TY Steve for the tunes. Ty everyone for your insights and help with this tread.

I've added the Brown Collection of NC Folklore and Abrams collection to my site 911 versions) and have stated adding more US versions here: http://www.bluegrassmessengers.com/us--canada-versions-1-gosport-tragedy.aspx

A few observations are found on the main page (headnotes) here: http://www.bluegrassmessengers.com/1-the-cruel-ships-carpenter-pretty-polly.aspx

The thrust of these observation are posted below (preliminary findings which may change) since they aren't so long. Comments are welcome:

Unless you agree with Fowler[5] that the first Gosport broadside was written around 1726 by a printer[6] from a story about a murder and ghostly visitation witnessed and transmitted by Charles Stewart (also Stuart), the broadsides could likely be based on an ur-ballad[7]. I find it unlikely that Stewart would have met Molly's parents and told them where she is buried and it seems unlikely that he witnessed her ghost and the other supernatural events that take place in the broadside and transmitted them to Cluer in London[8]. Aa certainly does not seem to be taken from tradition, but rather is a printer's elaboration on the ballad story. Bb, is shorter and closer to the likely tradition while Ca has the standard text with a new beginning and ending. Discovering the ur-ballad from the these printed broadsides can be done by studying available traditional versions that are not closely based on print[9].

The nature of the Ur-ballad raises these questions. Was the first print version, Aa London, circa 1720-1750, a wholly original work? Did other printers copy the first printing and subsequent folk singers learned the ballad directly or indirectly from these broadsides? My contention is that Aa was taken from tradition (the ur-ballad) and embellished. The embellishments could include additional stanzas and also the setting and names. As much as Fowler[10] has tried to prove the historical accuracy of the ballad as if it was based on an actual event, I assume that it was more likely that the specific details, which include William, Molly, Charles Stuart, Gosport, Plymouth (Portsmouth) and the HMS Bedford, were added to the ur-ballad by a printer. The ur-ballad does seem to be about a ship carpenter who charms and impregnates his lover while ashore and then before he sets sail, decides to murder her because she is pregnant. This murder would, according to tradition, have placed a curse on the vessel and its crew if the murderer went to sea. The appearance of his lover's ghost and his subsequent confession and death are predicable results of his crime. William dies because he is "raving distracted" which is another clue to understanding the ballad's oral history.

Fortunately there is a body of traditional versions that may be used to understand the underlying traditional ballad. One area of study is the Appalachians. Cecil Sharp and Olive Campbell collected 39 version alone of "The Cruel Ship's Carpenter" between 1916-1918, most of which date back to the 1800s by the informants and through their families date back to the times of their early ancestors settlements in the mountains. Although it may be impossible to pinpoint the exact date this ballad came into an area, like Madison County, NC, it may be assumed that the ballad was brought to the Virginia Colony and was taken by ancestors into the remote mountains, where it remained to be discovered by Sharp between 1916 and 1918. Many of the ballads brought to the Virginia colony were learned there before the Revolutionary War and taken into remote regions such as Flag Pond, TN and Madison County NC before 1800.

Most of the texts are the short Appalachian version which is similarly found in 6 stanzas of the broadside. It's clear that these stanzas are part of the ur-ballad. Only two texts, Sharp B and an MS fragment, are closely related to the Gosport broadsides. The remainder of the texts (and other Appalachian text) when compared show these results:

1. That first stanza of the broadsides is found enough times to assume that it is part of the ur-ballad. In Appalachia the damsel dwells in London and other locations and her beauty is mentioned and sometimes that her lover was a ship's carpenter.

2. That stanza two is usually present though not necessarily as the second stanza in that place. William wants to marry her and asks her to marry.

3. That her response, "I'm too young to marry" is present enough to warrant inclusion in the ur-ballad. Although it is rare, this is found in enough versions to believe it has been transmitted from the British Isles and is not a local variance.

4. That subsequent stanzas in the broadsides (4-13 Roxburghe) that include her sexual submission, her pregnancy, and his call to return to the sea are lacking in Appalachia except in very rare cases. It may be assumed that this may be in part due to taboos of transmission of sexual actions in general.

5. The visitation of Charles Stewart is missing in standard Appalachian version however, Polly's (Molly's) ghost does appear and William does die "raving distracted."

6. That additional stanzas constructed in Appalachia probably from the House Carpenter, where William boards a ship after the murder which sinks can be expected. The boarding of the ship is still in line with the broadside, it's the sinking and going to hell to pay a debt to the devil which is new.

The possibility that the early broadsides captured an existing traditional ballad (ur-ballad) is reasonable. It's also reasonable to assume that the Deming broadside and the "Polly's love" broadside were changed by the addition or subtraction of either new or pre-broadside traditional texts. I also know that some of the texts made by these printers entered tradition and this is proven by collected versions. My conclusion is that the ballad was reworked by a printer, or in this case printers, from a traditional tale about a murder and cosmic retribution upon the murderer for his horrific deed.

Richie