Melody | Remarkable modulations in tone that make speech tend toward song, chanting, or litany. At times uneven, repetitive, contrasted, or monotonous, melody is what gives speech its expressiveness.

School grammar books identify and teach three types of intonation corresponding to three modes of expression: declarative, interrogative, and exclamative. The first is associated with assessment or commentary: its melodic line “drops” at the end of a sentence. The second is for questions, the voice rising into a small upward curve at the end of an utterance to mark a suspension waiting to be resolved. The last case, for surprise, is also ascendant but more sharply so, as if suspended at its peak by the event that made it happen.

In reality, though, speech undergoes much more subtle modulations: it is obvious that we do not always speak in the same tone, and that even the most ordinary form of speech modulates its pitch and rhythm. Listening to this excerpt of a piece by René Lussier shows us just that.

As with timbre, certain speakers can be recognized by the melody of their speech; certain situations can also be recognized by the melodies they call for. Certain contexts produce more or less regular or uneven melodies, marking the affective or rhetorical nature of the enunciative situation in question. Finally, speech can also tend toward chanting or litany. In such cases, we say that it sounds like a speaker is singing when they talk, and the melodic aspect of their speech, freed of the necessities of address, will favor expressiveness over expression.

Standardized inflection

Certain forms of speech are built from pre-established tonal canons. Like this French railway announcement, for example, played on train station platforms. The strangeness of these words is due to the fact that they are artificially constructed, made up of pre-recorded words that form a coherent and instantly effective utterance (“travelers are to move to platform 1 without delay”), all while revealing their patched-together nature through gaps in the tonal lines they follow. This paradoxical excerpt is both regular and strangely inorganic.

In many cases, speech is very clearly structured (or implied) by certain melodic standards that have been more or less well assimilated by a speaker. The “proper speech” or “proper reading” these words tend to conform to are based on a certain vision of variety and modulation. This sermon, this response to an interview, this story, all aim to give life to a scene or utterance that will make them more lively and attractive.

These forms of speech can combat the monotony of someone “droning on” by playing with different voices or pitches (for an example of monotonous yet thrilling speech, see Mon père by Charles Pennequin).

In this excerpt from Jean-Marie Straub and Danièle Huillet’s film En Rachachant, conventional speech pitches are outrageously accentuated to produce a strange form of dialogue, trying so hard to sound correct that it feels “fake”—perhaps the very definition of the invented word rachacher which gives the film its title.

Sinusoidal forms

Some of this collection’s excerpts contain melodies remarkable for the way they modulate their pitches into almost sinusoidal forms. This process is characterized by repetitive utterances in which a speaker tries to make each iteration unique through pitch variation, as in shipping forecasts, this tongue twister, or this Gertrude Stein reading. In the same vein, we can cite the very strongly cadenced melody of an auctioneer’s words, compressed and accelerated to an extreme.

A sinusoidal melody also often ends up taking root by custom or wear, as in the routine musicality of this stewardess’s voice, identical in Arabic and in French. Or, sometimes the pattern and repetition of such modulations are structured around a strongly accentuated final note which highlights an utterance’s serial nature: lists, enumeration, arguments. This kind of melody gives speech rhythm, as can be heard in this excerpt from an Abdellatif Kechiche film or in this lecture by Vladimir Jankelevitch.

Uneven melodies

Whereas standardized speech modulates its pitch around a base frequency, some forms of speech are defined by uneven melodies which take form “by hops and by skips,” like the archetypal example of Vladimir Jankelevitch, who can be heard in this wonderful excerpt. These melodic jolts often mark problems in someone’s speech. Physiological problems for this transsexual person going through hormone treatments, or, more largely, ailments produced by an overflow of emotion (or the desire to produce some).

Such is the case with political speeches and their vehemence, pleas, poetic diatribes and, in an even more musical example, gospel preaching (see also C’est bizarre). These “eccentric” melodies are produced by overemphasis, seen as a particular means of dramatizing speech. Indeed, this example shows how a tragedian’s declamation is carried by the rise and fall of her lyrical intonations (see also Entre ici, Jean Moulin).

Speech with uneven melodies belongs to a system that exceeds the typical boundaries of intonation, calling out to the pre-verbal kingdom of children or animals, as with this cowherd from Poitou guiding his animals, or this man imitating birdcalls. This is also the case with babies babbling, a process that is both expressive (showing us the child’s happy or angry affects), intentional (babbling is addressed to someone, it counts on certain effects), and experimental. A child discovers and tunes its vocal apparatus through a series of intuitive scales that allow it to measure tessitura and learn to modulate expressiveness. This is not far from the approach of certain twentieth-century poets or composers like Raoul Hausmann or Luciano Berio, who dismantled and reinvented traditional modes of singing and habitual forms of speech.

Often, the extension of certain vowels pushes these forms of speech close to song, as in this hypnosis tape, this reading of a poem by its author, or this piece by John Cage.


Hence, the collection also provides us with forms of speech whose tonal variations push them close to song. This is the murky domain of sing-speech or speak-song. Opera recitatives, for example, are a means of playing with the tonal organization of speech. Sometimes speech functions like a residual mode of singing (see C’est fini là il est fichu il est mort or La bataille), or sometimes singing’s melodic structure allows for a message to be expressed, as in this recording by an Art Brut artist. Finally, certain unique utterances allow us to hear the way that singing’s aesthetic independence and communicative intent can be ritually bound to one another (see Mantra de la compassion, L’enfant du Coran, (Mh mh mh) yeh yeh eh): in order for speech to be truly effective as such, it must be sung; indeed, it is speech only insofar as it is song.


Trouver la belle

Writing workshop scene from Le Papotin, directed by Alexandre Plank for L’Atelier Fiction, France Culture, 2017.

À peine un filet pluvieux

Evelyne Dheliat, excerpt from the TF1 weather report, 2016.

We still need communism

Sceneable, excerpt of a video posted on Youtube, 2017.

Mais il y a des vilains peut-être

Street Scene, excerpts from the documentary On the Edge of the World, Claude Drexel, Arte, 2014.

Rire c'est bon pour la santé

Johann Schneider Ammann, President of the Helvetic Confederation, excerpt from a video address on the occasion of the Day of the Sick, 2016. 

Perja kupujem

Street buyer, Serbia, 2007. 

Court-circuiter la partitocratie

Patrick Buisson, excerpt from the radio program Répliques, 2016

Viete prečo

Excerpt from a church sermon, slovak television, 2014.

Proroshijile sunt grave

Excerpt of an orthodox monk's YouTube video blog, Romania, 2014.