Principles of Selection
The Irish Song Project is an exploration of the historical development of song in Ireland based on a representative selection of 220 songs from the Middle Ages up to the mid-nineteenth century. 'Irish song' here refers to songs composed or performed in Ireland in all of the languages for which evidence survives, thus not only songs in Gaelic and English but also, for the medieval period, in Latin and French as well. Thus we are seeking not so much to trace a single trajectory but rather to provide a comprehensive overview of all song types and genres. At its most ambitious, then, the Irish Song Project Database attempts to provide a snapshot of around 1000 years of singing culture in Ireland. From medieval chant to Gaelic-era poetry, improvised work-songs to nineteenth-century ballads, the database maps the wide varieties of song form and genre in Ireland, during a millennium of political, economic and cultural upheaval. The database does not promote a particular political viewpoint or promote one song genre above another. In this way we hope the database will challenge commonly held views about the history of song making and performing in Ireland, demonstrating that diverse forms of musical expression existed across social, linguistic, political, and religious divides.
Principles of selection
The central focus of this database is melody and thus we have included song texts only where a tune can be identified. Each song in the database is analysed textually and musically and presented together with facsimiles of original sources of both tune and song text, where available. Examples vary from those tunes which were transcribed or printed with an accompanying song text, to those for which we have collated tune and text from different sources. Transcriptions of both text and music in modern formats are also provided.
In selecting sources we have adhered to the following principles of selection:
- the earliest printed or written Irish source
- the earliest printed or written non-Irish source
- transcription from the oral tradition where appropriate.
Songs have been taken from manuscript and print traditions according to the criterion of the earliest Irish copy. However, although Dublin printings of ballads and songs survive from the early seventeenth century onwards (see 'Mount Taraghs Triumph', 1626), the greatest number of Irish sources date for the most part from the 1750s onwards. The most significant exception to this general rule is A Collection of Celebrated Irish Tunes (Dublin, 1724), which is the first known printed collection of Irish airs. We have thus chosen a generous selection of examples from this book, drawing upon Nicholas Carolan’s research in identifying song-texts for these tunes. A number of London-printed song-sheets from the early eighteenth century have been included where these claim that the song was performed in Ireland (e.g. 'Marinda's face like Cupid's Bow’, c.1710) or where they later entered a popular song repertoire as purportedly 'Irish' tunes (e.g. the tune for 'My bonny dear Shonny' 1683 which would become known variously as the 'Irish Jigg' and the 'Irish Trot'). We have also occasionally used a print source from beyond Ireland where its dating is known to be more consistent with performance in Ireland than a later Irish source (e.g. 'Shein shees').
By excluding songs for which there is no musical notation, we have thus omitted the earliest sources for liturgical song in Ireland which are found in medieval manuscripts such as the Antiphonary of Bangor (c.700 AD) which contains the oldest known hymn for St Patrick (‘Audite omnes amantes’). The Irish Liber Hymnorum (late 11th/early 12th century) is an equally important source, not least for its inclusion of hymn texts in Gaelic as well as in Latin. We have to turn to manuscripts from several centuries later to find Irish liturgical music with notation, and these are represented in the database by selected items from fourteenth- and fifteenth-century sources. The chants themselves may well be older, but are unlikely to be any earlier than the thirteenth century. Other exclusions include texts in English and in French found in fourteenth-century Irish manuscripts such as the Red Book of Ossory (RCB Library), the ‘Kildare Poems’ (BL Harley 913) and the Waterford collection (Corpus Christi College Cambridge, MS 405). We can be sure in the case of the Red Book of Ossory that these verses were sung, as they are referred to specifically as cantilenae (i.e. songs). However, whether this applies also to the Kildare Poems and to items such as 'Ich am of Irlaunde' from another fourteenth-century source (and so hauntingly adapted by W.B. Yeats in his poem of the same name) is unsure.
Several melodies have been included without song texts, either because they belonged to quasi-improvised songs (see 'Type 2: Simple Melody' below) or because they are identified as tunes which could carry several different texts. An example of this is the 'Popular Ballad Tune' published by George Petrie in 1855, which he states he often heard on the streets of Dublin (p.32).
