Article MT164, part 26 - newly written for MT.

Robin-a-thrush : Roud 2792

A study of a fascinating song and its relationship with 'The Wife Wrapt in Wether's Skin',
Child 277, Roud 117, and 'Kempy Kay', Child 33, Roud 32.

I have used the most common title found in published collections of folk songs for this article, and indeed my index of Master Titles, but on street literature it went by various names, 'Thrifty Housewife', 'The Tidy One' and 'The Tidy Hussey'.  It is of interest not just for its longevity, but also for its association with 2 Child Ballads.

This short chronology of 'Robin-a-thrush' spanning more than three centuries cannot pretend to be anything like comprehensive.  The evidence we have suggests many intermediate print and oral versions most of which have not survived, but the beauty of such an online article as this is that any new information can be added as it comes to light.  Indeed it has already been revised three times while I have been writing it for that reason.  Some of the print versions show evidence of both oral tradition and relatively sophisticated rewriting by hacks.

Whilst the earliest definitely related piece is dated 1679 the even earlier piece given below could have had some influence on the style of later pieces in the eighteenth century.

This simple song has some of the hallmarks of folksong, and although it has little in common with the text of 'Robin-a-thrush' it has some phrases in common and stanzas 1, 5 and 7 bear comparison.

Composed by William Hickes the next piece was a lengthy poem in couplets comprising progressively more gross descriptions of this slattern housewife and her domestic relations.  For instance some of it describes the mixing of excrement with food.  I had not the time to copy the full piece but the following extract contains in lines 7 to 12 enough to demonstrate its links to 'Robin-a-thrush'.

Unfortunately I have not yet been able to discover any eighteenth century versions but without doubt these once must have existed.

The next extant link in the chain appears to be the following slip song, probably printed by Kendrew of York c.1800.  It is printed without stanzaic divisions but looking at other versions it is easily divided into quatrains.  It appears that the 6th stanza lacks the first line, which could easily be supplied from other versions.  The first 9 stanzas describe the housewife's slovenly habits, but the following three take a different turn in describing her gross features.  As these have little influence on later versions and have other interesting connections I will deal with them in Appendix 2.  The final two stanzas are unique to this version.

A shortened version of the above was printed by T. Birt of 39 Great St. Andrew-street, 7 Dials, not long after and this certainly shows evidence of having come from oral tradition as opposed to a straight hack rewrite.  The order of stanzas is different and the altered phrasing is consistent with a piece that has been in oral tradition a short while, say a couple of years or so.  One would expect the order of stanzas to vary from version to version as the song is a description with little order to it. Another identical copy of the above is online at The Bodleian Broadside Ballads website, Firth c20 (55)

Yet another version quite different to both of these was printed about the same time by Croom of Sheffield demonstrating that there must have been at least one other version that precedes all 3.  Stanza 2, which doesn't occur on the other two slips, is almost word for word lines 8 and 9 in the 1679 version.

In some ways this version appears closer to the 1679 version than the other two.  Verse 4 here also appears in the Renfrewshire oral version (given below).

The Bodleian Broadside Ballads website attributes both this song and the following rewrite to Charles Dibdin the elder, but in all of his works I have seen I have not come across them.  It is highly unlikely he wrote both of them, but he may have had a hand in the rewrite which is more his style being aimed at a much more upmarket audience.  The version here was printed by John Pitts at his earlier address,14 Great St.  Andrew Street, 7 Dials, where he was active upto 1818.

From the date of this printing we know then that the more basic 'Tidy One' must predate 1819.  This rewrite was printed in 1843 in various published songsters by Turner and Fisher in America (Roud Collection).  The same Pitts broadside is also in Nottingham University Library, PR1181.  B2.  Special Collections o/size, 1:72; The Madden Collection, Cambridge University Library, London Printers 2 [VWML microfilm 75, item 1288], and The Bodleian Broadside Ballads website, Harding B16 (288c).  An identical copy printed by Croom of Sheffield under the title 'And wasn't she A Tidy One' is on The Bodleian website, Harding B28 (245).  The Thomas Hardy Mss, copy in the VWML acc.  No.  5173, item 171, has the tune only.

