on the tunes employed in certain conventional patterns like calling your child in in the evening — "John–ny!" — that was what, a descending third? That was the first I had ever heard about connections between 'tuneful speech' and common harmonic intervals in music. I don't remember any of the details, but I remember being intrigued and impressed by Mark's work!

  13. Eric said,

    September 9, 2011 @ 8:31 pm

    I wonder if anyone has spent much time studying the "tunefulness" of rap in hip hop? It's occurred to me before when listening to say, Tribe Called Quest" that there's definitely a strong melodic element to it, and it's not just "talking."

  14. Dan Lufkin said,

    September 9, 2011 @ 9:12 pm

    And then there's Swedish.

  15. Matt McIrvin said,

    September 9, 2011 @ 10:00 pm

    Leonard Bernstein wrote something about the descending minor third in the "nyah-nyah" taunt. But that's more sung than spoken.

  16. a said,

    September 10, 2011 @ 12:13 am

    Remember the post about "It's Gonna Rain" ? That was a song.

  17. James Martin said,

    September 10, 2011 @ 7:20 am

    I think it varies from speaker to speaker. This video harmonises a section of Sarah Palin's interview with Katie Couric; many of the passages are genuinely melodic: http://www.youtube.com/watch?v=9nlwwFZdXck

  18. McLemore said,

    September 10, 2011 @ 8:38 am

    @James, the overlaid piano in that video seems mostly to be getting the rhythm right… Palin's speech isn't any more tuneful than anyone else's.
    @Eric, I've studied the intonation in rap enough to give class lectures on it and would venture to say it's more like chanting than singing — e.g. like the childhood chants Mark used to formulate metrical stress — except of course much more complex… an absolutely fascinating, artful realm of pitch-rhythm play…

  19. J. B. Arnes said,

    September 10, 2011 @ 9:02 am

    There's a classic interview out there wherein Frank Zappa mentions how his guitar solos are "speech-influenced, rhythmically." Just an FYI for anyone who cares.

  20. Rod Johnson said,

    September 10, 2011 @ 11:46 am

    Zappa was a fan of Eric Dolphy, who famously tried to imitate the sound of speech in his solos. There are recordings of Dolphy and John Coltrane "talking" to each other with their saxes, for anyone who wants to assess their success.

  21. Mike Maxwell said,

    September 10, 2011 @ 2:23 pm

    @bfwebster: or a xylophone, see 1:25 here:

  22. Glenn Bingham said,

    September 10, 2011 @ 2:54 pm

    Now that we are on JazzSpeechLog, there is also Maynard Ferguson "preaching" through his trumpet. Gospel John:


  23. Wu Yong said,

    September 11, 2011 @ 6:51 am

    I've seen Chinese grammar books published in the west which have the tones of Putonghua and Cantonese demonstrated with musical notation…

  24. Matt said,

    September 11, 2011 @ 11:54 pm

    There's quite a bit in Berliner's "Thinking in Jazz" about soloists consciously approaching their solo as speech, or thinking up words for other people's solos that they memorize ("Remember 52nd street/ the sounds were unique/ yes/ the music had a message/ you see"). Great book.

  25. WillSteed said,

    September 12, 2011 @ 12:37 am

    Chao Yuenren used musical notation for his landmark Xiandai Wuyu Yanjiu (Studies of the Modern Wu Dialects). He used a slide whistle to imitate the speaker's tones and notated them as an key and octave followed by the notes within that scale for the tones.

    They've been used as acoustic data to compare Chao's record with modern recordings of the same varieties.

  26. phspaelti said,

    September 12, 2011 @ 3:04 am

    The alemannic dialects of the mountain area of Switzerland are often considered to be "singing" dialects. In her 1916 dissertation C. Streiff tried describing the tunes using musical notation as well. A copy of this document can be seen here:
    The description of the "musical accent" is in §22 on p. 17 ff.

    Streiff was also a pioneer in attempts at making recordings of samples for posterity. Samples of her speech were preserved on wax phonogramms at the University of Zürich:

  27. hans said,

    September 12, 2011 @ 3:24 am

    The translation of Janáček's comment is wrong (or there is a misprint in the original). The comment doesn't say "mluví k drUhé ženě", but "mluví k drAhé ženě", which means: talks to a dear woman, to someone who's close to the speaker.

