It's not unusual for new sources to turn up that make me go back and reconsider a reconstruction. It's a little irritating for it to happen less than a month after I finally get around to publishing one here on Kickery, and doubly irritating for it to be not a new source but old sources I simply hadn't looked at recently. Fortunately, this is less a change in my reconstruction than further background and options.
In reconstructing the fourth figure of the Mid-Lothians, an early 1820s quadrille, I wrote in my reconstruction notes that "I've never found any description of what step sequence to use for this figure," referring to the grand chain. Actually, I had come across such, many years ago, and they had simply slipped my mind. But I was looking through quadrille sources for a different project and found them again, so here is a little more information about performance options for the grand chain.
I'm not going to analyze here all the possible ways in which the grand chaîne may be danced. Limiting myself solely to the sixteen-bars-quite-round version, used in Mid-Lothians as eight bars for halfway round only, here are the two pertinent references:
(1) In Thomas Wilson's The Quadrille and Cotillion Panorama (first ed. 1816, second ed. 1822), he states that "...when Sixteen Bars are used, then a Chassé, Jetté [sic] and Assemblé is made with, and after taking each person's hand."
(2) In Duval's Lancers Quadrille (1817), the fifth figure (Les Lanciers) starts off with a grand chain, though it's not named as such: "Right & Left all round making ballotez every time the right and left hand is given." The French version is "A la droite et la gauche tout au tour ballotez chaque fois la droite et la gauche se donnent."
Wilson clearly contradicts my reconstruction, directing that the grand chain be performed with a series of chassé, jeté, assemblé repeated eight times, once for each change of hands. This has the virtue of being very clear and nicely regular to perform, but one has to keep the chain fairly tight and really travel on the chassé steps. I also don't find it as enjoyable to dance the same short step-sequence eight times in a row, but that is a matter of personal preference.
Duval is not as clear as Wilson, and his poor grammar doesn't help. Ballotez, as used previously by Duval in the first figure of his Lancers, as well as by other writers in other quadrilles, generally takes two bars of music. But in this figure he almost certainly uses it to mean a one-bar step-sequence. If one interprets "every time the right and left hand is given" as every time the right hand is given and every time the left hand is given, ballotez can be no more than one bar. Two bars of ballotez after every change of hands are simply impossible in sixteen bars of music. This interpretation of Duval would imply a sequence similar to Wilson's: chassé, ballotez repeated eight times. But that is now actually how I would read Duval's instructions. I think that if this is what he had meant, he'd have said "every time a hand is given" (chaque fois la main se donne) or similar.
It seems more natural to me as an English-speaker to read Duval as meaning that one is "making ballotez" after one has given both the right and left hands in sequence. With one bar for ballotez, this suggests a sequence of chassé x3, ballotez repeated four times. This is the same general pattern of "chassé x3, [something]" as a regular chaîne anglaise (right and left) in which one changes by both the right and left hand during three chassé steps before making a jeté, assemblé. It is typical for a full eight-bar figure to use the same steps as the four-bar (demie-, or half-) version performed twice, so it makes some sense for a sixteen-bar figure to use the same steps performed four times. It is notable that Duval calls his figure "Right & Left all round", which makes simply extending the same pace of changes (two in three bars) to the longer figure a reasonable interpretation. I also find this more enjoyable to dance than the one-chassé-per-change version and prefer the longer and more graceful paths one can use with more bars of travel.
It is also conceivable that Duval meant to employ two bars for ballotez, for a series of chassé x2, ballotez repeated four times. This would have a very different feel. Two bars of travel, changing hands twice, would alternate with two bars of dancing in place. I don't much like how this rushes through the changes, and it does not seem to fit the style or match the pace of other step-sequences of the era.
Given that one cannot determine with certainty what Duval meant, the only useful conclusion one can draw from all this is that there are three possible patterns for stepping through the grand chain:
chassé, [1b something] repeated eight times
chassé x2, [2b something] repeated four times
chassé x3, [1b something] repeated four times
The "something" might be either a one- or two-bar ballotez (Duval) or a one-bar jeté, assemblé (Wilson). The exact steps do not really matter to the pattern of the dance, since whatever it is is performed individually. Going into the nitty-gritty of exactly what ballotez means in this era is beyond the scope of this post, but it does not travel and clearly, per Wilson, jeté, assemblé is an acceptable non-traveling step-sequence.
So where does this leave me? I find the two-chassé pattern very unlikely. The one-chassé pattern is certainly documentable (Wilson, possibly Duval) and is a very reasonable option. The three-chassé pattern strikes me as the most likely reading of Duval, with the advantage of being congruent with the standard right and left (chaîne anglaise) step-sequence given in multiple sources of the era.
With this much ambiguity, I can't make any sort of final determination. It's interesting to have this kind of variation in a step involving multiple dancers; usually variant step-sequences are found in moves where the dancers move independently (traversez, balancez, etc.) I suspect that Duval and Wilson each regarded their step-sequence as the only one, but unless further sources turn up, we've no way to determine from our two-hundred-year distance what step-sequence was most common in practice.
The critical requirement in dancing a figure like this is that the all the dancers must agree, at least on the issue of a one-chassé pattern or a three-chassé pattern. So it comes down to a choice by the reconstructor. In general, for multi-person figures in a quadrille reconstruction used socially, I look for the simplest step-sequence that I find reasonably supportable from period sources. So, pending further evidence, I will keep the chassé x3, jeté, assemblé pattern, repeated four times, for the grand chain in the fourth figure of the Mid-Lothians and elsewhere.