Whilst we have no versions definitely identified as belonging to the eighteenth century the great variety amongst the earliest versions of the nineteenth century attests the great popularity the song must have had during the eighteenth.  Of the 27 stock stanzas identified in all extant versions, printed and oral, only 16 of these existed on the extant slip songs, but by c.1830 all 27 were in use.  This type of song is notorious for having a wide variety of stanzas and it readily lends itself to the addition of new stanzas and the adaptation of existing ones.  Other songs of this type are 'Brian o' Lynn', 'The Old Man from Lee', 'The Old Grey Mare' and a whole host of bawdy songs and children's songs.

I now move on to oral versions.  If we exclude the 5 stanzas tagged onto the end of 'Thrifty Housewife' which occur in few oral versions of this song, the similar 4th stanza of the longer Campbell Mss version (see Appendix 2), and the commonplace 'Cheese on the shelf, sing it yourself', we are looking at a total stock of 21 stanzas.  Just 11 of these are found in the slip songs comprising the first 9 stanzas of 'Thrifty Housewife' and the second and fourth stanzas in 'The Tidy Hussey'.  The other 10 stanzas are the last 3 stanzas of Sharp's 1909 Berkshire version; the second stanza of the English County Songs version; the last stanza of the London glee version; stanzas 2 and 7 of 'Willie went to Westerdale; the last stanza of the Renfrewshire version, and the last 2 in the Chambers Scottish version, all given in full below.

According to the great song historian, Frank Kidson, the song was sung by the theatrical clown Joseph Grimaldi in the early years of the nineteenth century and it may well have been this glee version that he sang. In all of the oral versions I have studied (listed at the end) the core stock of stanzas amounting to 13, i.e., those occurring 6 or more times, consists of all 9 stanzas of the Berkshire version above and stanzas 2, 3, 5 and 6 of the Renfrewshire version.  Basically the Berkshire version is arguably the most archetypal of all the oral versions, containing 9 of the 13 core stock stanzas.

Of those stanzas in the slip song 'Thrifty Housewife', apart from the last 5 to be dealt with later, stanzas 2 and 8 are not found in later oral tradition though they may have influenced stanzas 2 and 3 in the shorter version below collected by Hammond, which are unique to that version.

The last stanza of 'Willie went to Westerdale' is unique to it though there are similar stanzas in the 1679 version.
1679: 'And sent her husband for to fetch him a cap, But before it came he spewed up all in lap.'
Willie: 'She did a far dirtier trick than that, She let bairn wet in his neet cap.'
The last 2 stanzas of Chambers's 1829 version are unique to it but have similarities of sentiment with the last stanza in 'Thrifty Housewife'.

The second stanza in the Suffolk version given in Broadwood and Maitland's English County Songs, 1893, p92 is also found in Halliwell's 'The Nursery Rhymes of England' version and is the last stanza in the longer of the two Campell Ms versions.  It is a testament to the far-travelled and convoluted nature of the song that this scarce stanza is found in versions from opposite ends of the country. P> All of this confusing mixture is perhaps to be expected from a catalogue song with little or no narrative that has been in oral tradition for several centuries.  There is a strong likelihood that oral versions and longer print versions existed throughout the eighteenth century.  The strongest evidence for this is the existence of stanzas in the early oral versions which are not in the slip songs, the geographically widespread nature of the stanza distribution, and eight other stanzas that are common in oral versions that are not found in the extant slip songs.

On the face of it it could be argued that the London printed glee version of c.1815 was the origin of most of the southern English versions, but it has a different final stanza to all of the other comparable versions and these contain two other stanzas one of which is very common.  What is much more likely then, is that the glee was adapted from oral tradition.  More on this when we discuss the chorus in Appendix 1. 

