Navigation – Plan du site

AccueilNuméros39Folksong is not Poetry: Advocatin...

Folksong is not Poetry: Advocating for a Non-Modular Theory of Text-Setting in Folksong

Une théorie non-modulaire du text-setting dans la chanson traditionnelle
Rosalía Rodríguez-Vázquez
p. 19/34


La théorie le plus largement acceptée du text-setting soutient une vue de la chanson comme un composite qui combine un objet linguistique—le texte—et un objet musical—la mélodie—avec les structures métriques indépendantes régulées par différents ensembles des contraintes. Cet article avance des preuves contre une telle conception modulaire. J’applique quelques contraintes métriques de la Théorie de l’Optimalité à une chanson traditionnelle (folksong) du vingtième siècle pour démontrer que la prosodie du discours et la prosodie du vers sont secondaires par rapport au rythme de la chanson, qui est déterminée non seulement par le battu musical, mais aussi par la valeur des notes. Je soutiens qu’une folksong est une unité où la prosodie linguistique et le rythme musical réagissent réciproquement pour remplir les conditions rythmiques de la métrique de la chanson dans cette langue.

Haut de page

Texte intégral

1In recent years, several papers dealing with the relationship between linguistic prosody, verse prosody and musical rhythm from an Optimality Theory (OT) perspective have been written by linguists such as Hayes and MacEachern, Kiparsky, and Hayes, among others. These works seek to find out which constraints are at work in the process of text-setting in different languages, though most of them focus on song or chanted verse in English. Dell and Halle (63) define “song” as “a composite which combines two objects each with its own structure, a linguistic object—text—and a musical object—tune”, a view which implies that rhythmic and melodic settings of a given text emerge from two different sets of constraints, and text and tune are, in turn, two different objects regulated by separate constraint systems. Under such modular view, the relationship between text and tune would be mediated by text-setting, a concept explored by Halle and Lerdahl, which involves arranging lines against a metrical pattern or grid.

Figure 1. — Text-setting.

Figure 1. — Text-setting.

2Kiparsky (7) takes the modularity idea a step further by arguing that “the composer and performer of a song constructs a match between three tiers of rhythmic structure: linguistic prominence, poetic metre, and musical rhythm”. Those three tiers are organised according to the same kind of principles, as hierarchies of alternating prominence which are visually represented by trees or grids. Kiparsky (7) explains the concept of tier autonomy by pointing out that “a text has an intrinsic prosodic form independently of how it is versified, a stanza has an intrinsic metrical form independently of how it is set to music, and a tune has an intrinsic musical rhythm independently of the words that may be sung to it”.

3In this paper, I argue for a non-modular view of text-setting in folksong, whereby the prosody of speech and the prosody of verse are secondary to the rhythm of song, which is determined not only by the beat and where it falls, but also, as will be demonstrated, by note value and pitch height. I suggest that a folksong is a unit where linguistic prosody and musical rhythm interact so as to fulfil the rhythmic requirements of the metrics of song in that language. The paper is divided into three parts. In the first part, I introduce the main methodological principles applied by OT to the study of text-setting in folksong, concentrating on two specific constraints—MATCHSTRESS and PARALLELISM. In the second part, I illustrate how the afore-mentioned constraints interact and I expand the restrictions posed by such constraints, arguing for the need to revise their ranking for folksong. In the third part, I present my conclusions.

  • 1 “Universal Grammar” refers to the system of categories, operations, and principles shared by all hu (...)

4OT argues that Universal Grammar comprises a set of constraints, which are violable and hierarchically organised on a language-specific basis.1 The input candidate competes with other logically possible forms. The grammatical form is the one that best satisfies the given hierarchy of constraints.

5According to Hayes and MacEachern (1996), Kiparsky (2006), Hayes (2009) and Dell and Halle (2009), text-setting is regulated by two main groups of constraints, namely metrical and grouping constraints, which are present at both the prosodic level in language and the rhythmic level in song. For the sake of clarity, this paper will be dealing with one metrical constraint, MATCHSTRESS, and one grouping constraint, PARALLELISM, which are defined below.

6(1) MATCHSTRESS (Dell and Halle 69)

  1. All stressed syllables fall on strong beats.

    • 2 By “strong beat” we mean the most salient unit in a musical bar, which corresponds by default to th (...)

