How Many Times Do We Play That Tune?

One of those questions I get asked all the time by musicians and others is "how many times through the tune for this dance?"  The reference is to progressive longways country dances, which were the dominant social dance form in Europe and America from the mid-17th century through the early 19th century and are still enjoying widespread popularity in various living tradition and revival forms.

Modern English country dance and contra practice is for all couples to start the dance simultaneously, and the modern answer to the repeats question would be as many times as needed for everyone to enjoy the dance and fewer times than it would take for people to get bored.  Modern Scottish (RSCDS) practice differs in that their dances are generally performed in short sets and have a fixed number of repeats.  But if you truly wish to perform country dances in the historical style, it's a bit more complex!

Up until the mid 19th century, and possibly later, country dances were performed with one key difference from the modern style: the couples did not all start simultaneously.  The dance would be led off by the top couple in the set, who would dance down the set with each couple in turn.  As soon as the proper number of couples for the dance accumulated at the top of the set, a new couple would lead off, following the original active couple down the set in turn.  The dance would continue to "snowball" until the entire set was dancing.  As the couple who started the dance at the very bottom of the set reached the top and led off, the dancing would then slowly dwindle as that couple made its way down the set to the bottom until at the end only the last two couples were performing the dance.  This form of progression was the standard all the way from the first edition of Playford into at least the early 19th century, and there is some evidence it lasted into the early 20th century.

So what does this mean for the repeat count?  In a long set, the dance could potentially go on for an extremely long time - forty-five minutes to an hour in a set of twenty-plus couples, with the tune repeated 80 or 90 times.  Pity the musicians cursed with playing a sixteen-bar tune over and over and over and over again!  And just how many times over was that?  Happily, there's a formula!

For a dance in duple minor form (two couples needed to perform the figures), the repeats work as follows:
Three couple set: six times
Four couple set: nine times
Five couple or more: add three more repeats for each couple over four

So for a set of six couples, it would be 9+3+3 = 15 times through the dance.  Duple minors were the dominant form in the mid-to-late 17th and early 18th centuries.

For a dance in triple minor form, needing three couples to perform the figures, the formula is a bit different:
Three couple set (shortest possible): six times through
Four couple set: twelve times through
Five couples or more: for each additional couple over four, add four more repeats

So, for a set of six couples, it would be 12+4+4 = 20 times through the dance.  Triple minors were the dominant form from the late 18th century onward.  In the early 19th century, at least one dancing master added the suggestion that the original top couple should lead off a second time, which has the virtue of keeping more couples dancing longer but doesn't affect the total number of repeats needed for the dance.

As you might imagine, for reasons of sanity (dancers', musicians', and my own), when I call country dances in historical form, I cut sets off at six or seven couples, which is a long enough set to appreciate the "snowball" progression but short enough that no one goes completely batty.

Comments

[TexAnne]

Geez. No wonder Lizzy didn't want to dance with dear Mr. Collins!

[Fragano Ledgister]

I expect that the number of couples in the dance would be prearranged for each dance, by the use of dance cards. Still, it makes the upper middle and upper class entertainments of the eighteenth and early nineteenth century seem a trifle tedious.

[Abi]

Staggered starts...that also explains why the entire room was waiting for Fanny to start dancing in Mansfield Park. Since she was very low on the social order, she was used to standing around a good while after everyone else had started.

It also explains why she was so nervous. The whole room was watching her.

[Susan de Guardiola]

Goodness, I wasn't expecting comments on what I thought was a rather dry post!

TexAnne: Not only could a dance take 45 minutes or more in a long set, custom was to stand up two dances in a row with the same partner, so yes, standing up with Mr. Collins would have been not merely dreadful but endless. You'll note that when Mrs. B. describes Jane's experiences at the ball she counts off most of the dances in twos ("the first two with ..." etc.) Given the potential length of each dance, in the five or so dances and dance-pairs she mentioned she may well have summarized the entire ball, and Jane dancing two pairs of dances with Mr. Bingley being significant suddenly makes a lot of sense - they could easily have spent three hours together.

Fragano: Two things. First, I've not yet found any documentation of the use of dance cards that far back. The concept of dance card is so embedded in our culture that we tend to apply it to any historical period. (I made this same criticism of an alternate-Regency novel from Tor which was simply loaded with dance errors.)

