(Fr. and Sp.; It. romanza; Ger. Romanze).
From the 15th century, romance in Spain and romanza in Italy have nearly always signified a ballad; the narrative romance was, next to the villancico, the most popular song type in Spanish-speaking countries. In France and Germany the term came to indicate an extravagant, sentimental or ‘romantic’ tale in either prose or strophic verse. Since the 18th century vocal and instrumental settings entitled ‘romance’ have continued to express these ‘romantic’ and lyrical qualities (in this sense, the appropriate Spanish word is ‘romanza’).
3. The vocal romance in other countries.
JACK SAGE/SUSANA FRIEDMANN (1), SUSANA FRIEDMANN (2), ROGER HICKMAN (3–4)
As in other countries, Spanish romances (ballads), though probably story-songs at root, often dwell on a single situation taken from a story, the effect being rather more often a heightening of the dramatic tension than of lyricism. There is controversy about their origins. Menéndez Pidal has argued persuasively that romances are unique in European balladry in that they began as fragments of longer epic poems, and, though developing variants by transmission through oral tradition, have been preserved longer and more authentically than have ballads in other countries. Some objections to this theory (see Entwistle, Sage) are these: there is no evidence that Spanish epics preceded the romances they are presumed to have engendered, nor that epic ballads came earlier than other types of ballad; the earliest surviving texts are related not to any uniquely Spanish epic but rather to Carolingian, Arthurian and other cycles well known in European balladry generally, as well as to Spanish history (notably about border incidents involving the Moors or about King Peter the Cruel); and the earliest surviving tunes do not suggest an affinity with the postulated chant or recitation of epics. There is, however, general agreement that romances may be seen as folksongs, not in the sense that they were originally created by or for common people, but rather that they were adopted and continually adapted by common people.
There is strong evidence that, in the 14th century or even the 13th, romances were sung by professional musicians (perhaps jongleurs) for the entertainment of courtly, aristocratic and possibly urban customers. By the middle of the 15th century, though still in demand at court, they seem to have been taken up by amateur singers in other walks of life. By the middle of the 16th century, old ballads had become sufficiently plebeianized for cultured poets and musicians, while often imitating them in new ballads of a more artful kind, sometimes to speak disparagingly of them. On the whole, though, ballads clearly met with more favour in Iberian cultured circles of the 15th and 16th centuries than they did in other countries. Thanks to this patronage, ballads with their music were set down in manuscripts in Spain a century earlier than in most other countries. The poetic form of these early romances was based on an octosyllabic quatrain with assonance (vowel-rhyme) in alternate lines, though some of the more cultured romances from the later 15th century have full rhyme. Menéndez Pidal maintained that the earliest form was the couplet of 16 or even eight syllables; however, none of the earliest extant ballad tunes has a two-phrase musical structure. With some exceptions, such as some versions for voice and vihuela of the famous ballad on the taking of Alhama which interpolate a recurring lament ‘¡Ay de mi Alhama!’ (ex.1), Spanish ballads up to 1550 or so never had a refrain; by the 17th century they often had one, a change brought about not so much by the re-creative process of oral tradition or poetic fashion but rather by musicians in their search for expressiveness (see Montesinos).
