March 2016

Congratulations to Henry Pipe

BellBoard addicts might already have noticed that Henry Pipe has rung Norman Smith's 23-spliced on handbells. This is a tremendous achievement, all the more so at the age of 13. I had the pleasure of ringing one part for practice, with Henry, David Pipe and Alex Byrne, on our holiday last summer, and he was already ringing it very well.

According to the footnote, it was also a first for Alex Byrne, although he has previously rung Chandler's on handbells, which is much harder. Anyway, Henry and Alex become the 48th and 49th people to have rung Norman Smith's in hand. Well done indeed.

Separation Anxiety

How far apart do they get? That's often considered a factor in estimating the difficulty of a particular pair in a method. It explains the experience that ringing the slow work is the most difficult aspect of Kent on the tenors. Following on from Tina's recent article about the features that make a method handbell-friendly, we can look at the handbell-friendliness of each pair in standard methods, as measured by average separation. What do I mean by average separation? The separation of a pair (e.g. 7-8) in a change is the number of bells between them: 0 if they are in adjacent positions, 1 if they are in 2nd and 4th place (say), and so on. Adding up the separation of the pair in every change in the plain course, and then dividing by the length of the course, gives the average separation. Here are the results for the standard eight Surprise Major methods.

Average separation for each pair: Standard 8 Surprise Major

Method 1-2 3-4 5-6 7-8
Cambridge 2.00 2.11 2.27 1.63
Yorkshire 2.00 2.09 2.68 1.23
Lincolnshire 2.00 2.20 2.27 1.54
Rutland 2.00 2.21 2.27 1.52
Pudsey 2.00 2.05 2.57 1.38
Superlative 2.00 1.89 2.75 1.36
London 2.00 1.98 2.59 1.43
Bristol 2.00 2.21 2.82 0.96

The first interesting thing about this table is that the average separation of 1-2 is always 2. The second interesting thing is that for each method, adding up the average separations of all the pairs always gives a total of 8 (apart from rounding inaccuracies). In the comments I explain why this is.

For each method, the tenors stay closest together, then 3-4, then 5-6. This supports the usual perception that the tenors are the easiest inside pair, followed by 3-4, and 5-6 are the most difficult.

The smallest number in the table is 0.96 for the tenors in Bristol, but I'm not going to claim that Bristol on the tenors is easier than Yorkshire. However, if we look at the right-place methods, the smallest average separation for the tenors is in Yorkshire, and this supports the common approach of regarding the tenors to Yorkshire as the best starting point for Surprise Major.

The fact that the total of the average separations, for a given method, is always 8, means that if a method has one pair that stays close together, there must be other pairs that are further apart. For example, the very low 0.96 for 7-8 in Bristol is balanced by the high 2.82 for 5-6.

It's interesting to have some quantitative data to feed into the discussion about handbell-friendliness.

What makes a method handbell-friendly?

Going through Simon's occasional series of handbell-friendly compositions, I have been giving some thought to what features make the method itself handbell-friendly.  Some methods do lend themselves well to handbells, while other methods are surprisingly tricky.  Here are some of my thoughts on some particular features and a 'handbell-friendliness' score out of 5. 

Concurrent work

By concurrent work I am talking about methods where much or all of the pieces of work line up neatly, so that although you are ringing two bells, you are changing to new sections of work at the same time.  Plain Bob has this feature in a small way - all the bells are either hunting or reacting to a seconds place.  Kent Treble Bob has this feature - all the dodges line up, so if one of your bells is dodging, then the other bell is probably also dodging.  Stedman, especially above triples, also has this feature because of the sixes structure. 

The principle extends into Surprise methods as well:  in Cambridge, the 3-4 places begin and end at the same time as the Cambridge front work.  Yorkshire likewise offers smaller but similar concurrent sections of work, which can help stabilise the ringing and give you a mental break.  This works in less common methods as well:  Cassiobury has reasonable sections of concurrent work, and the concurrent work around the half lead in Whalley certainly eases the challenge it offers. 

Methods where there isn't concurrent work seem very difficult.  For example, Bristol is trippy because the fishtail work at the front and back overlap by just a small amount.  And let's not get started on the points that don't line up in Belfast! 

Handbell-friendliness: 5/5

Right places

Right places have a lot of advantages:  you know that whenever you make a place (or a consequent dodge), you will always do it a backstroke.  It offers a lot of opportunities for mental relaxation:  if you can manage it, you only have to consider your position every other stroke or only 50% of the time you are ringing. 

