Ballads/Romances. Origins. Classification.

The Spanish romances (or ballads) are widely recognised as constituting one of the largest and richest ballad traditions in Europe. Their popularity stretches from the Middle Ages to the twentieth century. But like other works of the Middle Ages, e. g. the Poema de mío Cid, the Libro de Buen Amor (The Book of Good Love) and La Celestina, the romances come with accompanying problems, in this case centred on questions of origins and classification.


The origins of the romances are contentious, with debates swirling especially over whether they derive from medieval epics or have another root.  The discussions hinge primarily on whether the ballads came first, appearing as anonymous poems sung by the “people” (el pueblo), or whether they were fragments extracted from cantares de gesta, epics composed by educated or cultured poets (whose names are unknown, as for example in the case of the Poema de mío Cid or the Poema de Fernán González). 

This epic/ballad issue was brought to the forefront in the 19th century, when German romanticism was in full swing and “popular” literature was appreciated for its closeness to “nature.” It was argued that the ballads --along with other kinds of anonymous lyric poems-- were spontaneous expressions of the people which in some way or other embodied the spirit of the nation. 

For the German romantics the ballads came first and the cantares de gesta were the later result of stringing together numerous ballads by artistically conscious and more cultured individuals.  By the mid 19th century, however, this idea of the “popular,” spontaneous origins of the ballads preceding the epics was gradually turned on its head. It was now argued that epic poems were the artistic products of individual authors, that they in fact preceded the ballads and that these were fragments removed from the epics. The popularity of the new form and its catchy rhythm were then quickly transferred to a wide variety of other, non-epic themes. The eminent scholar, Samuel Armistead, is quite categorical about this: “The first ballads undoubtedly started as fragments of medieval epic poems; non-epic historical events soon came to be narrated in the same verse as the newly independent epic fragments” (in Gerli 138).

This “fragmentist” theory was taken up and modified by Ramon Menéndez Pidal, the most authoritative early 20th-century figure in the history of the romance. Menéndez Pidal acknowledged the primacy of the individual author in the creative act but also insisted that the people (el pueblo) had a significant role in the subsequent re“creation” (i. e. retelling) and longevity of these romances. In other words, the people who had accepted a passive role when listening to the epics, now took on an active one in transmitting the fragments. Menéndez Pidal’s ideas --known as Neotraditionalism-- are still widely accepted, although not without challenge.

But how might this epic to ballad fragmentation have happened and when?  For the “how,” there are two major sources from which the fragments were extracted: the poems themselves and prose chronicles.  The latter contained prosified version of the poems, including some epics which are otherwise partially or totally lost (i. e. it is only through the chronicles and their use of “traditional epic motifs, or narrative elements” Deyermond 33) that we know of their existence. The poems and chronicles were then picked over by juglares (minstrels) who chanted or recited them to the public selecting the most dramatic and entertaining parts with most popular appeal.  Whereas the cantares de gesta poems were long and required professional minstrels, the fragments could be more easily memorised by the public and be passed on from generation to generation in homes or local communities.  However, since the world of oral transmission is by nature unstable, differences (i. e. variants) inevitably crept in. We know this because more than one version of these romances were copied when they were first transcribed late in the 15th or in the 16th centuries.  Colin Smith (14) puts it succinctly: “If modern scholars collect 500 version of a ballad, no two exactly correspond.” In addition, fragments from different parts of an epic or even from different poems might also be conflated as identities or events became mixed up.