The Cancionero Musical de Palacio (E-Mp 1335; ed. in MME, v, x, xiv, 1947–51, 1965; see Cancionero), compiled between 1505 and 1520, contains over 40 romances set for three or four parts. If one can assume that historical ballads were composed in the year of the event they describe, the earliest, Alburquerque, Alburquerque, dates from 1430 (ex.2). Others that can be dated in this way are Pascua d’espíritu santo (1435) and Sobre Baça estaba el Rey (1489). Most of the ballads in this cancionero and others of about 1500 show every sign of being arrangements by court composers (including Juan del Encina, Juan de Anchieta and Francisco de la Torre among others both named and anonymous) of pre-existing tunes for performance by musicians in the court’s employ. Over 50 of these romance tunes have survived in arrangements made during the period from about 1450 to 1550. Francisco de Salinas (De musica libri septem, 1577) implied that a single ballad tune served a variety of texts, a contention that conflicts strangely with the musical evidence, for, though there is certainly a remarkable homogeneity about these 50 tunes, every melody – Salinas’s examples apart – is distinct in its own right. The homogeneity of these tunes (almost certainly the superius in most cases) derives from several factors: they are all remarkably restrained, moving normally by step and jumping rarely by more than a 3rd; they all consist of four balanced phrases (ABCD), exactly matching the octosyllabic quatrain; they hardly ever indulge in any repeated or imitative passages or refrains; each phrase ends with a cadence often marked by a fermata; duple metres are used exclusively whereas other Spanish songs of the period indicate a variety of metres; most syllables are sung to one note, yet there is an equally characteristic fondness for expanded phrases, especially at cadences, of up to 12 notes per syllable. The striking similarity between these melodies and modern chorales (Stevenson SM, p.206) or hymns lends support to the theory that there were common roots between Iberian popular song and church music in medieval times. Almost as striking is the similarity to early English songs of popular type, such as Three Ravens in Ravenscroft’s Melismata (1611) or Western Wind and Cull to me in their duple-time forms of the 15th or 16th centuries (see J. Ritson: Ancient Songs and Ballads, London, 1829). This resemblance suggests that the lost music for English ballads of the 15th and 16th centuries may have shared at least some of the above characteristics with their Iberian counterparts, especially since there are obvious parallels, in the growing addiction to triple time and refrains, between the balladry of both countries in the 17th century.
The evidence about performance of the early Spanish romances indicates that there was no set way of presenting them, but that up to about 1550 they were most often sung in court as three- or four-part homophonic choruses. In most cases one part only is underlaid in the manuscripts, and this has led some students to conclude that such pieces would be performed by a solo voice singing the underlaid part to an instrumental accompaniment. Certainly at times in the 15th century ballads were sung to lute accompaniment apparently played by the singer himself (Menéndez Pidal, 1953, pp.19, 72); it seems reasonable to conclude that the pre-existing tunes were performed also as unaccompanied solos. The qualities most often praised in ballad-singing in this period were sweetness and intense emotion; there is barely a hint of any rough ‘folklike’ singing.
After 1520 the only known sources for old ballad tunes are the vihuela books from Milán’s El maestro (Valencia, 1536) to Esteban Daza’s El Parnaso (Valladolid, 1576), where they are adapted for solo voice and vihuela accompaniment, and two madrigals by Juan Vásquez in his Recopilación de sonetos y villancicos (Seville, 1560; ed. in MME, iv, 1946, nos.28, 44). The indisputable similarities between the melodies of particular ballads in these settings and those of the Cancionero Musical de Palacio and other collections of about 1500 (ex.3) may be attributed to common traditional sources or to the desire of 16th-century composers to re-create tunes taken directly from courtly songbooks collected as much as a century earlier. Milán advised singer and instrumentalist alternately to embellish (‘glosar’) some of these ballads and later vihuelists also recommended judicious ornamentation; this should be seen, perhaps, not as an innovation but as another stage in the traditional re-creation of the cadential flourishes found in the ballads of the Cancionero Musical de Palacio (ex.3b). Possibly the insertion of instrumental interludes at the ends of phrases and between strophes also reflects this re-creative process, though the vihuelists were clearly more concerned with providing opportunities for the exercise of vihuela technique than with preserving authentic traditions. Romance tunes of the older type appeared less and less frequently as the century progressed. They may be detected occasionally in the works of 16th-century composers such as Morales (Missa ‘Desilde al cavallero’) and Gombert (Dezilde al cavallero, a madrigal based on the same tune, as is Por vida de mis ojos by Juan Vásquez and Cabezón’s organ variations Diferencias sobre ‘El caballero’). During the latter half of the century the old tunes seem to have been subjected to such a degree of re-elaboration by oral transmission or cultured refinement, or both, that they are barely recognizable as derivatives of the older type. In the Cancionero Musical de Medinaceli (E-Mmc 13230), for example, these apparently new types have music in a variety of duple and triple rhythms and forms (ABCDD, ABCDCC, ABCDEFG etc.). By 1600 published ballads suggest that either the old type of tune had been altered practically out of recognition, or it had been replaced by a categorically ‘new’ type.
