[LONG] Characterizing Contradances

Skip to first unread message

Warren Pollans

Apr 23, 2004, 11:14:42 AM4/23/04
I'd appreciate hearing your ideas regarding what I describe below.

I have accumulated about 350 dances in my database and am looking for
a way to select similar dances - the problem is defining "similar".
(To have my software do the selection) For example, given dance X
(that I've called and danced many times) are there others in my DB
(with which I may not be that familiar) that have the same "feel"? -
another undefined word. My problem is trying to somehow quantify the
notion of similar dances. It's with this problem of "how to classify
dances" that I'm asking for your input. What follows is a brief
summary of tools that I use - all have helped me, but none solve my
(insoluble?) problem.

I have a parser that enforces certain rules for describing the dances
- among other things, that words are spelled the same everywhere (only
Allemande, not almd, allmd, Alemande, etc) - so that I can search for
dances containing the same sequences of figures (circle, circle, slide
OR circle left, circle right, women chain ...). This enables me to
take a favorite dance and look for others that contain some of the
same sequences. Or go through the entire database and list the dances
that contain each possible distinct subsequence of figures. This,
however, doesn't do what I want - it misses something. Doing this,
for example, indicates that Gene Hubert's Dance Gypsy and A Nice
Combination are the same sequence of figures - some done by different
people (partners vs neighbors) and in different directions (chain
across vs chain on diagonal).

I'm also able to generate a one-line summary of each dance that helps
to quickly scan through a list of dances - . From these summaries I
can get a rough estimate of the "complexity" of the dance - a
simplified version of that found in Zesty Contras (just the number of

Having said all this, it's my feeling that the music drives the dance
- "great" dances without the musical support don't work. On the other
hand, I remember dancing Broken Sixpence (just about as simple as
gets) with one the more well-known contradance bands several years ago
- absolutely wonderful. Perhaps, the solution is to make sure that
you always great music :-) I know that a related problem is matching
music to the dance but I'd like to stick with dance similarities for

I'd appreciate your input - this is as much of a programming project
(perl & mysql) as a dance project.


Mario Herger

Apr 23, 2004, 12:50:46 PM4/23/04
Good questions: in my database with 50.000 entries I have the same problem,
namely how to categorize and how to find automatically similarities.

Let's start with some thoughts
1) Name: does not help, as they are misleading (e.g. including "jig" in the
title does not mean that this is a jig, or polka in the title is not always
a polka)
2) Dance-type: many dances fit into several categories (e.g. Zwiefache are
waltzes and twosteps and Zwiefache; there are circle dances that are also
pure men-dances; dances with 8 pairs can also be dances with 4 pairs,
depends on the figure)
Furthermore many researchers might tend to categorize them different,
depending of their theory used (this dance is related with this type or
originates from this dance-form, therefore it is a blablabla)
3) Extract similarities in number of counts/measures, tempo, figure-types,
position (circle, line, couple, single...)
4) use of tools: sticks, hankies, swords, bell-belts (morris), horns, hats,
chairs, baskets,...
5) "non-dancers": fools (like in sword or morris dances), musicians involved
in the dancing (some swedish dances),...
6) dance-purpose: embedded and performed at a certain tradition, dance-game,
wedding dance, funeral/death dance, show dance, ritual dance,...
7) Dance-origin
8) Age of dance: new choreogaphy, variation of..., Playford-dance,
traditional dance,...
9) Special clothing required: for effects (e.g. morris dance),...
10) Tune: same dance with different tunes might change the character of the
11) execution of dance: slow, fast, celebratory, very relaxed,...
12) People: children dances, dances for older people, ....
13) Patterns: clapping on the shoes, hands and thights, kneeing down, step
dance, ...
14) hold: two hand hold, no hold, hold on shoulder (Sirtaki),
15) Artistry: jumping high, throwing legs, fast and sophisticated step
16) Use of body: leg-centered dances (irish step), arm-oriented dances,
dances with a lot of body-contact,...
17) ...

and so on and so one. Quintessence: I don't think you can automatise a
categorization, the instructions, tunes and names are too tricky. I tried to
do that a little bit in my database, but the outcome is rather poor. I think
you will have to manually categorize them (first specifiy categories), and
still then you will find people who do not agree with your categorization.




Apr 23, 2004, 1:36:18 PM4/23/04
I would also add the categories of Traditional, Revived, Playford or Modern.


