Are there language-independent aspects of the expression of emotion by intonation? More specifically, are there established relations between the expression of emotion by linguistic intonation and by musical means (i.e. musical intervals/phrases/melodies)? Quick summaries of what is known and/or pointers to the literature would be appreciated.

  • 2
    I initially answered this: > The only thing I know is that rising tone for questions is universal, even in tonal languages. Then, seeking a yummy link to back it up I learned even this not universal after all! > Some languages, like Chickasaw and Kalaallisut, have the opposite pattern from English: rising for statements and falling with questions. Sep 17, 2011 at 8:07
  • 2
    @hippietrail: also there is the phenomenon of HRT or uptalk in some speakers of English, where factual statements are given a rising intonation towards the end, sounding like a question.
    – Mitch
    Oct 5, 2011 at 14:54
  • @Mitch: Yes I'm from Australia where uptalk seems to have spread from. Oct 5, 2011 at 17:51

2 Answers 2


I'm not aware of any claims about universals when it comes to intonation and emotion. There are many studies (production and perception) out there that have tried to isolate the acoustic correlates of different emotions in speech in a single language. An example of one such perceptual study is "Verification of Acoustical Correlates of Emotional Speech using Formant-Synthesis" (Burkhardt and Sendlmeier 2000) for German--the researchers controlled various parameters like mean pitch, pitch range, vowel precision, and phonation type in synthesized speech and asked listeners to categorize the stimuli into a finite set of predefined emotions. As you might expect there were no absolute correlations--just tendencies (for example, a lowered mean pitch and a narrow pitch range as well as breathy or creaky voice was frequently judged to signal boredom).

I also read about one study from Emotion, "The minor third communicates sadness in speech, mirroring its use in music." (Curtis and Bharucha 2010) that recorded actors saying lines with various emotional affects and claimed that the descending minor third was a reliable cue to sadness. You can read about their study on the Scientific American website as well as hear audio samples from the study. I am dubious of their methodology and their interpretations of the results, however. Aside from their questionable methods of assessing the melodic intervals of the tokens, when I downloaded their samples and manipulated their pitch to expand and contract the pitch ranges of the utterances, my perception of the emotion being conveyed did not change.

  • Thanks! It sounds like this area could use some more data collection.
    – Azo
    Oct 5, 2011 at 18:23
  • Probably. But I would be skeptical of any study that attempts to tie any melodic interval to a specific emotion, since other factors such as focus and sentence type (e.g. declarative statement vs. echo question) are already known to affect pitch range and melodic contours. So, if from one utterance to the next the emotion of the speaker stays constant but one of those other factors changes, the melodic intervals in the utterance would change! Oct 5, 2011 at 21:41
  • I agree, one should carefully control for such factors. By the way, I just realized that the website that inspired me to ask this question (months after seeing it) is linked from the Scientific American page you linked to, suggesting there really aren't a lot of results along these lines.
    – Azo
    Oct 5, 2011 at 23:10
  • More broadly, I am always suspicious of any claims about emotion. While internal states can/should obviously be part of scientific discourse, who's to say what the difference is between sadness and melancholy, between rage and anger? Is one a kind of the other? Is there a universal "taxonomy" of emotion? I am not asking rhetorically! I just haven't seen any convincing arguments for emotion and so can't help but wonder if they are truly objects of scientific inquiry.
    – Teusz
    Jul 31, 2014 at 9:12

it is not an easy thing to state something is universal. However, there are trends of languages that have a phenomenon like 'final rising' to denote "Polar Questions". There are about 6000 languages around the world. We haven't gone deep on each and every language (i.e. we haven't intensively studied each portion of each component of the linguistic components of these languages.) One thing i am sure of is that, according to the theory of Iconicity "introduced by (Bolinger 1972)" -check the reference- that rising motion of anything denotes continuity, while falling movement of anything denotes finality. This is mention***ed in many papers and articles like Gussenhoven's*** Phonology of Intonation (Free online) or his book: Phonology of tone and Intonation, also a book like Suprasegmentals by Ilse Lehiste talks about prosodic features in general and describes Bolinger's Theory for unfinished business and the finality.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.