A consistent characteristic of the ‘new’ romance was the addition of a refrain. The ballads in a collection from the early part of the century, Romances y letras a tres vozes (E-Mn 1370–72; ed. in MME, xviii, 1956), appear with refrains (called ‘estribillo’, ‘buelta’ or ‘letra’). Musically these are given prominence over the quatrains, which are often set in a new quasi-recitative style with repeated phrases and agitated rhythms. The refrains are more melodic, use more imitation, are often in a different metre and are up to ten times as long as the strophes, even when the refrain texts are short. The refrain was often printed after each strophe, but it seems to have been sung increasingly at the beginning. By 1630, then, in the Cancionero Musical de la Sablonara (ed. J. Etzion, London, 1996) for example, the romance had become, from a musical point of view, virtually indistinguishable from the villancico. Later in the century different verses were frequently provided with different music, and the refrains were set for eight to 12 parts in contrast to the three or four parts of the strophes. They were accompanied by organ, harp and continuo.
In poetry, however, the romance form outshone the refrain song during the 17th century and beyond. Manuscript collections (see Wilson) occasionally contain ballad texts with guitar cyphers (at first numbers, later an ‘alphabet’) over the words, indicating chords to be strummed to a known tune. Such evidence shows that ballads were sung to guitar accompaniment in at least some households. Most 17th-century composers (e.g. Juan Blas de Castro, Juan de Palomares, Juan del Vado, Carlos Patiño, Manuel Machado, J.-B. Comes, Juan Hidalgo) set the new romances with their refrains to music for performance both in aristocratic circles and in the public theatre. Indeed, some of these cultured romances became very popular, especially when incorporated into musical interludes (bailes) in the theatre.
Thousands of sacred so-called romances for three to six voices and accompaniment are extant; they were sometimes interpolated into Matins and sung at the end of a Nocturn in place of the responsory.
There is little evidence to suggest that any kind of romance enjoyed a comparable vogue as a song in the 18th century. Works called romances, but properly more like villancicos, took on the appearance of cantatas, normally consisting of an instrumental introduction, recitatives, arias and choruses. In the first half of the century the accompaniment was predominantly the organ, basso continuo and two violins; later oboe, horn and other instruments were added. Composers of these cantata-like romances, mostly representatives of the Valencian school, include José Pradas Gallen, Pedro Rabassa, Pascual Fuentes and Francisco Morera. Estébanez Calderón, writing in 1847, named ‘un romance o corrida’ as one of the (presumably traditional) gypsy songs he had heard and seen performed; romances sung in 20th-century flamenco style, however, seem to bear no musical relationship to the old type of tune of the Cancionero Musical de Palacio. 20th-century Iberian ballads often have texts that go back as far as the 15th century, though their tunes, with the intervals, rhythms, triplets and cadences characteristic of other 20th-century folksong in the peninsula, are far removed (see Spain, §II, 3). There are many variants and versions of these 20th-century tunes, which can be grouped into families with interrelationships rather as they can in English balladry. The verse form is still predominantly the octosyllabic quatrain, though the romancillo with shorter lines (six syllables or even seven and five) is more in evidence. They are probably as often performed by choirs as by solo voice with guitar accompaniment. Romances de ciegos (sung and/or accompanied by a blind man or woman) sometimes retain something of the balanced shape and restraint of the old ballad.
In recent years many ballads have been collected from communities of Sephardi Jews, a people banished from Spain in 1492. The texts are, again, interesting derivatives of 15th- or 16th-century romances, yet the tunes are so far removed from the early melodies and so akin to modern Iberian folksong that they point once more to the conclusion that musical traditions are more volatile than literary traditions.
The romance was taken to the New World by the first explorers and colonists, and retained its roots in the Spanish form. Many romances such as Del gadina, Estaba Catalina and Muérete de Elena survive, and were performed by travelling fiddlers or payodores who accompanied themselves on the guitarrón (large guitar). Versions of traditional ballads are still sung by the Spanish population in the south-west USA, as are the décima (a type of romance that uses the octosyllabic verse in ten-line stanzas) and the canción. Variants have been collected in Mexico, Colombia (see Beutler), Venezuela, Argentina, Chile (see Danneman) and Uruguay. Like the old Spanish romances, Latin American romances often celebrate important events, both sacred and secular, although the religious romance prevails today, albeit in a fragmentary manner. The Mexican type shows both a literary and a musical affinity with the Spanish, especially the Andalusian with which it shares the name ‘corrido’. Although there are many metrical variants of the Mexican corrido, the octosyllabic quatrain is the most common, often with an added refrain (estribillo). The major modes are more in evidence than in Spanish ballads, but the melodic structure retains the same resemblance to older liturgical models. In both the Mexican and Andalusian types ternary rhythms are prevalent. The corrido is accompanied usually by the guitar, often the harp and, in Jalisco, an instrumental ensemble (mariachi) that includes violins. Late 20th-century studies show that the Spanish romance tradition is reflected not only in the song repertory of direct descendants of the Spanish population, but also in that of black rural communities that have appropriated the romance and incorporated it into their festive traditions (see Friedmann, 1996).