    All syllables that fall on a strong beat must be stressed.2

7In (2), extracted from the English folksong The Penny Wager (Kennedy, 1984), the text is set so that stressed syllables (“trav-”, “world” and “North”) fall on strong beats, while unstressed syllables (“I’ve”, “-elled”, “this”, “from”, “the” and “-ter”) correspond to weak beats.


9If we were to fully support the view that MATCHSTRESS is highly ranked in English folksong, the first and third syllables of the word “Counterie” would pose a problem, as the actual setting of the text assigns a strong beat to the third syllable, which is prosodically unstressed in spoken English (/’kntri/).

  • 3 In Fabb and Halle’s view, a syllable is projected when it is assigned a metrical position in the gr (...)

10Text-setting has been traditionally illustrated by means of a grid which provides us with an apparently quick, synthetic view of the degree of metricality of a specific line. For Hayes and Kaun, “metricality” refers to the extent to which a specific deliverance of a line corresponds to a pre-designed grid, a concept thoroughly explored by Fabb and Halle (26), for whom rhythm has to do with the arrangement of syllables as stressed or unstressed, while metre has to do with the grid, where certain syllables might not be projected (observe [3]), where Ø symbolises the non-projection of a syllable), while they are always pronounced in the reading or performance of a poem.3


12One can easily observe that the metrical grid does not work for song as it does for poetry, as in music, unlike in poetry, pauses are a part of the metrical grid. As Fabb and Halle acknowledge (36), albeit from a different theoretical framework, what is projected in music is not syllables, but “time intervals, determined by the value of the shortest note”. Crucial to their view is that “when a metrical text is set to music, a distinct musical grid is created which usually shows no particular resemblance to the grid of the text, as the two grids are generated by separate grouping rules. There is thus no necessary connection between the metre of a line of poetry and the way it is set to music” (Fabb and Halle 37).

13This links in nicely with Kiparsky’s view of text and tune as independent objects (14), as in vocal music “the same words are commonly sung to different musical measures, yet maintain certain invariant constraints on stanza form”. However, the possibility of assigment of a text to a number of different melodies in folksong does not mean that none of the texts has been created for a specific melody. We can rightly presuppose that each text was originally associated with its tune although, given the oral character of folk song, as time passed texts and melodies started to be dissociated and re-associated with new tunes and texts. Sharp explains this process in terms of variation stages (25-26), as “a frequent source of variation arises when the singer, having partially forgotten his words, has substituted corrupt and unmetrical lines. The attempt to adapt the tune to these irregularities will often lead to the inventions, unconscious of course, of interesting melodic changes”.

14Let us go back to the grid and observe the first stanza of The penny wager in (4), where the horizontal crosses symbolise the isochronic occurrence of time intervals. A time interval with more vertical crosses is more salient than one with fewer crosses.


16The musical grid is intended to reflect the rhythmic information provided by the music notation, particularly data related to beat relative strength and duration, and presence of pauses and silences. We observe a connection between MATCHSTRESS and the information provided by the grid with regard to the correspondence between syllables and positions. The grid also sets up a schema which should in principle remain invariable across the stanzas that make up a song. In (5), the pattern of metrical positions which get filled with syllables in the first line of the first stanza (S.1) is reproduced in the first line of subsequent stanzas, i.e. positional PARALLELISM (6) is observed.


Table 1. — The Penny Wager (first stanza and refrain)

Table 1. — The Penny Wager (first stanza and refrain)

18(6) PARALLELISM (positional) (Dell and Halle 70)

Two alignments are positionally parallel if the distributions of their syllables along the grid are identical.

19When applied to the study of text-setting, neither the stated OT constraints nor the grid take into consideration rhythmic elements beyond metrical position itself. This could lead to the mistaken conclusion that the melodic component of a song has no rhythmic function, thus giving rise to an incomplete theory of text-setting in song, which is equated to text-setting in chanted verse (see Hayes). A song is an integral entity, where the different components (chiefly, metrical position, note value and pitch height) can neither be neglected nor analysed in isolation, as the melodic contour is as decisive a parameter as is metrical position when judging the well-formedness of a setting. In Tovey’s words (212), “even with the full muscular strength of modern musical rhythms, the strong beats of the bar constitute only one form of accent, which may be easily counteracted by the length of a note, by its height, by its harmonic colour, and by incidents in the accompaniment”.

  • 4 Full text of The Penny Wager: I’ve travelled this world from the North Counterie//A-seeking, a-seek (...)