Let’s turn now to the question of when the fragmentation took place. The simple answer is we don’t really know. It is generally agreed that epic poetry declined in popularity during the 14th century, and that coincides with the appearance of the romances. The trouble is that we don’t have written evidence of the existence of these epic-based romances because they weren’t transcribed until, at the earliest, the end of the 15th century and most only in the 16th century (and by way of interest, some old ballads were not unearthed until the 20th century, early on by Menéndez Pidal and his wife María Goyri; another source has been Sephardic communities from Turkey and Greece to Los Angles and Seattle. See

In fact, the first recorded example so far of a romance does not appear until 1421 and it has nothing to do with any epic poem (it deals with an encounter between a young woman and a married man). But that doesn’t mean that romances were not composed before then. We can go back to the 14th century when ballads were made up during the civil wars (1358-69) between King Peter/Pedro el Cruel and his half-brother Henry/Enrique of Trastámara, or even earlier to 1328, to the first datable romance which deals with an episode in the life of King Alfonso XI of Castile-León, 1311-1350. In the case of the Pedro/ Enrique ballads, we know from their content that they were contemporary with historic events of the time, and for the Alfonso ballad internal references point to it appearing shortly after his death. These clearly are not based on epic material, and in fact are often classified as romances noticieros (i. e. bearing local or contemporary news). 

How can we then say that the romances originated in the epics if the oldest confirmable examples do not appear to support this idea? A possible explanation is that, apart from the single 1421 romance, we have no written record (yet) until late in the 15th century --during the reign of the Catholic Monarchs, Fernando/Ferdinand and Isabel/Isabella 1474-1504-- of any romances, not even for the Pedro/ Enrique cycle. It should not surprise us then that there are no textual examples of ballads from epic material either. There is a possible reason for this: comments by learned poets of the 15th century allude to the romances as inferior songs of the lowest category for the pleasure of people of low status. Presumably, then, such songs were not worth writing down or perhaps were written on pamphlets which have long disappeared.
By the end of the 15th century, however, tastes had changed.
The Spanish language became highly esteemed (witness the
publication of the first Spanish language grammar book, the
first Spanish-Latin dictionary, both published in 1492 by the
same author, the humanist Antonio de Nebrija) and the
romance form was appreciated as evidence of pride in their
native tradition

So, we simply don’t know with certainty which came first, the epic-based or non epic. Indeed, another scholar, Henk de Vries, argues that “It seems likely that the ballad in Spain originated as a means to spread news and propaganda. Once it had become popular –when the market for short narrative had been created-- ballads were composed about epic subjects” ("Literature and Historical Fact," 13). For de Vries, it is likely the romance noticiero that marks the birth of the Spanish ballad, but we are still left with the question: how or why did those who composed these "news" ballads choose the metrical form and rhyming scheme that they were written in?

Is there anything more specific that could give epic-based romances “bragging rights” as the first ballads? Possibly the metrical scheme and rhyming patterns.  The epics were written in segments (laisses) of different length.  In the early epics they generally varied around 14 syllables but by the 14th century (when the romances first appeared) they had stabilised at 16 syllables.  The rhyme was commonly assonance, that is the pairing of the last two vowel sounds were repeated irrespective of the intervening consonants. So, for example “castellAnO” rhymes with pasAdO” or “batAllA” with “AlmA.”

Both the octosyllable (i. e. half the 16-syllable line) and assonance are the most identifiable characteristics of all Spanish ballads, regardless of theme.  In textual form, a romance is written in lines of 8 syllables (sometimes, to save space, in lines of 16 syllables with a strong break -caesura- after the 8th syllable) with a fairly pronounced stress on the 7th syllable which, with other subsidiary stresses, produces a kind of galloping, easy to memorise beat or rhythm.

But neither assonance nor octosyllables were exclusive to the epic and ballad; both existed in earlier Spanish (or rather Galician-Portuguese) lyrical poetry but on a limited scale. A romance-like song (Cantiga 308) in octosyllables by Alfonso X in the mid 13th century in Galician-Portuguese can be found in his series of poems in praise of the Virgin Mary. It rhymes on the even lines, but the rhyming pattern is full rhyme rather than assonance. (Full rhyme is the repetition of identical vowel and/or consonant sounds following the last stressed vowel at the end of two or more lines of poetry: "amarillento" rhymes with "polvoriento", "casar" with "pasar.")

So, conclusion?  The combination of octosyllable and assonance and its regularity in both epic and ballad makes fragmentation from epic to ballad is the most persuasive argument. Following on the popularity of these epic based ballads, other themes were then adopted. Still, the question has not been conclusively resolved. 