See also Cancionero.
The French used the term ‘romance’ in the first half of the 18th century to denote a strophic poem recounting an ancient story of love and gallantry. Essential to the genre were the qualities of naturalness, simplicity and naivety. The earliest definition (Nouveau dictionnaire de l'Académie françoise, 1718) suggests a link to the Spanish romance: ‘Mot tiré de l'Espagnol, et qui signifie une sorte de Poésie en petit vers, contenant quelque ancienne histoire’. In the 1750s the works of the poet F.A. Paradis de Moncrif established the defining features of the genre; his most famous romance, Les constantes amours d'Alix et d'Alexis, accompanied by a simple, modal ‘air languedocien’, was cited as a model for the genre by J.-J. Rousseau (Encyclopédie, 1765). Rousseau (Dictionnaire de Musique, 1768) presented the first primarily musical definition of the term, suggesting that the melody should reflect the qualities of the poem: ‘point d'ornemens, rien de maniéré, une mélodie douce, naturelle, champêtre’. One of the most important examples of the new genre is Rousseau’s ‘Dans ma cabane obscure’ from Le devin du village of 1752 (ex.4). The strophic form, recurring three-bar phrases, thin texture and narrow range reflect the naive, natural state of the young peasant. Rousseau creates a sentimental mood through expressive devices such as the sudden expansion of the melodic range in the second half and the fuller texture and chromatic harmonic shifts near the final cadence. Mondonville's Titon et l'Aurore, composed in the same year, includes a gavotte en musette that alternates between major and minor, a common practice in the later romance.
The romance grew enormously in popularity during the next three decades. The genre's strophic form, unadorned melody, subordinate accompaniment and simple expression were ideally suited to exploit the vein of sentimentalism in opéra comique. Romances were composed by the leading opéra comique composers of the time, including F.-A. Philidor (Le sorcier, 1764), Monsigny (Le roi et le fermier, 1762) and Grétry (L'amitié à l'épreuve, 1770).
The romance also continued to appear in collections for drawing-room performance. In the first printed set, Recueil de romances historiques, tendres et burlesques tant anciennes que modernes avec les airs notés (1767), Charles de Lusse distinguished between three types: romances historiques, romances tendres and romances burlesques. Other notable collections include Romances d'Arnaud Berquin mises en music par de Blois et Gramagnac (1776) and Consolations des misères de ma vie ou Recueil d'airs, romances et duos de J.-J. Rousseau (1781). These works are distinguished from opera romances by their limited accompaniment (unaccompanied or basso continuo only) and by the frequent use of eight or more stanzas (Rousseau's Pour quoi vompre has 29). J.-P.-G. Martini's Plaisir d'amour (1783/4) is often cited as a landmark in the development of the romance: it has a realized accompaniment, and the normal strophic form is replaced by a rondo pattern, which became increasingly popular in the 19th century. By the end of the 1780s the sentimental romance was very fashionable, especially at the court of Marie Antoinette, who is credited with the composition of the romance Ah s'il est dans mon village.
With the advent of the Revolution the romance changed rapidly. The charm and sentimentality of Rousseau's conception were replaced by patriotism (Gossec's Ode sur l'enfance), reflection of current events (François Devienne's Romance patriotique sur la mort du jeune Bara), tragedy (Marie Stuart, by Martini) and terror (La mort de Werther, by L.-E. Jadin). A new and freer lyricism developed, and the accompaniment, generally indicated for ‘piano, harp or guitar’, became more elaborate. By the turn of the century there was a continuous flow of these songs, affirming their immense popularity. Leading composers of the genre include Gossec, Méhul, Rodolphe Kreutzer, C.-H. Plantade, P.J. Garat and Boieldieu. One indication of the extent of romance production is the appearance of ten volumes of romances entitled Collection de morceaux de chant, extraits des meilleurs acteurs avec accompagnement de guitare published in Paris in the early 19th century. Although the romance was well received by the general public, many contemporary critics held the genre in complete contempt. The romance continued to appear in opéra comique, especially in the works of Méhul (Joseph, 1807, contains two excellent examples), Cherubini, Boieldieu and Auber. As the century progressed operatic settings became more elaborate, often including choruses and exploiting orchestral colour. The extensive ornamentation in Rossini's ‘Sombre forêt’ from Guillaume Tell (1829) exemplifies an attempt to dramatize the romance.