A brief look at the geographical distribution of versions is worthwhile at this point even if it only shows how diverse and widespread the song is.  The longest printed version comes from York, with the two shorter slip versions from Sheffield and London.  This does not tell us a great deal, nor do I think it particularly significant as any of those three could also have been printed elsewhere without being currently extant.  However when we turn to oral versions a pattern of quite distinct regional types emerges.  Of these regional types the southern English type seems to have been the most stable, in the length, stanza order and consistency of stanzas, typified by Sharp's Berkshire version.  The most notable departure from this is Baring Gould's Devon version below which is much closer to Scottish versions.

Whilst this version has several second lines which are unique to it, each stanza is recognisable as belonging to the general stock.  The last to lines are unique to it and very likely unique to the singer.

Scottish versions tend to have much more fluidity.  Even the longer earlier versions such as Campbell Mss, Chambers and Crawfurd are widely varied in the combination of stanzas used and in the order of stanzas.  Later Scottish versions are very brief, the longest of five stanzas being one of the versions in the Greig-Duncan Collection.

When we get over to America, versions are well spread out.  There is some affinity between Southern versions utilising the same 6 stanzas, but there are even more versions that use a wide variety of stanzas from the general stock.  All of the stanzas in American versions can be traced back to British variants though some British stanzas have been extended to make two stanzas as in stanzas 2 and 3 in the following version given in Randolph's Ozark Folksongs.

Whilst the last stanza is close to the last stanza in the English archetype it is too much of a commonplace in this type of song to say that it derived from the English version.

The are two survivals from northern England 'Willie went to Westerdale' and the following fragment from north east England c.1800-1830

The chorus obviously harks back to the slip songs but this verse, whilst it is common in oral tradition, is not found in this form in the slip songs.

'Willie went to Westerdale' in many ways is much closer to Scottish versions and there is some evidence to suggest worker migration from Scotland to the Esk Valley to work the mineral mines found there.  Other Scottish songs have long been popular in the area, such as 'The Bonny Hawthorn that Blooms in the Vale'.  The affinity of the chorus to some Scottish and American versions will be dealt with in Appendix 1.

Before moving on to the choruses it needs mentioning that the song has twice been appended with a series of stanzas which to some degree in both cases do not follow the main theme of the song.  It is very likely that both sets of appended stanzas were grafted on some time in the eighteenth century, probably towards the end of that century, and these could easily have occurred at about the same time.  The sole theme of the original song is the slattern behaviour of the housewife, and though some of the descriptions are pretty gross they are by and large quite plausible.  The making of the bed by the bugs and the cheese turning by itself should be taken simply as sarcasm and not literally.  The whole thing is a joke typical of the jests of the period and indeed the song is typical of a large genre of such songs, nowadays more likely to be found in collections of bawdry and in milder forms in collections of children's songs.

One group of these appended stanzas is best shown in the slip song 'Thrifty Housewife' stanzas 10 to 12, which also survive in a few scarce oral versions, indeed along with a fourth stanza in the same vein.  These stanzas are actually a description of the wife's physical features and they are far from humorous although still gross.  As they relate to another Child Ballad they are dealt with in more detail in Appendix 2.

Quite the opposite is the group of stanzas appended to the common English oral version, typified by the three stanzas at the end of the Berkshire version.  These stanzas give the song a quite different turn, giving it a jolly nonsense twist in the tail.  We can skip over the last of these as a commonplace found in many other songs of this genre particularly in America, but the other two extend the idea of the cheese turning by itself, having it sprout legs and run through the door, and then to Charing Cross followed by the lady on a white horse.  The first of these may have been suggested by the old wives's saying that a particularly mouldy cheese is 'ready to walk off the table'; and the second by the children's rhyme 'Ride a Cock Horse to Banbury Cross', indeed several versions name Banbury Cross and in one the lady rides a cock horse.  However the idea of an inanimate object suddenly sprouting legs and running off, followed by its adventures, is common in rhyme, song and folktale,, 'The Gingerbread Man' and 'The Old Lady of 92 who blew a Fart'.