20The Penny Wager (Figure 2) is a strophic song, i.e. it sets different lines of text to the same musical pattern. This version of the song, recorded by Kennedy in 1950, comprises six stanzas, each of which consists of four lines. The refrain comes after the last line of every second stanza.4

Figure 2. — The Penny Wager.

Figure 2. — The Penny Wager.
  • 5 “Scansion” is the process of measuring verse, that is, of marking accented and unaccented syllables (...)

21Table 1 sets out the metrical scansion of the first stanza, where a cross stands for “strong metrical position” and a hyphen stands for “weak metrical position”.5 Positions group into feet. The shaded area (lines 5, 6) corresponds to the refrain.

  • 6 A syllable is said to be “strong” when it is made to correspond to a strong metrical position (a st (...)

22The four lines in stanza 1 comprise four feet starting with a weak syllable and ending with a strong syllable.6 As Attridge observes (83), the structure of four lines of four beats each is “by far the commonest rhythmic structure of popular song [. . .] and [. . .] it occurs in, or underlies, nearly all popular verse in English”. Despite the popular view that four-beat patterns in verse owe their existence to the music with which the verse was originally associated, Attridge follows Burling’s (1425) view that “we have general rules of rhythm, which are neither predominantly musical nor predominantly poetic, but stand equally behind both music and spoken verse”.

23Although the number of metrical accents in this song stays constant throughout the stanza, the fact that the number of syllables varies between lines has musical implications that get reflected in the text-setting of the song, but not in its scansion. If the premise for text-setting is that, as a general rule, one syllable corresponds to one rhythmic note, then a line with more syllables will require more rhythmic notes (i.e. more metrical positions to be filled), which means that the exact tune of the first stanza will not necessarily be preserved throughout the whole composition—PARALLELISM will be violated. It has been stated that PARALLELISM is ranked lower than MATCHSTRESS in English folksong (Dell and Halle 75), an idea that will be qualified later on in this paper.

24The only factor that must remain constant is the occurrence of strong syllables at equal intervals—metrical feet must be isochronous, a fact arguably derived from the speech prosody of English, where stresses fall at perceptually equal intervals (stress-timing). This type of isochrony is automatically achieved in much Western music since there is always a strong accent on the first beat of each bar and a secondary accent on the second or third beat of each bar (depending on the time signature). In this respect, neither the scansion presented in Table 1 nor the grid adds any relevant data to the information provided by the musical notation.

25The key question is to analyse what type of linguistic materials fills the different rhythmic slots. In fact, the difficulty in song arises when strong metrical syllables, that is, syllables that fall on strong metrical positions, must be made to correspond to both strong and weak beats. In this sense, the musical setting of the first stanza of The Penny Wager provides us with instances of unstressed syllables being promoted to strong metrical positions (“-rie” in line 1, “-nie” in line 2, “to” in line 4) as well as stressed syllables being demoted to weak positions (“count-” in line 1 and “com-” in line 2). In OT terms, MATCHSTRESS is violated on several occasions in the song. Mismatches result from misalignments of a syllable to a strong beat if it is unstressed (“to” in line 4) or to a weak beat if it stressed (“count-” in line 1).

26OT does not contemplate those cases where MATCHSTRESS is not determined by misalignments of stress and position, but by misalignments of stress and duration. This is what Rodríguez-Vázquez calls “duration mismatches” (2010a, 25; 2010b, 249), whereby note values are in disagreement with prosodic stress and/or syllable duration. Let us observe (2), where the grammatical word to, realised as /t/ in connected speech, not only falls on a strong beat—stress mismatch—but is made to correspond to a long note value—two quavers or a crotchet. This setting renders the function word to as salient a syllable as the lexical word please in the previous bar, which, in spite of containing a long vowel that never gets reduced in connected speech, is assigned a quaver. In this respect, the mismatch of the grammatical word to is doubly salient. On the other hand, in stanza 5, line 1, the lexical monosyllabic word “caught”, with a long vowel as its nucleus (/kɔ:t/), is made to correspond to a semi-quaver (not provided in the musical notation, as stanza 1, line 1 does not make use of it), which renders it less salient than function words such as “the” and “a” (realised as /ðə/ and /ə/, respectively) in the same line.


28According to this, MATCHTRESS should be implemented bearing in mind mismatches between stress and note length. My first extension of the restrictions established by MATCHSTRESS has been formulated in (8b) below.