After 1830 the romance as a song type began to give way to the more dramatic mélodie. Several attempts to dramatize the romance (such as Le songe de Tartini which contains a virtuoso violin part) were unable to revitalize the genre. The most successful composers during this period were Antoine-Joseph Romagnési, Pauline Duchambge, Auguste Panseron, Loïsa Puget and Francesco Masini. In 1846 Romagnési suggested the following classifications for the romance: romances sentimentales, mélodies reveuses et graves, chants héroique et fortement rhythmés, romances passionnées et dramatiques and chasonettes. The broadening application of the term during the Second Empire (1851–70) meant its dissolution as an independent genre, and the term ‘romance’ soon became interchangeable with ‘chanson’ and ‘mélodie’. Of the composers who continued to produce these works in quantity, Lamartine and Monpou were the most outstanding. Other leading 19th-century composers who used the term include Berlioz, Gounod, Bizet, Saint-Saëns, Lalo, Duparc and Fauré. Since 1870 the term has been applied to a variety of works with no common form or expression, but characteristic traits (such as references to the troubadours, old Spanish or pre-revolutionary French settings, amorous adventures, simplicity and lyricism) survive.
French influence on the German Romanze in the 18th century was considerable; the earliest German examples imitated the strophic form, simple melodies, phrase structure and harmonies of the Parisian models. Elements of folksong were also assimilated, the Romanze giving a distinctive German character. The term ‘Romanze’ was introduced in 1756 with the publication of three poems by J.W.L. Gleim. The earliest published collections of poems, Romanzen mit Melodien, und einem Schreiben an den Verfasser derselben (Friedrich Lowen, 1762) and Romanzen mit Melodien (Daniel Schiebeler, 1767) were both set to music by J.A. Hiller. Other 18th-century Romanze composers include Johann André, G.W. Gruber, F.W. Weis, C.G. Neefe and E.J.B. Lang. The folk quality of the Romanze was ideal for expressing simple sentiments, and it soon became a standard feature in Singspiel, as in Hiller's popular ‘Als ich auf meiner Bleiche’ in Die Jagd (1770), which attained an extraordinary level of popularity. After 1770 most German comic operas contained between one and three Romanzen, although they were less common in Viennese comic opera. Romanze texts often suggested ancient or exotic places. Mozart exploited this element in Pedrillo's Romanze in Die Entführung aus dem Serail with its modal features and pizzicato accompaniment, suggesting a Spanish guitar.
Many of the leading German poets, such as Herder, Goethe, Schiller and Tieck, provided Romanze texts, and settings of their works often appeared in collections, such as Göthe's Lieder, Oden, Balladen und Romanzen by J.F. Reichardt (1809) and Lieder, Balladen und Romanzen von Göthe by Peter Grönland (1817). The terms ‘Romanze’ and ‘Ballade’ were frequently interchanged, and even contemporary theorists had difficulty in differentiating them. One proposed distinction was a matter of emphasis: the Ballade maintained an epic character, as the story was the essential feature; the Romanze, though less dramatic and lyric than the lied, placed more emphasis on musical elements, especially the vocal line. The Ballade also had the character of a gloomy, northern European work, while the Romanze had that of a warm love poem from southern Europe. Settings of romances with French texts and Romanzen with both French and German texts were published sporadically throughout the 19th century. Among the works bearing these designations are Beethoven's Que le temps me dure (text by Rousseau), Schubert's Sah' ein Knab and Das Wasser rauscht, and Wagner's Mignonne, allons voir and Adieux de Marie Stuart. Many 19th-century stage works contain two or three Romanzen. A common feature is the alternation of major and minor sections, as exemplified in Schubert’s ‘Die Vollmond strahlt’ from Rosamunde (1823). Gruesome texts (‘Ein Vampyr nimmt wohl die Gestalt von jedem Menschen auf’ in Lindpainter's Die Vampyr, 1828) and effective use of orchestral colour (‘Nero, dem Kettenbund’ in Weber's Der Freischütz, 1821) also became prevalent characteristics of the operatic Romanze. Like the French romance, the term ‘Romanze’ lost any specific formal meaning in the 19th century: strophic, rondo, ABA and through-composed patterns were all common.
