Article MT164, part 24 - newly written for MT.

The Maid on the Shore

(Roud 181, Laws K27)


I was inspired to write this article after coming across a fairly short ordinary version of the ballad in Stuart M Frank's new book Jolly Sailors Bold which is a collection of songs and ballads written down by American whalemen in their journals in the nineteenth century (See Pete Wood's review on this website).  It prompted some discussion online and caused me to look through other early versions I have access to.  In several of the notes to these, editors have expressed the opinion that it is rather unusual in its structure and content and have come to the conclusion that it might be a translation from the Irish or the French.

However Phillips Barry in an article in the Bulletin of the Folksong Society of the Northeast, Bibliographical and Special Series, Volume XI (1960), p.12 -13 presents a strong case for it being a rewrite of a ballad story well-known in many European countries.  In fact its plot particularly resembles versions found in Scandinavia.  During the eighteenth century many of the longer broadside ballads were written by people with rather more literary knowledge than those of a later period.  They often seem to have used a wide variety of sources for the plots of the longer ballads like those printed by William and Cluer Dicey in three to four columns with between 20 and 40 stanzas.  Some of them are reprints or remakes from the seventeenth century, but a significant portion seem to rely on foreign tales for their plots.  A typical example would be The Bramble Briar (Roud 18, Laws M32) which is obviously based on the well-known tale of 'Isabella and the Pot of Basil' found in Boccaccio's Decameron.  Quite a number of the Child Ballads are likely rewrites of ballads and tales from the continent, Scandinavia in particular.  Phillips Barry's article is so well researched and relevant that I make no excuses for reproducing it here:

'… The rather close relationship of the texts to each other suggests derivation from a lost broadside; the traditional divergencies are less in the textual variants than in the sets of the air.

The Fair Maid by the Sea-Shore is the representative in English of a group of Nordic and Romance ballads in which a maid is enticed on board a ship and abducted; the motif is found in Middle Irish popular tales.

The Romance form is in French, Italian and Catalan.  A French ballad L' Epee Liberatrice (V Smith, Romania, VII, 69) is of a Prince's daughter who sees a ship with thirty men on board, coming to land; she goes on board to learn the song sung by the youngest of the crew.  The song puts her to sleep; when she awakes, she is on the high seas.  She asks a sailor for a sword and kills herself.  The Italian ballad Il Corsaro (C Nigra, Canti Popolari del Piemonte, p.106) is much the same, except that it omits the sleep-producing effect of the music.  In Catalan tradition (Lo Mariner, in Briz y Candi, Cansons de la Terra, I, 113-20), the ballad has a happy ending.

The Nordic form, found in Danish, Norwegian and Faroese, has been studied in detail by R Berge, Skipar Holgje, in Norsk Folkecultur, I, pp.71-84.  The best form of it is the Swedish Skon Jungfrun (A I Arwidsson, Svenska Fornsangre, I, no.41, pp.288, ff.)  A maid on the beach sees a ship coming to land and asks the skipper what he has to sell.  He takes her by the snow-white hand and leads her into his cabin, where she passes the night.  When she awakes on the high seas, she laments that she will never see her children again - a taunt for the skipper, implying that she has deceived him - lets down her yellow hair, jumps overboard and swims to her native land.  The skipper pursues in his ship, but is left cursing on the beach, as he sees her sitting in her father's window.  Norwegian B (Berge, I.c., pp.72-3) has the 'happy ending' of the Catalan ballad, found also in the version of The Fair maid on the Sea Shore in the Belden MSS.

Our ballad resembles the Nordic form rather more closely, but approaches the Romance form in the retention of the sleep-producing music.  We have the magic music in the Middle Irish Dindsenchus of Cleena's Wave, Glandore, County Cork, to the effect that Cleena (M. Ir., Clidna) was beguiled by Iuchna Curly-locks, who played sleep-producing music to her in his boat of bronze, from which she was washed overboard and drowned.  (Revue Celtique, XV, 437; Todd Lecture series, X, part 3, pp.206, ff.)

Elsewhere in the Dindsenchas (Rev Celt, XVI, 32), the sleep-producing music is called 'the mermaid's melody': the fact that in our ballad, it is the maid and not the seaman who sings it enables us to trace the ancestry of the ballad to ancient mythology of the ghost-soul.