29(a1) All stressed syllables fall on strong beats.
(a2) All syllables that fall on a strong beat must be stressed.
(b) All stressed syllables fall on a note-value which minimally corresponds to that indicated by the time signature denominator.

30From a close analysis of the text-setting in the six stanzas, we observe that the melodic contour of the song—that is, the tune itself—anticipates the need of syllable insertion at certain points. This entails the violation of positional PARALLELISM. The grid presented in (5) shows the matching of syllables and beats in stanzas 1, 2, and 5, line 1, where two weak syllables (“from the”, “and a-”, “and I”, respectively) are made to correspond to two semiquavers. On the other hand, line 1 in stanzas 3, 4 and 6 assigns a quaver to syllables “sat”, “asked” and “un-”, respectively, thus causing a violation of positional PARALLELISM.

31A second anticipating mechanism can be seen in the setting of a single syllable to more than one musical note (a melisma), which also makes the distributions of syllables along the grid not identical.


33The violations of PARALLELISM observed above are linked to the fact that different lines show a varying number of syllables across the song—line 1 has 11 syllables in stanzas 1 and 2, 10 in stanzas 3, 4 and 6, 12 in stanza 5. Such violations are thus a by-product of the relative rigidity of the musical grid, which remains stable throughout the whole song.

34The concept “melisma” leads us towards the melodic element that is necessarily present in song—not in chanted verse. This song presents a really simple melodic-harmonic structure, which interacts with its metrical structure. Table 2 sets out the analysis of the first stanza using melodic, harmonic and metrical parameters.

Table 2. — Mapping melody, harmony and metre

Table 2. — Mapping melody, harmony and metre
  • 7 A cadence is a stylised close in music which divides a given composition into periods or brings it (...)

35Roman numbers indicate the harmonic progression, where “N-A” means “Non-Authentic” cadence and “A” stands for “Authentic” cadence.7 Asterisks correspond to those positions where a harmonic movement takes place (on the first beat and sometimes also on the second beat), and a prosodically unstressed syllable falls. This gives rise to a double mismatch, as an unstressed syllable falls both on a strong beat and a harmonically-critical point. As can be observed in Table 3, when the parameters relative note height and relative note duration are added, the most salient mismatch is “to” (dark shading), while “-nie” in “pennie” is a relatively salient one (light shading) and “-rie”/“-nie” in “Counterie” and “companie” stand as minor ones.

Table 3. — Mapping melody, harmony and metre

Table 3. — Mapping melody, harmony and metre

36Below is my final extension of the restrictions made by MATCHSTRESS (see 10b, c).

(10) MATCHSTRESS (final)

37(a1) All stressed syllables fall on strong beats.
(a2) All syllables that fall on a strong beat must be stressed.
(b) All stressed syllables fall on a note-value which minimally corresponds to that indicated by the time signature denominator.
(c) All syllables that fall on a cadence must be stressed.

38In conclusion, there are two types of mismatch if we bear in mind considerations other than purely accentual ones. On the one hand, we encounter a group of polysyllabic nouns with initial stress (“companie”, “pennie”), which are nevertheless set so that the musical beat falls on their second syllable. The mismatched nouns end with a <y>—or its spelling variant <ie>, both of which correspond to the phoneme /i/. This is a peculiar phoneme, known as “neutralised /i/”, which shows two unique characteristics, namely (a) it is neither short nor long and (b) it is tense (see Halle and Keyser 29). In other words, this phoneme cannot be reduced to schwa (/ə/), which might be a reason why it can be made salient when set to music. This phenomenon, referred to as “Country mismatch” (Rodríguez-Vázquez 2010 a, 30; 2010 b, 241), usually takes place at the end of a musical and textual phrase (where cadences occur), which is marked with a long note value. From my point of view, this type of mismatch is idiosyncratic of folksong, to the extent that it should be taken as a style or genre marker. On the other hand, monosyllabic prepositions (“to”, “on”, “for”) are sometimes mismatched, although not exclusively from an accentual/positional point of view—melodic and harmonic issues play an important role, as has been shown—, a fact which parallels the promotion of unstressed syllables in poetry.

