Friday 6 November 2009

neonate intonation

A reporter from the Times rang me up yesterday afternoon to ask for my reaction to the claim made by German biologists that
French newborns tended to cry with a rising melody contour, starting at a low pitch and ending on a high note, whereas German babies preferred a falling melody. …
These patterns are consistent with characteristic differences between the two languages, according the researchers.

This was the first I had heard of this claim, and I had not had an opportunity to read the journal article on which it was based. Not surprisingly, my immediate reaction was to rubbish it. After all, both French and German use both rising and falling pitch contours (as does English).
No one would want to dispute that very young children acquire the appropriate suprasegmentals for their native language, including in particular appropriate rhythm and intonation, well before they acquire segmentals (vowels and consonants). It is entirely plausible that this should be detectable even in newborns, and even that French newborns should vocalize differently from German ones.

The problem arises in the claim that French has rises while German has falls. Both have both.

The report duly appeared in today’s Times. It makes the claim in rather greater detail, including reference to loudness.

While the average volume of crying was the same, the French babies started more quietly and built up to a crescendo, while the German babies did the opposite. …

“When you say the word ‘Papa’ in German, for instance, you stress the first syllable, whereas in French it is the other way round,” explained Professor Angela Friederici, of the Max Planck Institute for Human Cognitive and Brain Sciences, Leipzig. The same pattern is typical for longer phrases, she said.

This appears to be plain wrong. According to Mangold’s Duden Aussprachewörterbuch, German Papa is usually stressed on the second syllable, paˈpaː. (That is if it means ‘father’. If it means ‘pope’, it has initial stress. I don’t think German newborns would be concerned with the Pope.)


  1. First of all, when I read this this morning (with an ill and indifferently crying baby on my arm no less), I thought I should send you the link.

    But ad rem, this sounds like far too much interpretation of what is probably a coincidence.

    Nevertheless, G Papá is about like E papá, maybe less old-fashioned and more genteel, while G Pápa is neutral and rather wide-spread, like póppa in America.

    Anyway, every dialect's accent in French and German has its own intonation. Perhaps more importantly, this is true for regional non-dialectal standards at least in German, isn't it?

  2. Agreed, G paˈpaː has a very prewar feeling to it. No commoner non-Austrian infant under 60 would pronounce it like that.

  3. Read the article, interesting though probably difficult to determine. Possibly an indication of prenatal prosodic processing, but JW is right. Both French and German have both types of prosody, to say one is preeminent or predominate in either language is false.

    Great blog!

  4. Yeah, pa'pa definitely have that 'posh' feel to it. I suspect I know it from mocking portrayals of the Danish royal family. But it's been too long since I watched any satire.

  5. See now .

  6. Having lived in Western Austria, Germany and Eastern Austria for all my life (in very distinct and different dialect areas), I can honestly say I have never heard Papa being stressed on the final syllable. Neither in conversation nor on television/radio. I agree with earlier posts that an initial stress certainly appears old fashioned (I could dearly imagine this being featured in a 19th century costume drama in Vienna) but I highly doubt this being a preferred variant of German speakers.

  7. The 4th edition of the DUDEN Aussprachewörterbuch of 2000 has both versions /pa'pa:/ and /'papa/. My observation (and gut feeling) is that the latter version is the predominant one, at least in modern colloquial German speech.

  8. Someone needs to tell Mangold, then. In the 6th edition:
    Papa (Vater) pa'pa:, auch: 'papa

  9. David Marjanović6 November 2009 at 23:12

    Agreed, G paˈpaː has a very prewar feeling to it. No commoner non-Austrian infant under 60 would pronounce it like that.

    Even in Austria, it occurs exclusively in films about the upper class during monarchy times.

    The Ausspracheduden is full of nonsense as far as I can tell.

    Anyway, every dialect's accent in French and German has its own intonation.

    Yes, except there's not much left of the French dialects, except that the Parisians speak faster than anyone between there and China.

    Perhaps more importantly, this is true for regional non-dialectal standards at least in German, isn't it?

    If Switzerland is ignored, this is a rather small effect, but it probably exists. One reason I despise the Ausspracheduden so much is that it presents just one standard accent. The variety of standard accents is easily as large in German as in worldwide English.

  10. David Marjanović6 November 2009 at 23:20

    Oh, and, the German word for "pope" is Papst /paːpst/*. Papa is absolutely never used for this meaning.

    * One of the few exceptions to the rule that requires short vowels in front of consonant clusters.

  11. /habe:mus pa:pam/ and we have the German word "Papamobil" (= Pope's papamobile), which I pronounce /pa:pamobi:l/

  12. Look out for Horizon on BBC Two, Tuesday.

    All about language acquisition, featuring Noam Chomsky.

  13. David Marjanović8 November 2009 at 17:34

    which I pronounce /pa:pamobi:l/

    Interesting. Short first vowel for me.

  14. This comment has been removed by the author.

  15. @David Marjanović:
    My long a-vowel is probably due to the fact that I'm 'spoilt' by my training in Latin. Using a short a-vowel makes the word sound funny to me because it automatically reminds me of the German equivalent to Daddy.

  16. It's just struck me that this "research" is reminiscent of the sort of thing that is inspired by "Nihonjinron", the idea that the Japanese are a race apart, uniquely attuned to Yamatodamashii and all things spiritual. I am thinking of one "result" to the effect that the Japanese favour the right hemisphere of the brain and other races the left. So to master the language you have to use the right brain. No doubt this is the reason for what I reported in connexion with your question about Scouse, Shige:

  17. Something related:

  18. Fascinating, but Baby B doesn't give us much to go on! So the apparent "Daddy" at the end of Baby A's performance might prompt a process of elimination. But my disinclination to settle for that was indeed prompted by cadence. In which I was vindicated, but no wonder they didn't get very conclusive results from speakers of the languages who hadn't been screened for explicit linguistic training!

    I don't think any of this can have been new to most of us, but it doesn't seem like a very well-designed bit of research.


Note: only a member of this blog may post a comment.