G Weiker, in Der Seelenvogel in der Antiken Litteratur und Kunst, shows that the Sirens, birds with human heads, originally represented spirits of the dead.  From very ancient time, the ghost soul was believed to desire passionately the love of a living mortal: this belief, through the lore of Lilith, Lamia and similar types of supernatural women, has left innumerable traces in popular tradition.  The magic music, that is, ghost and spirit music, of which Sirens' song and angels' song are but particular types, has been held to possess for mortal ears an attraction so irresistible that the consciousness of time and space might be extinguished by it.  In the eighth century, as we learn from the tract De Monstris et Belluis, (ed. Berger de Xivrey, p.25) Sirens are described as mermaids, women with tails of fishes, 'who deceive sailors by their lovely form and by the charm of their singing'.  We suspect the change is due to Arabic influence, since the Latin synonym of siren, namely puella marina, is in Arabic bint al-bahr, 'maid, (literally daughter) of the sea', a phrase applied by Arab naturalists to the manatee, popularly believed to be half fish, half human.  Through the Irish muirgen, morigain - the latter a gloss to lamia in Isa., XXXIV, 14, Thesaurus Palaeohibernicus, I, 2, - the character of the siren-lamia is fixed for Morgain la Fee, the case type of supernatural mistress, sometimes devoted, sometimes cruel and malicious.

A Greek ballad printed by A Passow, Carmina Popularia Graeciaecentiore Ris, No.524, p.401, has some affinities with our ballad.  A water-woman, who calls herself the Lamia of the Sea Shore and of the Sea wagers that she can dance longer than a young shepherd can pipe for her.  He pipes for three days, then falls asleep: the Lamia robs him of his flocks.'

Whilst all of this is very interesting what we really want to know is the likely immediate inspiration for our ballad, and despite the lack of the sleep-producing music, the Scandinavian ballads - motif for motif - have more in common with our ballad than any of the other possible sources suggested here.  It is also quite possible that our ballad was inspired by some other continental version not yet uncovered, and that the Scandinavian ballads are derived from our ballad.  Versions of Child 20 The Cruel Mother that clearly derive from the English ballad have been found in Denmark.

Since Phillips Barry wrote this article, other earlier and longer versions of the ballad have surfaced.  The two fullest and earliest versions were noted down within a couple of years of each other in the early nineteenth century.  The first is the only extant broadside printing.  It was printed in a single column on a two ballad broadside by W Armstrong of Liverpool c.1820-24 (its companion being the traditional ballad Will the Weaver).  It is not set out in stanzas though I have attempted to split it up into stanzas here using other versions as a guide.  Even so it appears to have an extra half stanza or one of the stanzas has six lines which is a frequent occurrence in older ballads.  It appears to be somewhat garbled from about half way through and could well have come from oral tradition.  The order of the stanzas seems also wrong towards the end, for instance, the last stanza here, in other versions comes further up the order and would seem to fit better there.  It is worth pointing out that at least some of Armstrong's output was obviously taken down from Irish immigrants into Liverpool, although from its style and language there is nothing here to suggest that this ballad is Irish.  There is some evidence to suggest that several well-known Irish ballads that appeared widespread on English broadsides entered this country via Liverpool and were printed by Armstrong and then copied by other English printers.

The second ballad was noted down c.1825 by Thomas Macqueen in Ayrshire adding the note 'I learned some years ago from a Kilburnie man it has a fine tune.'  Macqueen was one of the collectors employed by Andrew Crawfurd, who in turn collected material for William Motherwell.  It was published in Emily Lyle's Andrew Crawfurd's Collection of Ballads and Songs , Volume II (1996), p.88.

Though the ballad was written down in lowland Scots there are no specifically Scottish words used, i.e., those words with Scottish spellings could easily be adaptations of Standard English.  Stanza 4 here does not occur in the broadside but is found in a New York State version given below and partially in a Newfoundland version 1.  Stanzas 8 and 9 here are spliced into one stanza (8) in the broadside and the first three lines of the last stanza are unique to this version.  They could well be an interpolation of the singer, having forgotten the original lines.

It is pretty obvious that neither of these two versions derive from the other, which presents us with the prior existence of a longer version.  Although I normally steer clear of presenting tune evidence as this is not my field of study, it has struck me and others that the tune and metre are quite closely related to Green Broom (Roud 379) which would appear to date from at least the early eighteenth century.  (See Pills to Purge Melancholy, Volume 6, 1719-20, Thomas D'Urfey, p.100)  If indeed the two are related this presents the possibility that one was modeled on the other and possibly their origins are roughly contemporary with each other.