An initial focus of the project was on the identification and classification of melodic types as a methodology by which continuities of tradition and cultural practice might be made newly visible. For example, the singing of a funeral elegy continued in Ireland and Scotland until the mid-twentieth century and keys into a pattern of singing and social ritual well attested throughout medieval Europe (one which continues in parts of Central and Eastern Europe today). While a given elegy may not in itself be particularly old, its ‘type’ (i.e. melodic elements and textual topoi) is unchanging and thus recognisable as belonging to an older style of singing (and social expression). Such melodies are characterised typologically as formulaic or cellular in structure and relate to oral-tradition performance for which the earliest European textual evidence survives in ninth-century French and German sources (the date of the oldest surviving manuscripts containing Latin song texts with neumatic music notation). It is hoped that future research might expand upon this initial exploratory selection in probing these kinds of question.
The database there includes representative samples from each of the following song types:
- chant-like melody
- simple melody
- complex melody.
Type 1: Chant-like Melody
Songs in this category originate primarily from the mid-to-late medieval period. They are defined by a mostly step-wise melodic pattern, typical of the period. Illustrative examples include 'Audate omnes' and 'Brigit bé bithmaith'.
Type 2: Simple Melody
Songs in this category demonstrate a greater degree of melodic development than those in Type 1, but without the complexity of range and structure which defines examples of Type 3. Selections here are drawn from a wider array of sources than Type 1 and were often printed or written down between c.1790 and c.1900. They include examples of a number of genres and song types which disappeared from their original context during the twentieth century. These include ploughwhistles, spinning songs, lullabies, keens, lays and some forms of quasi-improvised songs. An example of the latter category is the song 'Ullulu mo mhailín' ('Alas my little bag') which was published by P.W. Joyce in 1873. Its relatively simple tune consists of an A phrase and a B phrase of four bars each, both of which are repeated.
Type 3: Complex Melody
Songs in this category form the most extensive part of the collection. Most of these are reproduced from eighteenth- and nineteenth-century sources. Song melodies in this category demonstrate significant musical complexity. They range in genre and form from songs which we consider to be part of the sean nós tradition to four-part odes with full orchestral accompaniment, from printed song sheets to broadside ballads. Something of the diversity here can be seen by comparing, for example, the political song 'The Drapier’s Ballad' with 'Dochtúir Boyle'.
Notes on Main Sources
The database draws on the small number of medieval sources which contain music considered by scholars to be Irish in composition. The earliest sources, The Irish Liber Hymnorum and The Bangor Antiphonary, while they contain words in an Irish dialect, do not contain any music. Scholars and performers have reconstructed this work using contemporary tune settings, but we have not included such reconstructions in the database.
The Kilmoone Breviary and the Christchurch Antiphonary (early fifteenth century) and the Clondalkin Breviary (late fifteenth century) are the sources for our earliest selections. However, like many later sources, it should be borne in mind that these important references contain hymns and another religious music that had longer performance histories. For example, the hymn 'Ecce Fulget Clarissima', found in the Kilmoone Breviary, contains the same words as a piece in the twelfth-century Irish Liber Hymnorum, thus allowing us to view an older form of melody.
Material written or printed in the seventeenth century has come from three main types of source: (a) lute books, (b) formal odes and (c) a small number of ballad sheets. The lute books contain a number of items which Aloys Fleischmann considered 'songs' or 'vocal tunes', and while these sources are mostly Scottish in origin, Fleischmann selected them for his collection of Irish music. While there may be some debate as to whether these particular songs were sung in Ireland, the musical links between Scotland and Ireland create sufficient ambiguity to allow us to include a few examples from these sources. These may well demonstrate the type of English language vernacular song which was sung in this period.
Odes were written and performed for formal occasions, such as state, monarchical or institutional anniversaries. Although they were composed in the style of European art music, these pieces are included as examples of Irish song because of their specifically Irish context; they were never performed outside Ireland and thus represent a specific type of Irish song that differs significantly from other forms selected in the database. Currently we display one of these but will be adding two further examples in the next few months.