39After having analysed The Penny Wager as a song, my conclusions are the opposite to those reached by Hayes, among others, who ranks MATCHSTRESS higher than PARALLELISM. In my view, fidelity to the musical grid in terms of beat strength, note-value and cadential action regulates text-setting in folksong, which implies loosely adhering to PARALLELISM, while MATCHSTRESS bears the consequences of the former, which entails a tension between the musical grid (established by the time signature and the rhythmic and melodic-harmonic schemata set in the first stanza) and its rhythmic and melodic realisation (how the remainder of the text is set).

40When referring to folksong, both the use of the grid and the application of modular constraints might be seen to entail a series of simplistic observations and conclusions related to how stress and beat relate to each other in folksong. As Boone suggests (212), there have always been two major pitfalls in theories about the relationship between music and text: “the first pitfall is overemphasis on tonic accent as the prime agent of rhythmic activity in linguistic rhythm; the second is overemphasis on a simple alignment between tonic accent and musical accents as the hallmark of “good” prosody”.

41The view of the relationship between music and language as a stress game is distorted because it does not really tell us much about what is going on in language, poetry and music and, more importantly, yields a partial account of the infinite ways of interacting between text and music, be it in the enormously rich and varied traditions of folksong, or in the intentionally imbricated realms of art song.

42I hope to have shown that a folksong should not be analysed as poetry (where metricality plays a key role) to which a melody is adapted, but as a separate entity where speech, verse and music come together and respond to folksong constraints.

Haut de page


Attridge, Derek, The Rhythms of English Poetry, London: Longman, 1982.

Boone, Graeme M., Patterns in Play, Lincoln: U of Nebraska P, 1999.

Burling, Robbins, “The Metrics of Children’s Verse: A Cross-linguistic Study”, American Anthropologist 68 (1966): 1418-1441.

Dell, François & John Halle, “Comparing Musical Text-Setting in French and in English Songs”, Towards a Typology of Poetic Forms, ed. Jean-Louis Aroui and Andy Arleo, Amsterdam: Benjamins, 2009, 63-78.

Fabb, Nigel & Morris Halle, Meter in Poetry: A New Theory, Cambridge: CUP, 2008.

Halle, Morris & Samuel J. Keyser, English Stress: Its Form, Its Growth, and Its Role in Verse, New York: Harper and Row, 1971.

Halle, Morris & Fred Lerdahl, “A Generative Text-Setting Model”, Current Musicology, 55 (1993): 3-23.

Hayes, Bruce, “Text-Setting as Constraint Conflict”, Towards a Typology of Poetic Forms, ed. Jean-Louis Aroui and Andy Arleo, Amsterdam: Benjamins, 2009, 43-62.

Hayes, Bruce & Abigail Kaun, “The Role of Phonological Phrasing in Sung and Chanted Verse”, The Linguistic Review 13 (1996): 243-303.

Hayes, Bruce & Margaret MacEachern, “Are There Lines in Folk Poetry?”, UCLA Working Papers in Phonology 1 (1996): 125-142.

Kennedy, Peter, Folk Songs of Britain and Ireland, London: Oak, 1984 [1975].

Kiparsky, Paul, “A Modular Metrics of Folk Verse”, Formal Approaches to Poetry, ed. Elan B. Dresher and Nila Friedberg, Berlin: Mouton de Gruyter, 2006, 7-52.

Rodríguez-Vázquez, Rosalía, “Text-Setting Constraints: a Comparative Perspective”, Australian Journal of Linguistics 30.1 (2010a): 19-34.

Rodríguez-Vázquez, Rosalía, The Rhythm of Speech, Verse and Vocal Music: A New Theory, Bern: Peter Lang (Linguistic Insights Series vol. 110), 2010b.

Sharp, Cecil, English Folk Song, London: Methuen, 1954 [1907].

Tovey, Donald F., “Words and Music: Some Obiter Dicta”, The Main Stream of Music and Other Essays, New York: AMS, 1979 [1949], 202-219.

Haut de page


1 “Universal Grammar” refers to the system of categories, operations, and principles shared by all human languages and considered to be innate.

2 By “strong beat” we mean the most salient unit in a musical bar, which corresponds by default to the first pulse in the bar. There might be more than one strong beat per bar, depending on the time signature.

3 In Fabb and Halle’s view, a syllable is projected when it is assigned a metrical position in the grid. Projection thus belongs in the abstract realm.