Appendix 1: Robin-a-thrush (Roud 2792) and The Wife Wrapt in Wether's Skin (Child 277, Roud 117)

The first and important point to note is that although the subject matter in these two songs is similar, they have no text in common worth noting other than the occasional sharing of a first line, and indeed appear to have totally independent origins.  No versions of Robin-a-thrush have any mention of ploughing or sheepskin; indeed the only mention of thrashing occurs in one of the shared choruses.  In no version of Child 277 is the wife a slattern, she is merely lazy and/or contemptuous of doing menial tasks.  Both songs are quite long in their fuller versions but at no point have any stanzas from one strayed into the other apart from the first line in a few American versions of Child 277, despite sharing two choruses.  It is a fact that both songs have early versions that share the common name Robin and there are often similarities in the wording of the first stanza in that the male narrator, or Robin, is bringing home a new wife.  Both songs vary between using first and third person narrator; earliest versions of 'Robin-a-thrush' use first person and earliest versions of Child 277 use third person (Robin).  However, early oral versions of 'Robin-a-thrush' also have Robin and this could possibly have contributed to the sharing of choruses and some of the confusion.

Both songs seemingly started out with nothing at all in common.  Indeed it would appear that what they later shared, be it tunes or choruses, were not part at all of the earliest versions of both songs.  The A and B texts of Child 277 could well have derived from printed copies by Johnstone of Falkirk, or all of these could derive from an earlier version.  The Falkirk copy is slightly longer than the oral versions.  The refrains in the slip versions of 'Robin-a-thrush' and the two early versions of Child 277 do not appear to have influenced the choruses of the two songs that they have in common in oral versions, with the possible exception of the name 'Robin'.

British Library edition: 11606.aa.22(45) 3 also printed by Johnstone.  The BL version employs capitals on some nouns which is generally an earlier custom.

Various sites give Johnstone's dates as c.1810-1820.  Despite the Child A and B versions predating this, indications are that the printed chorus is the earlier.  The first refrain in the printed version 'Holland's green, Holland' could well be a street cry from a fine cotton seller: Holland handkerchieves and shirts are common in balladry.  This appears to be corrupted to 'Hollin, green hollin' in Child A and B but holly is indeed green so either could take precedent.  The second refrain 'Benty bows Robin' simply translates as 'bandy-legged Robin' but in the oral versions it becomes 'Bend your bow(s), Robin'.  Again it is not clear which takes precedent chronologically but the printed version seems more reasonable as a phrase to me.

Before moving on to the similarlities in the choruses it should be noted that much extra knowledge would likely be gained by studying the extant tunes of both songs, particularly those versions that share choruses.  Bertrand Bronson in his 'Traditional Tunes of the Child Ballads' has made an excellent start on this in comparing 56 versions of Child 277 and 6 versions of 'Robin-a-thrush'.  To give an idea of the extent of variation in tune types he identifies 5 different groups, and the 'Robin-a-thrush' group makes that 6.  Broadly speaking these different groups of tunes correspond to different choruses as well, so in grouping them into tune groups he has also effectively grouped them by chorus.  Unfortunately I am not qualified to comment on the tunes here.

As the main point of this study is to compare and distinguish the two songs I will only mention briefly those of Bronson's groups that have no bearing on the comparison.  Group A consists of the two tunes from the Child 277A and B type.  Group B is the 'Wee Cooper of Fife' versions corresponding with Child 277 C and D.  In Scotland this highly popular variant seems to have quickly displaced the Group A types at an early stage.  Group C types, only found in America, have adopted the 'Jennifer, gentle and rosemary/as the dew flies over the mulberry tree' chorus widely used in other Child Ballads.  We can dismiss these as having no bearing on our current comparison; Group D likewise, having the chorus 'the draggle draggle daughter of Peggy-o' found only in a few versions in Somerset.  Group E is particularly fascinating, mainly found in America, but with some curious isolated versions from England and Scotland.  It is those with the 'dandoo' chorus, not dissimilar to the 'Kemo, kimo' variants of that ancient song 'The Frog and the Mouse', which also employs a wide variety of tunes and choruses.