The term appears frequently in Italy and Russia. In Italian tragic operas of the early 19th century, ‘romance’ was a common title for short arias in a slow tempo. Examples can be found in the works of Bellini (I puritani, 1835), Donizetti (L'elisir d'amore, 1832) and Verdi (Il trovatore, 1853). Expanded romances appeared in later works by Verdi, Ponchielli and Puccini before the genre vanished with the advent of the continuous musical flow of verismo. At St Petersburg, where Parisian manners were much admired, the romance was cultivated in the 18th century both in French comic operas and as a song type. In the 19th century settings of poems by Pushkin were common, and composers in the genre included Glinka. Because of the loose application of the term it is difficult to specify how many works were composed, but in general, lyricism and folk qualities were important elements. Many later Russian composers, including Tchaikovsky, Rachmaninoff and Shostakovich, continued to use this title. English composers, preferring ‘ode’ and ‘ballad’, rarely employed the term.
The
simplicity, lyricism and form of the vocal romance were easily adapted
to instrumental composition. In the 18th century the term was most frequently
applied to slow movements with a rondo, ABA or variation structure.
Gossec's Symphony in E
op.5 no.2, written for Paris in 1761 or 1762, contains the first known
appearance of such a movement in a symphonic work; Dittersdorf is credited with
introducing it to Vienna in his Symphony in E op.7 no.1 (1773). The variation structure, as
exemplified in Haydn's Symphony no.85 in B, ‘La reine’, was most popular in French symphonies
and in the quatuor concertant. In Germany, instrumental Romanzen in ABA
and rondo forms were common and appeared in a variety of genres: Hoffmeister's
String Quartet op.14 no.2, Mozart's Serenadek525 (Eine
kleine Nachtmusik) and, in the 19th century, Schumann's Symphony in D
minor.
The first solo instrumental romance was composed by the violinist Pierre Gaviniès in about 1760, and the term became a common title for the slow movements in his sonatas and concertos. The genre flourished in the violin and cello schools centring on Paris well into the 19th century, with numerous examples composed by violinists including Viotti, Rode and Kreutzer. Beethoven employed this form in three single-movement works, Romanze cantabile for piano, flute, bassoon and orchestra and two Romanzen for solo violin and orchestra, opp.40 and 50. The latter works, both in rondo forms, are models of the balance between lyricism and virtuosic display that can be found in later works, such as the slow movement of Wieniawski's Violin Concerto no.2. Both European and American violinists continued to cultivate the genre up to the end of the century.
The earliest Romanzen for piano solo are the simple tunes in D.G. Türk's Handstücke für angehende Klavierspieler (1792–5) and Reichardt's Kleine Klavier- und Singstücke (1783). More elaborate settings appeared in southern Germany during the late 18th century, using ABA or rondo structures. Variations on a Romanze theme, such as Hummel's Thèmes variés and Clara Wieck's Romance variée, appeared occasionally in the 18th and 19th centuries. Mozart used the term for the slow movement of his Piano Concerto in D minor k466; its rondo-like structure is highlighted by the return of the simple, unadorned tune after a contrasting section of stormy virtuosity. Other concertos with movements of a similar construction include one of Haydn's for two lire organizzate (the movement later reappears in the ‘Military’ Symphony, no.100), several of Mozart's horn concertos and Chopin's Piano Concerto in E minor. The most frequent application of the term in the 19th century was to small character-pieces with no common formal pattern, for example Schumann's Drei Romanzen op.28. In such works the Romanze bears connotations of love or antiquity and is predominantly lyrical. It is this last feature that has remained constant in the wide variety of instrumental Romanzen.