Second, tedium is relative. In the 18thc, country dancing would have been preceded by a series of minuets and other formal dances where one couple at a time danced in strict order of precedence while everyone else sat around and watched. For hours. Compared to that, waiting a few minutes for a dance to come down the set is nothing. And we have also lost some of the social context: keep in mind that the dance floor was the only place where young men and women could be more or less unchaperoned together. Lizzy & Darcy using the breaks in the dance to talk in Pride and Prejudice is (naturally) dead on. The waiting-out bits were valued chances to flirt and talking during them prevalent enough that dancing masters preached against it.

Abi: Leading off a dance was a high-pressure job. Everyone watched intensely because the first lady in the set would be the one who determined the tune, the dance figures, and the steps to use. The entire rest of the set picked it up by watching. It's amusing to watch at my assemblies which dancers try to get to the top of the set and which ones try just as hard to make sure they're at the bottom so they have a chance to watch before having to do it. This element of performance-for-peers is something else that has vanished in most modern dance contexts and is worth a post of its own sometime.

[Fragano Ledgister]

Susan: Of course tedium is relative (and life in the 18th century moved at a far slower pace than our own, and was more driven by the cycles of nature).

I can't claim to be much of a historian (except in my own narrow areas of geographic and axiological interest), so I shall have to revise my image of what a formal dance would have involved in the days of the Regency or earlier.

[Lee]

FYI, I linked to this from my LiveJournal. Turns out one of the people I was thinking might be interested already knows you! (That would be Dr. Becky from Boston -- I know her by 2 different online names, and am completely blanking on her mundane last name.)

[Marilee J. Layman]

But repeating it until everybody is satisfied is very boring for the musicians!

[Susan de Guardiola]

Marilee:
Yup, and there are occasional disapproving mentions in early 19thc dance manuals of musicians, ah, taking the bit in their teeth and changing the tune partway through; this seems (at least to the dancing masters) to have been regarded as disrespectful of the lady who determined the tune.

When doing period-style country dances (for me, primarily when doing Regency, since I've no good opportunity to do entire events set in the late 17th or early 18th centuries) nowadays, along with limiting the set so there are no more than 24 repeats (seven couples, though I prefer only six and 20 repeats), I also explicitly let my musicians change the tune midway through. It's always an interesting process trying to balance period style and modern sensibility, but I feel that the "snowball" starts bring in at least a smidgeon of that performance element which is so little a part of modern country dance.

At the assembly I ran in October, I had ladies "call" (state) the tune and dance figures, too, though we pre-rehearsed the specifics since ladies today are prone to freeze in terror if you just suddenly turn to them and tell them to call a dance.

[Paul A.]

This post has also given me an "oh, so that's what that bit in [book] is about!" moment. For me, it's the description of Fezziwig's party in A Christmas Carol, where there's a dance that starts off properly and ends up in chaos (but everybody's having so much fun they don't care); I'd been able to make out that much, but now I can also interpret the details of the scene as well.

[Cathy]

Oh how I love the minutiae of dance geekery. I really appreciate how you try to bring forward all these interesting little details of how dancing actually *worked* in the past. Not just the figures and steps, but the cultural context is so interesting and important to understand.

Some time I think it would be amusing to run a ball that begins with hours of tedious minuets (oh, the performance anxiety). Then again there might not be *hours* because I bet I might be the only person who wanted to attend such a ball!

[Marc Hartstein]

The formula can be generalized to:

r = (c+1)(s-1) [if s>c]
r = c (s-1) [if s=c]

Where:
r = the number of repeats through the music required
c = the number of couples required to perform the dance
s = the number of couples in the set.

(The case where the number of couples in the set equals the number of couples needed for the dance is degenerate.)

I can provide a derivation if anybody is interested.

If someone is aware of a dance form which uses n-tuple minor sets for n greater than three, this should work correctly.

If you use the Regency option of having the lead couple dance a final time through to the bottom, the formula becomes:

r = (c+1)(s-1) + c [if s>c]
r = (c+1)(s-1) [if s=c]

[Susan de Guardiola]

Marc:
I haven't actually tested the formulas, but very cool. Two notes, though:

(1) When the top couple leads off a second time, they do not dance all the way to the bottom. The original bottom couple dances all the way to the bottom and that ends the dance. The top couple ends a few places higher and then they drop to the bottom for the next dance, everyone else in the set moving up one place. So there would be no difference in the length of the dance and no need for a separate formula.