4 Full text of The Penny Wager: I’ve travelled this world from the North Counterie//A-seeking, a-seeking for good companie//Good companie I never could find//Nor that which will please me to my mind//Singing: Whackfoldeday//whackfoldeday//and I had in my pocket//just one pennie.//I saddled my horse and away I did ride//Till I came to an alehouse all on the wayside//I boldly got off and sat myself down//And called for a jug of good ale that was brown//Singing. . .//There were two young men sat playing the dice//And I thought they were playing so gay and so nice//As they were a-playing and I looking on//They took me to be a young nobleman’s son//Singing. . .//One of these young men asked me if I’d play//and I immediately asked him what wagers he’d laid//He said: Five guineas and the other: Ten pound//The wager was bet, but no money put down//Singing. . .//I caught hold of the dice and I gave them a fling//And it happened to be my good fortune to win//If they had a’won, and I had ’ave lost//I should have to have sold my little black horse//Singing. . .//I tarried that night until the next day//when I thought it was time to be jaggin’ away//I asked the landlady what I had to pay//Come give me a kiss, and go on your way//Singing: Whackfoldeday//whackfoldeday//and I had in my pocket//just ten pounds free.

5 “Scansion” is the process of measuring verse, that is, of marking accented and unaccented syllables, dividing the lines into feet, identifying the metrical pattern, and noting significant variations from that pattern.

6 A syllable is said to be “strong” when it is made to correspond to a strong metrical position (a strong beat in music).

7 A cadence is a stylised close in music which divides a given composition into periods or brings it to a full conclusion. An authentic cadence is a chord progression where the dominant chord is followed by the tonic chord (V-I, where “V” represents the chord based on the fifth step of the Western scale and “I” represents the chord based on the first step of the scale). All other cadences are considered non-authentic.

Haut de page

Table des illustrations

Titre Figure 1. — Text-setting.
Fichier image/jpeg, 152k
Fichier image/jpeg, 68k
Fichier image/jpeg, 48k
Fichier image/jpeg, 72k
Fichier image/jpeg, 92k
Titre Table 1. — The Penny Wager (first stanza and refrain)
Fichier image/jpeg, 128k
Titre Figure 2. — The Penny Wager.
Fichier image/jpeg, 324k
Fichier image/jpeg, 76k
Fichier image/jpeg, 120k
Titre Table 2. — Mapping melody, harmony and metre
Fichier image/jpeg, 136k
Titre Table 3. — Mapping melody, harmony and metre
Fichier image/jpeg, 99k
Haut de page

Pour citer cet article

Référence papier

Rosalía Rodríguez-Vázquez, « Folksong is not Poetry: Advocating for a Non-Modular Theory of Text-Setting in Folksong »Études britanniques contemporaines, 39 | 2010, 19/34.

Référence électronique

Rosalía Rodríguez-Vázquez, « Folksong is not Poetry: Advocating for a Non-Modular Theory of Text-Setting in Folksong »Études britanniques contemporaines [En ligne], 39 | 2010, mis en ligne le 11 février 2016, consulté le 06 février 2023. URL : ; DOI :

Haut de page


Rosalía Rodríguez-Vázquez

University of Vigo.
Rosalía Rodríguez Vázquez was born in Vigo (Galicia, Spain). She graduated in English Language and Literature (Honours and Prize) and in Music Performance from the University of Vigo. In 2003, she moved to the University of Edinburgh (Scotland, UK), where she finished an MSc by Research in Linguistics and Music (Distinction) funded by the Fundación Caixa Galicia. In 2004, the Arts and Humanities Research Council awarded her a three-year grant to complete a PhD (Phonology and Music) at that same Scottish university, where she also taught English Linguistics for four years. She now holds a permanent lecturing position at the Department of English, French and German (University of Vigo). In 2010-11, she will carry out a twelve-month postdoc at the “Structures formelles du langage” research lab (Paris VIII/CNRS), funded by the Fundación Pedro Barrié de la Maza. Dr Rodríguez-Vázquez’s first monograph, The Rhythm of Speech, Verse and Vocal Music: A New Theory (Bern: Peter Lang), appeared in January 2010.

Haut de page

Droits d’auteur


Creative Commons - Attribution - Pas d'Utilisation Commerciale - Pas de Modification 4.0 International - CC BY-NC-ND 4.0

Haut de page
  • Logo Presses universitaires de la Méditerranée
  • Logo ERIH +
  • DOAJ - Directory of Open Access Journals
  • OpenEdition Journals
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search