StevensonSM
E.M. Torner: ‘Ensayo de clasificación de las melodías de romance’, Homenaje ofrecido a Menéndez Pidal, ii (Madrid, 1925), 391–402
J.M. de Cossío, ed.: Romancero popular de la Montaña (Santander, 1933)
J. Bal y Gay, ed.: Romances y villancicos españoles del siglo XVI (Mexico City, 1939)
W.J. Entwistle: European Balladry (Oxford, 1939/R)
R. Menéndez Pidal: Romancero hispánico (hispano-portugués, americano y sefardí) (Madrid, 1953, 2/1968)
M. Querol Gavaldá: ‘Importance historique et nationale du romance’, Musique et poésie au XVIe siècle: Paris 1953, 299–328
D. Devoto: ‘Sobre el estudio folklórico del romancero español’, Bulletin hispanique, lvii (1955), 233–91
J.F. Montesinos: ‘Algunos problemas del romancero nuevo’, Ensayos y estudios de literatura española, ed. J.H. Silverman (Mexico City, 1959, 2/1970), 75
E. Gerson-Kiwi: ‘On the Musical Sources of the Judaeo-Hispanic Romance’, MQ, l (1964), 31–43
P. Bénichou: Romancero judeo-español de Marruecos (Madrid, 1968)
R. Goldberg: ‘Un modo de subsistencia del romancero nuevo: romances de Góngora y de Lope de Vega en bailes del Siglo de Oro’, Bulletin hispanique, lxxii (1970), 56–95
D. Catalán, S.G. Armistead and A. Sánchez Romeralo: El romancero en la tradición oral moderna (Madrid, 1972)
I. Katz: ‘The Traditional Folk Music of Spain: Explorations and Perspectives’, YIFMC, vi (1974), 64–85
J. Sage: ‘Early Spanish Ballad Music: Tradition or Metamorphosis?’, Medieval Hispanic Studies Presented to Rita Hamilton, ed. A. Deyermond (London, 1976), 129–214
S.G. Armistead and others: El romancero judeo-español en el Archivo Menéndez Pidal: catálogo-índice de romances y canciones (Madrid, 1978)
A. Sánchez Romeralo, S.G. Armistead and S. Petersen: Bibliografía del romancero oral, i (Madrid, 1980)
J. Etzion: ‘The Spanish Polyphonic Ballad in Sixteenth-Century Vihuela Publications’, MD, xxv (1981), 179–97
R. Goldberg: Tonos a lo humano y a lo divino (London, 1981)
K. Kish: ‘The Spanish Ballad in the Eighteenth Century: a Reconsideration’, Historical Review, xli (1981), 271–84
M. Frenk: ‘Lectores y oidores: la difusión oral de la literatura en el Siglo de Oro’, Actas del VII Congreso de la Asociación Internacional de Hispanistas (Bulzoni, 1982), 101–23
I. Katz: ‘The Musical Legacy of the Judeo-Spanish Romancero’, Hispania Judaica: Studies on the History, Language and Literature of the Jews in the Hispanic World, ii, ed. J. Solà-Solé, S.G. Armistead and J. Silverman (Barcelona, 1982)
M. Trapero and L. Siemens, eds.: Romancero de Gran Canaria (Las Palmas, 1982–90)
M. Trapero: ‘Romancero y teatro popular en la tradición oral castellano-leonesa’, Revista de dialectología y tradiciones populares, xxxviii (1983), 37–53
J. Díaz: ‘Melodías prototipo en el repertorio romancístico’, AnM, xxxix–lx (1984–5), 97–106
M. Manzano: ‘Estructuras arquetípicas de recitación en la música tradicional’, RdMc, ix (1986), 357–97
M. Cruz de Enterría: ‘Romancero: cantado, recitado, leído?’, Edad de Oro, vii (1988), 89–104
J. Etzion: ‘The Spanish and Sephardic Romances: Musical Links’, EthM, xxxii (1988), 1–37
J. Etzion and S. Weich-Shahak: ‘The Music of the Judeo-Spanish Romancero: Stylistic Features’, AnM, xliii (1988), 221–56
I. Katz: ‘Contrafacta and the Judeo-Spanish Romancero: a Musicological View’, Hispanic Studies in Honor of Joseph Silverman, ed. J. Ricapito (Newark, NJ, 1988), 169–87
J. Cohen: Judeo-Spanish Songs in the Sephardic Communities of Montreal and Toronto (diss., U. of Montreal, 1989)