Before proceeding further I feel the need to state that most of the following is based on supposition, possibility and probability, but even if it only provokes debate and further study I will have succeeded.  With such a widespread, varied and long-lived pair of songs it is unwise and misleading to suggest otherwise.  In addition, trying to pin down lines of descent in largely nonsense choruses is notoriously difficult and fraught with dangers, one of which is that seemingly meaningful phrases becoming nonsense in the oral process should not be taken for granted.  The reverse process where nonsense phrases are rationalised into something meaningful can be just as common.

Child's C and D versions of 277 are the earliest versions of this ballad to utilise Bronson's Group B 'Wee Cooper of Fife' chorus.  The D version sent in a letter by Robert Eden Scott, Mrs Brown of Falkland's nephew, to Robert Jamieson in 1805 is the earliest extant version of the chorus.

Jamieson-Brown MS, Appendix p.iii, letter of R. Scott to Jamieson, June 9, 1805.  Stanza 1 of 8.

Of particular relevance here are the 'now, now, now' repeats, the reference to a 'petticoat' and 'Robin a Rashes'.

The C version, first printed in Whitelaw's 'Book of Scottish Songs' in 1844, would appear to derive from the D version above or some similar version, but this is only a probability as chronological publishing detail proves nothing as we shall see.

The second lines of the two versions show enough resemblance to be closely related, likewise the mention of personal names in the fourth lines.  It needs little leap of faith to assume that 'Rushety' in C is either derived from 'Rashes' or both from the same source.  I assume that Child is repeating the capitals on the two words from the manuscript and book which would tend to suggest that the writer/publisher assumed they were proper names, particularly in the latter where Rushety is actually speaking ('quo').  This would be the limit of our knowledge and supposition were it not for further oral versions of both songs widespread in the English-speaking world that present further possibilities.  Before we move on to these, the great variety found in the use of tunes and choruses in both songs, suggests that this process was happening quite early, at least the eighteenth century, in those versions where the group B choruses are present.

At this point I take a leap of faith and suggest that the D version sent to Jamieson in 1805 could be the hybrid that has led to at least some of the chorus crossovers.  It is plain that stanzas 1 and 3 to 8 all belong solidly to Child 277, but the second stanza below and the chorus actually belong to 'Robin-a-thrush'.

It is also possible that the first stanza given earlier, which was repeated in all of the later 'Wee Cooper' versions, was made up by someone in the Brown household in order to localise the song.  Robert Scott obtained the song with two other pieces from an old maidservant in the family.  The idea that this chorus was from 'Robin-a-thrush' was proposed as early as 1906 by Anne Gilchrist in Journal of the Folk-Song Society No.  9, the 4th part of Vol II, p224, although it's unlikely she had access to the Crawfurd version (above) so didn't recognise that the second stanza was also taken from 'Robin-a-thrush'.  This stanza appears in no other version of Child 277 either earlier or later. The chorus of the Crawfurd Renfrewshire version of 'Robin-a-thrush' appears to be somewhere in between Child's D version and C version and is so chronologically (c.1830).  The 'neigletie, neigletie' has taken the short hop to 'nickety, nackety' and the 'Heich, wullie, williecoat' has become 'Hey, Willie Wallacky'.  Although the 'Wee Cooper' has lost his 'petticoat' completely the vestige of 'coat' is still there in the intermediate version.

Yet another Scottish oral version , that published by Chambers from recitation in 1829, has a chorus that appears to be a corruption of the Renfrewshire version above.

Whilst it's mostly nonsense the word 'ben' seems to be derived from 'bang' in the other version, and of course 'willie wallets' is 'wullie willicoat', later 'Willie Wallacky' in 'The Wee Cooper'.  The 'niddle noddle' also corresponds to 'neagletie, neagletie' and later 'nickety, nackety'; the 'noo, noo, noo' being common to all.