Particularly of note in the Macqueen version is the mention of conjuring in stanza 10.  It certainly backs up Phillips Barry's assertion that there are supernatural elements to the ballad and that singing the crew to sleep is not just a clever wile.  The only other version I have seen that mentions the conjuring aspect is one of the Newfoundland versions collected by Maud Karpeles: 2

The middle of the nineteenth century presents us with three American versions, two of them from North East whalemen.  The earliest of these comes from the log of the Ocean Rover, 1859 (See Gale Huntington, Songs the Whalemen Sang, p.136).  Apart from the reduction in stanzas one would expect of oral tradition over forty years or so, stanza 6 is a splicing of two of the earlier stanzas.  This same spliced stanza also occurs in the two other versions here given but rarely in later versions which have the full stanza (9 in the broadside and 10 in the Macqueen version).  Here is the Ocean Rover version. The next version written down in 1866 comes from the Old Album of William A Larkin (Reproduced in Journal of American Folklore, 60, No.237 (1947) Item 26, p.230).  Larkin started to write down these songs when he was twenty.  His family had come from England, first settling in Pennsylvania then moving to Ohio where Larkin was born in 1845, then on to Illinois where he wrote down his songs.

Differences between this and the preceding version are that stanza 6 is a splicing of two stanzas.  It also presents for the first time the complete final stanza as found in most later versions, with the incremental repetition from the preceding stanza.

A few years later whaleman George W Piper, ship Europa of Edgartown, Massachusetts, wrote down in his journal dated 1868-70, a version which consists of stanzas 1, 2, 4, 7, 8 and 9 of the Larkin version with no significant variation other than it sets out the repeats as a separate chorus as follows: Some twenty or so versions were collected in North America in the twentieth century, along with a couple in Ireland.  I have inspected most of these and no two have the same combination of stanzas.  Most of these have between one and four stanzas consisting of two earlier stanzas compressed into one.  Nearly all of the North American versions were found in the North Eastern states or the Canadian maritime provinces, which brings me to a point worth including here, that Macqueen who collected the c.1825 Scottish version, emigrated to Hull in Canada in 1828 and Hull is not far up the St Lawrence from the North Eastern seaboard.  It is worth reminding readers that this part of the world was heavily settled by Scottish and Irish emigrants.

The longest and arguably most interesting version collected in the twentieth century comes from the Catskill Mountains in New York State 3, about 200 miles as the crow flies from Hull.  It was recorded in the 1940s in two almost identical versions from near neighbours, Elson Van Wagner and George Edwards.  I reproduce here the twelve stanza version from Van Wagner of Clayville.  The eighth stanza is unique to this version, likewise the first half of the final stanza.

There has been some speculation on the song's origins by the likes of Alan Lomax and Bertrand Bronson in an attempt to link the ballad to Child 43 The Broomfield Hill.  Admittedly both ballads bear the motif of the heroine tricking her male would-be seducer by sending him to sleep, but there is no mention of singing in Child 43.  The two ballads also share a commonplace in that when the seducer awakens he blames others for allowing it to happen.  However, in Child 43 he blames his horse, his hound and his hawks and, rarely, his serving man, whereas in Maid on the Shore he blames his crew.  Similar stanzas appear in many ballads, more usually when a powerful figure has killed someone he loves or admires in a fit of passion, after which he blames his men in similar wording for not staying his hand.  Bronson 4 includes half a dozen versions of our ballad in an appendix to Child 43 which 'have seemingly little to do with any of the rest and go to a curious marine adaptation of the text.'  To my mind the connection between the two ballads is very tenuous, but to be fair his expertise was in tune variation, of which I have little knowledge.

Others 5 have sought to link the ballad to Child 4 Lady Isabel and the Elf Knight.  In an Ozark version of this the seducer is charmed asleep, but again there is no mention of singing, and in the same version the heroine uses his sword to dispatch him.  Again the link is very tenuous and it would be appropriate to say this particular version of Child 4 probably borrowed the charm idea from Child 43.

Anyone interested in tune relatives of the ballad should seek out a copy of that marvellous book Folk Songs of the Catskills, pp.276-280 3, although it makes no mention of Green Brooms which I think may be related, but then I'm no expert.

Further Research:

Given the fact that Phillips Barry's findings obviously have substance, the bulk of the English story follows closely that of Scandinavian ballads, but he did not have access to versions that included the 'singing to sleep' motif found in other continental versions.  He therefore couldn't present us with a direct source for the English version.  It is my belief that this must exist, i.e., it is highly unlikely that the author went to two separate sources to compose his version, though it is possible: For instance poets like Herder, Schiller and Goethe were conversant with Scandinavian ballads and translated much material from other European languages.  It is possible that one of them or one of their contemporaries compounded a German version which inspired our ballad.  Hey, I'm now getting well out of my depth!  What I'm suggesting here then is someone with the necessary skills (obviously not me) check out other European versions to see if something closer to our English ballad can be found.  Pure guesswork, but it seems likely that Scandinavian versions once existed that do contain the 'singing to sleep' motif.

Versions Consulted:

Those preceded by numbers in brackets refer to numbers in the text.

Dungbeetle - 6.12.10

Site designed and maintained by Musical Traditions Web Services  Updated: 8.12.10