S. Armistead: ‘Bibliografía crítica del Romancero (1984)’, Actas del Congreso Romancero-Cancionero, ii (Madrid, 1990)
M. Frenk: ‘La poesía oralizada y sus mil variantes’, Anuario de Letras, xxix (1991), 133–54
J. Dias Marques and W. Brunetto: ‘O romanceiro e as cantigas da segada’, Estudios de folklore y literatura a Mercedes Díaz-Roig, ed. B. Garza and Y. Jiménez de Báez (Mexico City, 1992), 171–208
A. Cauffriez: La perennité du romanceiro dans la musique populaire du Trás-os-Montes (Paris, 1993)
D. Devoto: ‘Un millar de cantares exportados’, Bulletin hispanique, xcvi (1994), 5–116
J. Vicuña Cifuentes: Romances populares y vulgares, recogidos de la tradición oral chilena (Santiago, 1912)
C. Poncet y de Cárdenas: El romance en Cuba (Havana, 1914)
V.T. Mendoza: El romance español y el corrido mexicano (Mexico City, 1939)
R. Menéndez-Pidal: Los romances de América y otros estudios (Madrid, 1958)
G. Beutler: Studien zum spanischen Romancero in Kolumbien in seiner schriftlichen und mündlichen Uberlieferung von der Zeit der Eroberung bis zur Gegenwart (Heidelberg, 1969)
M. Danneman and R. Barros: El romancero chileno (Santiago, 1970)
G. de Granda: Estúdios sobre un área dialectal hispanoamericana de población negra (Bogotá, 1977)
G. List: Music and Poetry in a Colombian Village (Bloomington, IN, 1983)
M. Díaz Roig: ‘El romancero tradicional en América’, El romancero: tradición y pervivencia, ed. V. Atero (Cádiz, 1989), 651–68
S. Friedmann: ‘Estrategias orales y la transmisión musical del romancero en las tierras bajas de Colombia’, El romancero: tradición y pervivencia, ed. V. Atero (Cádiz, 1989), 701–17
M. Herrera-Sobek: Northward Bound: the Mexican Immigrant Experience in Ballad and Song (Bloomington, IN, 1993)
S. Friedmann: The Festive Song Repertory of Barbacoas, Colombia, and its Implications for Ballad Transmission (diss., U. of London, 1996)
M. da Costa Fontes and I. Katz: O romanceiro portugues e brasileiro: indice temático e bibliográfico (Madison, WI, 1997)
I. Katz: ‘Romance, romancero’, Diccionario de la música española e hispanoamericana, ed. E. Casares Rodicio (Madrid, 2000)
J.G. Sulzer: Allgemeine Theorie der schönen Künste (Leipzig, 1792–4)
H. Engel: Das Instrumentalkonzert (Leipzig, 1932, 2/1971–4)
S. Goslich: Beiträge zur Geschichte der deutschen romantischen Oper (Leipzig, 1937, enlarged 2/1975 as Die deutsche romantische Oper)
H. Gougelot: Catalogue des romances françaises parues sous la Révolution et l’Empire (Melun, 1937–43)
H. Gougelot: La romance française sous la Révolution et l'Empire (Melun, 1938–43)
B. Lupo: ‘Romanza, notturni, ariette nel primo Ottocento’, RaM, xix (1941), 81–91
F. Noske: La mélodie française de Berlioz à Duparc (Amsterdam, 1954; Eng. trans., rev. R. Benton and F. Noske, 1970/R)
M. Pincherle: Discussion in Les influences étrangères dans l'oeuvre de Mozart: Paris 1956, 251ff
D.C. Ossenkop: The Earliest Settings of German Ballads for Voice and Clavier (diss., Columbia U., 1968)
D. Heartz: ‘The Beginnings of the Operatic Romance: Rousseau, Sedaine and Monsigny’, Eighteenth-Century Studies, xv (1981–2), 149–78
A. Caswell: ‘The French Romance’, Music in Paris in the Eighteen-Thirties, ed. P. Bloom (Stuyvesant, NY, 1987), 97–115
H. Gülow: Studien zur instrumentalen Romance in Deutschland vor 1810 (Frankfurt, 1987)
R. Gstrein: Die vokale Romanze in der Zeit von 1750 bis 1850 (Innsbruck, 1989)
D. Charlton: ‘The romance and its Cognates: Narrative, Irony and vraisemblance in Early Opéra Comique’, Die Opéra Comique und ihr Einfluss auf das europäische Musiktheater im 19. Jahrhundert: Frankfurt 1994, 43–92