    Janáček was more Moravian than Czech, and indeed Czech listeners (such as me) sometimes percieve the Moravian dialects as more melodic than Czech.

    Another Czech composer named Alois Hába, who composed quarter-tone music, captures the melody of speech to some extent in his opera "Matka" (Mother).

  28. Alex Temple said,

    September 12, 2011 @ 9:35 am

    One of my favorite composers to use speech samples for their melodic content is Jacob Ter Veldhuis, a.k.a. JacobTV. Here's an excerpt of his piece "The Body of Your Dreams," based on samples from weight-loss infomercials:

    Sometimes the piano part matches the speech samples only in rhythm and contour, but there are spots where it matches them in pitch as well ("by pushing this button"). There's a trick, though: the pitches in speech are usually vague enough that you can manipulate how a listener perceives them by giving them a different accompaniment. I spent some time studying another JacobTV piece, "Lipstick," and I found that he could make the exact same sample of a voice saying "jumping" sound like E-B or like E-Bb depending on how he "doubled" it in the flute.

    I wrote a piece based on speech samples about five years ago, and when I was going through the source recordings (people talking about their dreams), I found that certain phrases jumped out as clearly more "melodic" than their surroundings. I didn't use any instrumental accompaniment, or any sounds at all that aren't from the source recordings, so I think you can hear (especially after about 45 seconds in) that clear melodies do sometimes exist in speech:

  29. Joseph said,

    September 12, 2011 @ 10:35 am

    @Tom Recht: I haven't rigorously tested it, but generally speaking I'd say yes. It should be noted that I am a native speaker of a tonal language who's also had formal musical training, so my perception is probably a little biased. But the other thing that I find is that the notes that I'm hearing in speech don't necessarily always map neatly into the deodecaphonic scale. For example, although I could confidently map the initial note for a given Mandarin 4th-tone, I wouldn't be sure how far down the scale I should stop (whether it'd be a perfect fifth, or a full octave, for example).

  30. Ray Dillinger said,

    September 12, 2011 @ 10:43 am

    I think that the perception of tonal variations in inflected phrases are probably how tonal languages sometimes arise spontaneously from non-tonal precursor languages.

    ObDisclaimer: I am a computer linguist dealing mainly with English, and my exposure to other languages is limited to (non-tonal) romance languages. My training and knowledge of the processes by which languages differentiate themselves from one another, is limited to having read a few books that don't treat the origins of tonal languages explicitly. So this theory may or may not be mainstream, and may or may not have already been advanced or shown or even tested by someone else, and I wouldn't know.

    My theory is that phrases which have a characteristic tonal intonation colloquially, become lexicalized as words the same way English "God be with ye" became "Goodbye" — but the lexicalized item, for some reason (possibly to disambiguate with an existing similar word) preserves the characteristic intonation. When several such items accumulate, you have a set of words distinguished primarily by tone, and awareness of tone becomes part of how the language is understood. Add a few centuries and stir, and you may wind up with a tonal language.

    I've noticed that tonal languages tend to have much simpler syllables otherwise than non-tonal languages. I speculate that a precursor non-tonal language with a simpler syllable structure may much more frequently than otherwise give rise to a situation in which a lexicalized item needs to have some characteristic intonation in order to keep it distinct from an existing word. Thus tonal languages may be one of the consequences of otherwise simpler syllable structure, rather than just a correlated phenomenon.

    But this is a chicken-and-egg speculation. It seems just as likely that users of tonal languages can drop syllabic complexity as an unnecessary ornament given the ability to distinguish words by tone, in which case syllabic simplicity may be a consequence of tonalism. Someone has probably done the hardcore research needed to find out which scenario is true (if either is true), but it isn't me.


  31. Rod Johnson said,

    September 14, 2011 @ 1:04 pm

    Ken Pike used to carry a slide whistle around too, to try to capture intonation contours. Supposedly it was quite irritating at times.

  32. Don’t speak to me in that tone and/or melody. | Josh McNeill said,

    September 16, 2011 @ 4:46 pm

    […] this, despite him being one of my favorite composers. I also didn't expect to come across this information on Language Log, a linguistics […]

RSS feed for comments on this post