Most of the oral versions, 27 out of 36, of 'Robin-a-thrush' have Bronson's group B chorus.  Looking at the last line of the chorus 3 English versions are particularly interesting in that they give much greater meaning to this line.

Whilst it is possible that these versions are the result of rationalising nonsense lines it is very tempting to come to the conclusion that something like this was the predecessor of all the extant versions that use this chorus.  If this was the case it would give some weight to the possibility that this chorus was part of Child 277 before it became attached to 'Robin-a-thrush', in that thrashing of the wife is part of the former whereas no version of 'Robin-a-thrush' mentions thrashing other than in the chorus.  In other words it may well be that 'Robin-a-thrush' has preserved an earlier chorus of Child 277.  This place is as good as any to point out that all of the Group B versions in Bronson are clearly derived from Child C, Whitelaw's extremely popular 1844 version of Child 277.

Another feature of the chorus that is common only to the Sumner and Hammond versions mentioned above is the inclusion of 'hey down, ho down'.

Similar refrains are found in seventeeth century ballads like The Keeper and Robin Hood ballads.  Of course it may have been altered simply for this reason for effect, but taken in conjunction with the last line it adds weight to the possibility that this is an earlier form.

Another interesting feature of the English 'Robin-a-thrush' choruses is the remarkable persistence of the mention of a petticoat, later lost in the 'Wee Cooper' forms of Child 277.  Almost all of the English versions have it albeit described in a different way in each one.  In addition to the two above we have:

'Tops and petticoats' (English County Songs)
'Ruffety petticoat' (Sharp and George Withers, Somerset)
'Shocking green petticoat (My Song is my Own)
'Rough guff petticoat, guff stuff petticoat' (Broadwood Mss)
'Stuffajum petticoat' (Glee)
'Ruppety petticoat' (Jeff Wesley)

It may be worth noting the description 'green' in three of these versions.  It is tempting to compare this with the 'green gown' or the unlucky meaning of green in other ballads, but this is more likely a red herring.  It is even tempting to connect its use to that in the Child 277, 'Holland's green, Holland' but again this is unlikely.

Looking for explanations for the origins of the personal names in Child C 'The Wee Cooper' there are a few observations to be made.  'Willy Wallacky' is in the Crawford Renfrewshire version 'wullie willicoat' and as we have seen this is likely a corruption of 'petticoat in the English versions.  'John Dougall' in the Crawford version is 'John Douglas', and 'jingle John Doo' in the shorter Campbell Mss version.  They may be attempts to localise or simply to rationalise nonsense phrases.

Incredibly American versions of 'Robin-a-thrush' appear to actually derive choruses from the 'Wee Cooper' versions of Child 277.  This chorus of the following North Carolina version is typical.

I now move on to Bronson's Group E, the 'Dandoo' types where there has also been some crossover between the two songs.  Whereas it is fairly common for two or more songs to share the same chorus and tunes, I must confess to being baffled as to how two separate songs have come to share two separate choruses, and would welcome explanations as to how this could have come about.  The mere fact that the songs have similar subject matter is insufficient explanation for me.  This Group E chorus shows considerably more variation between versions than the Group B choruses, particularly in Britain where it is very rare.  All but one of the Child 277 versions of Group E come from America, indeed they are very widespread there being found in at least 14 states.  I have chosen 5 versions that demonstrate closest links with the 2 British versions of 'Robin-a-thrush'.  These are numbers 37, 46, 47, 49 and 52 from Bronson, Volume 4. Now here is the only extant British version of Child 277 that uses the E Group chorus. Again with the last version we have the anomaly of subject matter from one song (dairy produce) being mentioned in a chorus of the other.

There are only two 'Robin-a-thrush' versions with Group E chorus found in Britain, although one of them, 'Willie went to Westerdale' has been recorded in at least four slightly differing versions.  However these all come from one small area within a few miles of Whitby, Yorkshire, and they are all close enough in text and chorus to be considered as one version for the purposes of this article.