Important eighteenth-century collections which give an insight into musical tastes and fashions during this period include A Collection of the Most Celebrated Irish Tunes (Dublin: John and William Neal, 1724); two collections of airs edited by Burk Thumoth, Twelve Scotch and Twelve Irish Airs with Variations Set for the German Flute, Violin or Harpsichord (London: Simpson, [c.1745]) and Twelve English and Twelve Irish Airs […] Book the Second (London: J. Simpson, [c.1746]); and The Hibernian Muse; a Collection of Irish Airs (London: S. A. & P. Thompson, ). Of these, the most significant is the Neals’ collection as it is the earliest-known printed collection of Irish music. Only one copy of this book is known to have survived and this unique copy – once owned by Edward Bunting – is now held in the Special Collections of the McClay library at Queen’s University Belfast. A modern facsimile edition, edited by Nicholas Carolan (1986; revised 2010), identifies song-texts for twenty-two of the forty-nine tunes. We are grateful to Nicholas Carolan for his permission to reproduce the text transcriptions and translations for these songs from his edition. This period also saw a significant growth in the publication of song-sheets: only an illustrative sample has been selected from these. The earliest examples are printed in London, although they often indicate performances in Ireland (such as the three songs by Richard Leveridge identified as being ‘sung in the theatre in Dublin’). Other London song-sheets printed tunes which would circulate to different song texts throughout the century, including within Ireland (the earliest example included here is the 1683 song ‘My Bonny dear Shonny’). Dublin-printed song-sheets begin to proliferate in the 1770s, many published by Walker’s Hibernian Magazine. The earliest Dublin-printed song-sheet known to survive is ‘The Tipling Philosophers’ (1720) and as such is an obvious selection for the database. A sample of broadside ballads has been included where we have been able to source a printed or transcribed tune for the song. Among this category are three songs on the Wood’s Halfpence affair (1724-25). Research for this project has identified at least eighteen Dublin-printed songs on this controversy, so we have chosen a representative sample of those to which a tune can be assigned.
A sizeable portion of the selections in this database are taken from the Edward Bunting Notebooks held in the McClay Library’s Special Collections at Queen’s University Belfast. Text and tune source references given in the database refer to the item as catalogued in Colette Moloney, The Irish Music Manuscripts of Edward Bunting (1773-1843): An Introduction and Catalogue (Dublin: Irish Traditional Music Archive, 2000). Readers should note that index numbers in this catalogue do not always map onto the library’s internal catalogues, and these alternative listings are given in the item’s individual note. Moloney’s system gives each example an integer, allowing items to be simply marked according to their individual notebook and item number. The Bunting notebooks are known as MS 4 in the library catalogue. The collection straddles the eighteenth and nineteenth centuries, beginning with Bunting's transcriptions from the Belfast Harpers' Assembly of 1792. Bunting subsequently sponsored several field collecting trips on the west coast and in Gaelic-speaking regions of the north. The manuscripts contained source material for three major publications in 1796, 1809 and 1840 and many of the tunes he collected were adapted by Thomas Moore in his Irish Melodies.
The songs of Thomas Moore remain some of the most enduring examples of Irish song from the nineteenth century. The database includes a number of items from Moore's various publications, specifically those written to instrumental melodies, as opposed to tunes that previously carried a text (e.g. 'Lesbia hath a beaming eye' to the tune of the piping jig 'Nora Criona') and those which establish a tune from another tradition as 'Irish' (e.g. 'Believe me if all those endearing young charms' the tune of which was previously considered to be English). The publications of the antiquarians George Petrie (1790-1866) and P.W. Joyce (1827-1914) provide us with a source for many examples of song from the nineteenth century. Although these books (Joyce: 1873, 1888, 1909 and Petrie: 1855, 1902-5) were published after 1850, they are reflective of repertoires and practices that were historically grounded. For example, Joyce published numerous songs that he learned from his father in childhood, thus placing them well within the project's date range. Petrie collected and published material from a variety of sources, many of which were verifiably in practice before 1850. Royal Irish Academy MS12 Q13, written by Pronsiais Ó Cathain of Dublin between 1873 and 1876 is an essay on the state of the Irish language in Ulster at this time. It contains a number of work-songs collected by the author across the north of Ireland, most of which are not found elsewhere in these forms. It is probable that this is the same Frank Keane whom Petrie cites as a source for more than twenty tunes or songs in his manuscripts. Some examples from the manuscripts of William Forde and John Pigot, a major source in the holdings of the Royal Irish Academy, have also been included in the database. While much of this material consists of a collation of other major sources (such as Bunting's publications), many items are found uniquely within the collection. Some settings of lullabies and keens are included in the database.
A number of selections from twentieth-century collections have also been made, for example, from The Songs of Elizabeth Cronin (Dublin: Four Courts Press, 1999). This book contains transcriptions of both song-texts and melodies, of which this publication represents the earliest evidence. In selecting from such publications, the dating of the item has been verified as pre-1850 and modern collections have been cited only if no older source has been forthcoming. In a small number of instances we have transcribed words from the aural tradition where no earlier publication exists, such as in the example of the series of 'lullaby' tunes transcribed by William Forde. The text in this instance was transcribed from the singing of John Doherty in 1953.
The Irish Song Project team welcome corrections or suggestions for further selections. These can be sent to the following e-mail address: