It’s always interesting to find a roughly contemporary version of a classic dance that is recognizably the same dance…but not quite the same.

I generally use Thomas Wilson’s version of “Sir Roger de Coverley” as my default version for the Regency era, mostly because it was the first one I encountered.  But it was not the only version of the standard figures in the Regency era, even leaving aside the standard country dances and the other whole-set dances (such as the very odd one I described here) set to the same tune.  The version published in Platts’s popular & original dances for the pianoforte, violin &c., with proper figures. Vol. 3, no. 25 (London, 1811) is almost precisely contemporary with the version Wilson was publishing from at least 1808 (in An Analysis of Country Dancing) onward.  The description has three notable differences, one of which makes me want to seriously reconsider how I teach and perform the dance.  I’ve transcribed the description at the bottom of this post for those who want to see for themselves.

The essential pattern of the Platts’s version is the same as Wilson’s: a series of diagonal figures, each performed first by the top lady and bottom gentleman of the set then by the bottom lady and top gentleman.  This is followed by a “shoelace”-type weave by the top (active) couple through the other couples to the bottom of the set, then by a promenade up with the other couples falling in from the bottom, and all the couples casting off, leaving the original top couple at the bottom and all the others moved up one place.  A more detailed breakdown may be found in my post on Wilson’s version.

So where does the Platts’s version differ?  Aside from some minor differences in wording, there are three places where the performance diverges from Wilson’s.

First, and least significantly to me, the diagonal figures have been trimmed.  Wilson’s sequence is advance and retire, right-hand turn, left-hand turn, two-hand turn, and “allemande”, which in Wilson’s terminology is a dos-à-dos.  The Platts’s version deletes the right- and left-hand turns, jumping directly from advance and retire (“meet in the middle”) to the two-hand turn, and then, interestingly, to “Pousette round each other”.  I don’t believe this is a poussette in the sense of couples circling each other – it’s being performed by only two dancers.  Instead I think it is merely the two dancers following a poussette-like track by “pushing” forward past each other then “pulling” back around the other side while still facing the original way.  In other words…a dos-à-dos.  So the only variation here is the elimination of two of the diagonal figures, a precursor to later versions which would eliminate even more of them (see here and here for examples).  

The second difference is more interesting to me: instead of just weaving through the other couples in Wilson’s style, the Platts’s version has the couples taking hands every time they cross – first right hands and then left.  The move is described as a series of turning halfway and casting off.  The actual weaving track is the same as in Wilson, who doesn’t specify which shoulder the dancers pass by.  I tend to default to “always right shoulders” because asking people to alternate passing shoulders does not tend to go well.  But I think it would be easier to remember to alternate if the dancers actually take hands on each pass.  

The third difference is not even in the actual dance instructions, but it is the one that makes me want to reconsider how I perform the dance.  At the end of the description, there’s a note about the music specifying that the first part of the dance (the diagonal figures) is performed to the first part of the music, the second part (the weave with hands) to the second part, and the third part (promenading up and casting off) to the third part.

Oh.

Oh.

Ow.

I know, having danced this dance in many different places with teachers other than myself, that I am not alone in not correlating the three strains with the dance and just having musicians play AABBCC over and over.  Every recording I have of it follows that pattern as well.  But while Wilson did not specify that the parts of the dance correlate with the strains, I think it is so obvious a thing to do that perhaps he didn’t feel a need to mention it.

And that presents some serious practical challenges. 

Most obviously, if there is more than one set of dancers, they would need to have the same number of couples and keep very precisely correlated with each other to make this work.  Anyone who has taught this dance or its American descendant, the Virginia Reel, at a ball knows that that plan is, ah, very optimistic.  The sets end up different lengths, some dancers move faster than the others, and after once or twice through they end up skewed.

The other significant challenge is that how many times each strain is played depends on the precise speed at which the dancers perform the figures.  With some drilling of the dancers, the diagonals could be standardized to a specific timing.  But the weave and promenade would be much harder, since how long they take depends partly on the length of the set.  Even the diagonals would be difficult to perfectly standardize – the difference in travel time in a set of five couples vs. a set of ten couples would matter.  Calculating this on the fly at a ball would be, well, challenging.

One solution would be to have a single set and just have the master of ceremonies signal the musicians when to change strains.  I suspect that’s about how it would have worked in period, possibly because not every single person danced.  And it would be fine at a house party.  But it could be very unwieldy at an event with a large number of eager dancers.  Ten couples, perhaps.  But twenty?  Thirty?

Another approach to the problem would be to settle on a specific number of couples in a set, enforce it among all the sets, and drill the dancers in advance to perfect the timing.  That’s great if one actually can drill the dancers in advance.  It would work for a performance team.  But it is not going to work at public balls where people are just walking in with no previous study.  And it completely wrecks the utility of “Roger de Coverley” as the easy finishing dance that everyone can do.

The only even semi-workable approach I can think of would be to assemble some dance-testers and work out the number of repeats of each strain that fits sets of X number of couples (possibly for several different values of X) moving at a typical pace, enforce that length of set, and have the musicians play that repeat pattern while the dancers adjust themselves to it on the fly.  That might involve a little bit of waiting for the musicians to catch up or a bit of scurrying to finish a figure in time, but it ought to more-or-less work.  I’m not thrilled by the “more-or-less” part of it, but I think it’s better than turning “Roger de Coverley” into a dance only for those who’ve spent time drilling it intensely.

There would be even less flexibility if using recorded music, but if one edited a recording or had one custom-made, it could be made to work.  Having to rearrange the strains this way would be an unpleasant editing challenge, though, much more complicated than just looping extra repeats.

Given the difficulty of implementing the dance/strain correlation, I should perhaps ignore this as a weirdness specific to the Platts’s version, decide that Wilson did not mention it because he did not correlate figures with the strains of the dance, and just keep doing it as I’ve always done it…but I’m not sure I’m entirely comfortable with that approach now.  The dance/strain correlation just makes so much sense.

Fortunately or otherwise, I don’t foresee many opportunities to teach “Roger de Coverley” right at the moment, so it’s not an immediate issue.

If anyone else experiments with this dance/music correlation, or even just with using the right and left hands in the weave, I’d be very interested to hear the results.

~~~~~~~~~~~~~~~~~~~

The original instructions from Platts’s popular & original dances for the pianoforte, violin &c., with proper figures. Vol. 3, no. 25 (London, 1811):

The top Lady & bottom Gent: meet in the middle; the top Gent: and bottom Lady do the same  the top Lady and bottom Gent: meet in the middle and turn with both hands; the top Gent: and bottom Lady do the same; The top lady and bottom Gent: meet in the middle and Pousette round each other; the top Gent: and bottom Lady repeat the same figure   The top Gent: turn his Partner half round with the right hand & cast off 1 Cu: turn again with the left and cast off another Cu: and so continue to the bottom of the sett;  The 1st. Cu: having arrived at the bottom lead up the middle the other Couples following in rotation: being at the top the Ladies cast off on their own sides and the Gentleman on theirs.

NB. The 1st part of the tune must be repeated ‘till the 1st. Cu: are ready to cross over and the 2d. part ‘till they are arrived at the bottom; the last part to the remainder of the Figure.