Here is a variant from my own collecting:

The second and fourth lines only vary from one variant to another in the phonetic spelling, but variants of the last line are: It may not be significant but as I have been singing the song myself for almost 50 years I have always assumed, given the nature of the lyrics, the last line in standard English was 'Take her among you' but I never got as far as asking source singers what they thought of it.

Comparison of this chorus with American versions of Child 277 Group E are pretty easy to spot in all three lines.  Almost all of the American versions have 'dandoo' in the second line and here we have 'do a dandy'.  Bronson's 52 version has for fourth line 'Clish ma clash ma clingo' which compares well with our 'Clish clash ma clandy'.  The last line is not so easy to compare but Bronson's 47 version has 'clarmango'.

The only other British version is a fragment given in the Journal of the Folk Dance and Song Society No.  40, 1937, p120.  It was sung to Anne Gilchrist by a Scots friend in about 1912.

If anything it is even closer to American versions of Child 277.  There is no great leap from 'dhu and dhu' to 'dandoo' and indeed in Bronson's 47 it is even spelt 'dan dhu'.  Its last line is very close to the fourth line in almost any of the American versions. 

Where does this take us in evolutionary terms? The Esk Valley, where 'Willie went to Westerdale' is rigidly localised (Westerdale lies at the head of the valley to the west), shows evidence of plenty of Scottish influence.  Scottish songs are certainly embedded in the local repertoire, 'The Bonny Hawthorn' being one example that springs to mind.  In fact most of the source singers in the area who sang 'Willie went to Westerdale' also sang it.  If we assume that this chorus and tune were set to both Child 277 and 'Robin-a-thrush' some time in the nineteenth century in Scotland that gives us a neat line of transmission being taken to America by emigrants in the nineteenth century.  It has been found in England in use for both songs albeit rarely.  However there are too few versions extant in Britain to come to any hard and fast conclusions.  Here is an enigma to be solved by someone with much greater analytical powers than I possess.

Before leaving the chorus analysis I should mention that the three versions of 'Robin-a-thrush' in Volume 7 of the Greig-Duncan Collection at p.56 all go to the same chorus which is different again to all of the choruses dealt with above.  Whilst they are brief some of the stanzas are not found in earlier Scottish versions but belong to the general stock.  This only adds to the amazing geographical dispersion of the general stock stanzas.

Appendix 2: Robin-a-thrush (Roud 2792) and Kempy Kay (Child 33, Roud 32)

In reading the slip ballad 'The Thrifty Housewife' presented near the beginning of this article scholars familiar with Child Ballads cannot fail to have noticed the familiar extra 5 stanzas tagged onto the end which are not normally attached to versions of 'Robin-a-thrush'.  The first of these is repeated in Birt's printing titled 'The Tidy One'.  The last two of this sequence are unique to 'The Thrifty Housewife' and have no bearing on the comparison being made here so I just give the three relevant stanzas.

It is tempting to suggest that the Birt printing has come from oral tradition, but hacks were used to mixing and matching to make new versions by butchering existing ones, even those recently printed.  When you're scrabbling for a living at the bottom of the strata of poesy recycling is a lot easier than original creation.  The printers weren't too bothered either about the niceties of plagiarism and the 'poet' got the same shilling whether his piece of scribble was original or not; it was purely a commercial process, even with some of the nobler pieces.  Would it sell on the streets was the prime driving force, perhaps the equivalent of the twentieth century single releases, of which the most popular were immediately covered by other artistes.  Even in today's litigious rampant commercialism any popular piece is immediately copied, imitated and/or parodied.

Had the three stanzas in 'Thrifty Housewife' only been associated with these two slip versions, and had not found their way into oral tradition with this song, we perhaps would have not found it too remarkable that they also happen to be part of yet another Child Ballad; even perhaps have come to the conclusion that they must be derived from the Child Ballad.  However two of the three, and a fourth stanza in the same vein, form stanzas 2, 3 and 4 of the longer Campbell Mss version given earlier.