David Smukler

May 8, 2004, 5:33:18 PM5/8/04
Hi Warren,

One feature I put into my database of dances was a field that indicates
where the balances fall with regard to the tune. I thought it might be
useful for helping me communicate with the band, since band members
sometimes ask me if there are balances, and if so where they fall. I haven't
ended up using it that often in the way that I intended. However, it turns
out that it is interesting to sort dances according to this field. By doing
so, for example, I discovered that Revolution Reel (Tom Hinds), Rant and
Roar (Russell Owen), I Wish They All Could Be California Twirls (Jim Kitch),
and Al's Advice (Paul Taylor) all have the same structure to their B-parts
in regard to balances as Chuck the Budgie (Rick Mohr), namely a B1 with no
balances, and a B2 with a balance at the beginning and another halfway
through. Chuck the Budgie was written to go with the tune "Pat the Budgie,"
and sure enough all those other dances seem to fit nicely to the same tune
-- although this is not to say that they don't all work well with other
tunes also. But it does create some food for thought. Similarly I discovered
that Snake in the Hey (Cary Ravitz) and Triplet for Felix (Philippe Callens)
have the same (and rather distinctive) balance structure as Wizard's Walk
(Ruth Ungar).

David Smukler

Jim Taylor

May 11, 2004, 9:32:48 PM5/11/04
Received this from the Folk Alliance and (subsequently) Phila. Folksong
Society. Posted here because there are musicians in our midst...
Apologies for the bandwidth...

Back to Lurking now,
Jim Taylor
Chuck Brodsky's stolen instruments - descriptions

I just got home from a trip overseas to discover that a bunch of
instruments were stolen from my house near Asheville, North Carolina,
sometime between April 11th and May 6th. If anybody knows of any other
listserves, discussion groups, hotlines, or websites where I can list
the following descriptions, please let me know.

Please help spread the word. Here's what was taken:

Larry Pogreba resonator guitar
Very dark green, with a hubcap from an early '60's Rambler covering the
resonator cone. The hubcap is cut in the shape of a butterfly. A
pick-up is installed under the cone, with a quarter inch jack at the
bottom end. There are "crude" beads where the metal back and sides
were welded together, and on the back all of the ballpeen hammer marks
are visible from when the back was first shaped.
A photograph of a similar enough Pogreba to mine can be viewed at
http://www.kummersvintage.com/images/nashhub.JPG (This is NOT the one
stolen from me...there will be slight differences between
mine and the one in the photo...but it's very close in looks. A photo
of the actual instrument that was stolen can be e-mailed on request.)

Steve Wise Long Neck Dulcimer
This is an instrument that is played like a guitar, NOT like a lap
dulcimer. It has a thin, teardrop-shaped body (with a cut-away on
either side at the neck), about the size of a mandola, f holes, a
pick-up, small rectangular black plastic battery cover on the side.
The neck is long and is fretted like a dulcimer, but with the strings in
reverse order from a lap dulcimer. It has a low string, a middle
string, and then the high string being doubled. This instrument was
stolen in its white "Calton" fiberglass case with a plush red
interior. The outside of the case has a name plate with "Chuck
Brodsky" engraved in it along with the serial # 8086. There is an odd
piece of plush red padding built into the inside of the case's top to
compensate for the dulcimer's raised bridge.
A photo of this very instrument can be viewed at

Ovation "Elite" acoustic guitar serial # 1868
Thin body, black, rounded back, many small sound holes. A repaired 7
inch crack in the guitar's top will be apparent on inspection. The
guitar strap was multi-colored. There might be some corrosion around
the battery terminal & controls for the pick-up.
A photo of a similar guitar to mine can be viewed at
However, my guitar does not have the different colored wood trim at the
top of the headstock.

Madiera acoustic guitar
The label inside says "Made by Guild". This guitar has maple back and
sides,and "Madiera" inlayed on the headstock. It's a well-worn
instrument, my first guitar. There are at least 2 small screw holes in
the top of the guitar from where an old pick-up used to be mounted.
The area near the pick gaurd has been worn down through the finish, as
are areas around the edge of the sound hole. There were cracks around
the bindings on the back & sides, with one small spot where the guitar
had been dropped and the wood was crushed in a little. There was a thin
braided dark leather strap tied to the bottom end and to just before
the nut at the headstock. Tiny blood spatterings from long ago dot
the inside wood of the back of the guitar when you look through the
sound hole. A photo of this guitar can be e-mailed on request.

Banjolin (brand name unknown)
Possibly has an inscription somewhere "Made in Philadelphia." Belonged
to my great uncle. This instrument has a banjo head (not sure if it
was plastic or skin) attached to the neck of a mandolin (4 sets of
doubled strings). The wooden tone ring is a pattern of various colored
woods, either inlaid or interlocked (not sure). It probably dates back
to the 1930's or 1940's. Some of the metal hardware around the head
might possibly be recent replacement parts(not certain). There is a
thin shim adjusting the angle of the neck at the point
where it attaches to the banjo head. This instrument was stolen without
its original case.

Cheers -
Chuck Brodsky
Asheville, NC

Reply all
Reply to author
0 new messages