(2) Wilson (first quarter of 19th century) actually recommends that all dances include neutral couples between the dancers, which means that his triple minors are functionally danced in quadruple minor form, with one couple never doing anything at all. The object was to provide a "spacer" couple so as not to get figures tangled between adjoining minor sets and to provide a sort of static background for the dancers. In a triple minor dance where the third couple already had nothing to do (possible; the Regency default was triple minor even if the third couple was entirely inactive), then there would be no need for a fourth couple to act as spacers.

Since (1) I've never found anyone other than Wilson espousing this idea of spacer couples; (2) even Wilson doesn't espouse it that heavily, as his diagrams don't show them; and (3) sheesh, not more repeats in a dance, I don't use spacer couples and thus, happily, never have any need for formulas for quadruple minor sets. There are other period formats for dances, but to the best of my knowledge none that involve larger than triple minor sets.

[Marc Hartstein]

Susan:

1) Ah, my mistake. I had somehow assumed that the lead couple would dance all the way to the bottom, rather than dancing most of the way then dropping. It made sense to my aesthetic for completeness, leaving them conveniently in the right place.

2) I'd forgotten about the spacer couples. I suppose if you could handle the sorts of very large sets previously discussed, adding 1/4 again to the number of repeats wouldn't be such a big deal. I wonder if it would avoid confusion more than it would result in confusion when people who were supposed to be spacers tried to dance. Given how often we see that at the top and bottom of the set, even among experienced dancers . . ..

(Note that for practical purposes, what I wrote can be remembered as one formula [r = (c+1)(s-1)] plus memorizing that in the unusual case of 3 couples for a 3 couple dance you need only 6 repeats)

[pecunium]

I happen to much like the out time in the sets. My habit is to dance not more than one dance, per section of the evening (assuming there to be intermezzo), with someone.

Admittedly, being male this is easier for me to pull off, because as a rule, men are under-represented at the events I attend.

[Susan de Guardiola]

pecunium:
What sort of dancing do you do? (just curious)

[Marnen Laibow-Koser]

In case anyone is interested in a comparison figure, modern American contra and ECD groups seem to go somewhere around 13 times in most cases, unless the sets are extremely long (over 20 couples or so, perhaps?). I understand that in the UK, they run longways dances shorter.

[Susan de Guardiola]

I am constantly being frustrated by recordings which run 7 times through, which is one repeat too many for a three-couple Regency set. I can usually trim them down fairly easily with editing software for 6x through, which is perfect, but it would be nice to have recordings long enough for a longer set (12x would allow a 4-couple set) since it's harder for me to loop them cleanly with my nonprofessional software and modest ability with it.

[Susan de Guardiola]

I should add that two CDs aimed at Regency dancers have the correct number of repeats on several dances:

Lady Caroline's Regency Romp has three country dances at 6x and two waltz country dances at 6x, plus one at 8x; no idea what the idea was with that last.

The Spare Parts CD The Regency Ballroom (see the far right column for the cover) has a 32b country dance 12x for a four-couple set, a 24b country dance run 6x for a three-couple set, and a 40b waltz played 6x for a waltz country dance ("Spanish dance", as they were known in the late 1810s.)

(I served as dance advisor to The Regency Ballroom and Lexington folks who did Lady Caroline's had taken a Regency workshop from me and had access to these numbers.)

[Donna Hatch]

This is really fun! I attended a Regency Write's conference years ago where a dance master taught us to dance, but I guess the modernized it because I don't remember waiting around very much. We were in groups of two couples but we did move down the line until we got to the end and could take a breather (much needed by then!) before we got drawn into the pattern again. It sure was fun!
Thanks for further clarification.

[Michael]

Hmmm. At least one version of Pop Goes the Weasel implies spacer couples as an option. Knowing that Wilson advocated them makes that direction make more sense to me than it had.

[Susan de Guardiola]

Michael:
Interesting; which source? What would make sense to me is not spacer couples per se, but the dance itself being considered triple minor, with the third couple simply having nothing to do and thus being a de facto neutral couple.

Given short modern attention spans and the apparent difficulty of counting to three in mid-dance, I don't think neutral couples have a big future in reconstructing period dances.