Whatever the exact origins of the romance, it was a form that came to incorporate a wide variety of themes from various sources: 1. epic verse/ material (from e. g. Poema de mío Cid, Mocedades del Cid, Poema de Fernán González, Siete Infantes de Lara); 2. events from Spanish history (e. g. the defeat and fall in 711 of Rodrigo (Roderic), the last Visigothic king, or the civil war between King Pedro (Peter) and his half brother, Enrique (Henry) of Trastámara, 1358-69); 3. the Moors and frontier life (in many of which the Moors are viewed sympathetically); 4.  early tales from France, loosely inspired by the court of Charlemagne or Arthurian legends; and 5. more universal lyrical themes: romance (i. e. love), eroticism, mystery, magic, adventure, vengeance, death (known as novelesque ballads). All these medieval ballads fall under the general heading of romances viejos (old ballads). They are often sub-classified –although not without debate-- into romances literarios (numbers 1 and 4), romances históricos (number 2), romances fronterizos (number 3) and romances novelescos (number 5).

Given their undocumented origins, and wide thematic variety, the early or medieval romances have resisted a complete and authoritative classification, although all fall under the general rubric of romances viejos. Many readers still find useful the grouping suggested by the British scholar W. J. Entwistle in 1939. He divided the romances into three general groups: historical, literary and adventure ballads.  The historical (romances históricos) are those inspired by contemporary (or near contemporary) historical events during the late Middle Ages, including those dealing with the civil war (1350-69) between King Pedro/Peter and his half brother and later king, Enrique/ Henry of Trastámara, and frontier relationships between Christians and Moors. The former as sometimes subclassified as romances noticieros, the latter as  romances fronterizos. The literary ballads (romances literarios) include those of epic origin, even when the epic poems deal with historical figures, e. g. the Cid, or Count Fernán González (the founder of Castile) and others. Literary too, are those ballads drawn from legends including those surrounding historical figures about whom no epic was written (e.g. Rodrigo, the last Visigothic king), as well as more exotic material from French Carolingian literature and even distant Arthurian themes reworked by French poets. The historical and literary ballads are often classified in “cycles”, i. e. being a series of ballads based on the same story.  The adventure ballads (romances novelescos) are less homogeneous and have more universal themes.  More lyrical than narrative, these romances--dealing with love --faithfulness, unfaithfulness, seduction, frustration -- vengeance, mystery, death, or simply adventure-- are strongly influenced by folklore and often have a strong symbolic content.

Another suggested classification is to divide the romances viejos into three groups: 1. ballads deriving from the cantares de gesta including those derived from French epics, e. g. the Chanson de Roland which inspired the Carolingian ballads; 2. romances históricos and fronterizos; and 3. the romances novelescos.

There are those who prefer a simpler and even more general division whereby the term romances viejos is limited to those ballads derived from the cantares de gesta. The other early romances fall under the heading of romances juglarescos. These include ballads dealing with historic events, Moorish and Carolingian themes, and fictional ballads. See

Armistead, Samuel G. “Ballads” in Medieval Iberia: An Encyclopedia pp. 138- 140. Ed. Gerli, E. Michael,  New York, London 2003
Deyermond, A.D. A Literary History of Spain: The Middle Ages. London 1971.
Gies, David T. ed. The Cambridge History of Spanish Literature Cambridge UP paperback edition (corrected) 2009
Pinero, Pedro M   Romancero Madrid 1999
Smith, Colin    Spanish Ballads 2nd ed. Bristol 1996
Vries, Henk De  Ballads, "Literature and Historical Fact,"  Jahrbuch für Volksliedforschung 44 Jahrg (1999), pp. 13-23
Walters, D. Gareth The Cambridge Introduction to Spanish Poetry: Spain and Spanish America Cambridge UP 2002
Wright, Roger    Spanish Ballads Warminster: Arid & Phillips 1987  PQ/ 6196/ W76/ 1987