As stated earlier once a theme has been set up in a song of this sort communal composition can easily come into effect with fresh stanzas added in a twinkling so it comes as no surprise to see a fresh stanza in there.

So far we have versions using this theme in England and Scotland.  However an unusual version was collected in Alabama with parts of two of these stanzas included.  It has a Group B chorus and I give it in full here as it is different to all of the other American versions and in several ways closer to the earlier Scottish versions.

In this last stanza we have two of the earlier stanzas shunted together.  The song was sung to Arnold by Janie Barnard Couch of Guntersville, Ala.  In 1938, who learnt it from her mother who came from Buena Vista, Georgia.  The song had been passed down in the female line for several generations.

In his headnotes to Child 33 (Volume 1, p.301) Child states 'All these versions (7) of "Kempy Kay" are known, or may be presumed, to have been taken down within the first 3 decades of this century (nineteenth).  A is traced as many years back into the last.' He also refers to the Campbell Mss version of 'Robin-a-thrush' quoted above and in his footnotes quotes the 3 stanzas given above.  Here he also mentions the other Campbell version 'The Queen of Sluts' and the Chambers version.  In the headnotes he states 'Some verses from this ballad have been adopted into one form of a still more unpleasant piece in the Campbell collection concerning a wife who was "the queen of all sluts".' However, in the very last section of Additions and Corrections in ESPB Volume 5, p.289, when he was wearying of his task and having doubts about the veracity of many of the ballads, he stated 'I have serious doubts whether this offensive ballad has not been made too important, whether not-withstanding the points at p.301 (Volume 1), it is anything more than a variety of The Queen of Sluts'.(In other words our song, 'Robin-a-thrush'.)

For comparison purposes I give below lines from 'Robin-a-thrush' that correspond in some degree with lines from 'Kempy Kay' given alongside.  Note that in 'Kempy Kay' the description is in some versions being applied to Kempy Kay, in others to his bride and in some cases to both.

Robin-a-thrush    Kempy Kay
She has two teeth in her headHis teeth they were
Like two harrow tineslike tether sticks
(Thrifty Housewife, st 10)(Child 33, A7)
She has a nose and faceShe had a neis upon her face
Like a bugle hornWas like an auld pat-fit
(Thrifty Housewife, st 11)(Child 33, B8)
And the slaver it hangs down sirThe slaver that hang between their twa gobs
As long as my armWad hae tethered a ten year auld bill (=bull)
(Thrifty Housewife, st 11)(Child 33, B12)
She has eyes in her headShe had twa een intil her head
Like rotten plumbswar like twa rotten plums
(Thrifty Housewife, st 12)(Child 33, B9)
She is covered with scabsAnd aye her hand was at her neck
As big as my thumbAnd riving up the scabs
(Thrifty Housewife, st 12)(Child 33, C7)
The hair in her headIlka hair intil her head
Was like heather crowsWas like a heather cowe
The louses were in'tAnd ilka louse anunder it
Thick as linseed bowsWas like a bruckit ewe.
(Campbell 'What a bad luck had I' st 4)(Child 33, A9)

Most of the Child 33 lines quoted above are present in some form in most of the versions given in Child.

Despite Child's last comment it is very difficult to discern in which direction these stanzas moved, there is so little evidence from the eighteenth century when these ballads were being formed.  All we can really state with any confidence is that one song gave material to the other or that they both derive from something earlier..

Two other versions of 'Kempy Kay' surfaced shortly after Child died and these are in the Greig-Duncan Collection, Volume 7 at p483.  One is an adaptation by William Walker of Peter Buchan's concoction (Child G) and the other is from the prolific ballad reciter Bell Robertson who actually knew Jamie Rankin, Peter Buchan's itinerant blind fall-guy.

Any ballad collected in Scotland in the early nineteenth century was more likely to have been decades old rather than centuries and to have passed through more sophisticated hands than unsophisticated ones.  Many scholars have hinted at this in recent years but few have been bold enough to openly subscribe to this thesis.

Stock verses for Robin-a-thrush

Sources and versions consulted.