And yet, once you get accustomed to ringing wrong places, they can make a method easier:  take Oxford and Kent.  Oxford has nice right places in 3-4, which at first seems much easier to manage, whereas the cascading (and wrong) places in Kent are always a problem when trying to learn it.  However, the effect of the right places in Oxford tosses you around pattern-wise, whereas you always leave those Kent places in the same pattern you entered them. 

Handbell-friendliness: 4/5


Symmetry always sounds like a good thing, but symmetry in handbell ringing is not really much of a help, and I'm not sure why.  Possibly it is because you are ringing two bells each at different points, but I think it may be because all the pieces of method are a little too much the same as their counterpart, and it is easy to lose track of which bit of work you are actually doing.  Also, the whole 'do the same work backwards or upside down' feature of symmetry which is so helpful in learning methods in tower seems to be too mentally tiring to work for a handbell pair.

Superlative, for example, always seems to be trippy.  While Bristol is challenging for all kinds of reasons, the symmetry doesn't really offer much help.

Little pieces of symmetry within a method are also not so very helpful.  The symmetrical lead of any Surprise method often seems to be the hardest, and certainly tempts you to swap your bells, especially around the half lead.  The symmetrical lead for the tenor pair in Yorkshire major has the least amount of coursing in it.  Where symmetrical leads are easier (like in the 3-4 course in Bristol or Belfast), they are simpler for other reasons (keep reading). 

Handbell-friendliness: 2/5

Distinct front and back works

There are many methods which have a very distinct division in work on the front and work on the back.  St Clements has this feature in a very simple way, with dodging on the front and plain bob above.  Kent and Oxford are also very obvious examples of this.  More complex methods like Bristol or Belfast have very distinct front and back works, with place bells that never go above 4ths or below 5ths place. 

When both of your bells are in either the front or the back part of these methods, nothing could be easier.  They are dream leads.  When you are ringing methods on higher numbers, this feature really comes into its own, because you spend more time inside the backwork.  It can be a real bonus for some very hard methods like Phobos (Simon has discussed this in detail in a previous post on structural ringing).

However, when you have one bell on the front and one bell at the back, the difficulty increases quite a lot.  Remember how much the ringing can destabilise when someone hesitant hits the frontwork leads when ringing Kent - one bell in the frontwork and one bell in the backwork. 

Handbell-friendliness:  3//5


Staticness is where you spend a long time doing trivial variations (combinations of places, dodges or points) in one dodging position, so you spend a long time in a lead not really going anywhere.  It sounds like it should offer a nice easy lead or two, letting one bell waft in a fairly static place which concentrating on the other. 

It never works that way.  Take a little method called Kelvinbridge Surprise Minor (x38x14x12x1236x1234x36, 16).  Nobody moves very much:  the frontwork is an entire lead of places and dodges (not symmetrical), with a lead-long crankshaft work in 3-4, and an easy and regular backwork that fills in the gaps.  You can't get longer pieces of concurrent work.  However, when we rang it some years ago it proved a little tricky, and was potentially trippy, as we tried to keep our direction straight and do the dodges and places in the correct order.

Superlative is a method which is also notoriously tricky, despite having very long pieces of concurrent work and a stabilising 5-pull dodge at the half-lead.  The pieces of work are just too long and it is easy to lose track of where you are on the line.  The frontwork in Belfast is very easy to get out of sync, and usually results in trips on the front.  Double Dublin is a minor variation of Bristol, so if you can ring Bristol, Double Dublin ought to be a wheeze.  And yet, the small variation creates extremely long works at the front and back, and it is so easy to get out of sync.

Handbell-friendliness: 1/5

Handbell Compositions: Superlative Major

Next month's peal with the Edinburgh crossover band is going to be Superlative. It has interesting possibilities for handbell-friendly compositions, because the 24 courses with 3-4, 5-6 and 7-8 all coursing are mutually true. These are the coursing orders of the form xxyy2, xx2yy, 2xxyy where xx can be 34 or 56 and yy is the other pair, 56 or 34, and the order within each pair can be swapped independently.

To combine these courses into the ultimate handbell composition, the first step is to start and finish at the treble's backstroke snap, so that the coursing order is 65234. The difficulty is swapping 3-4 and 5-6 into each other's places. For example, how can we go from 65432 to 43652 without separating 5-6?

One possibility is to use a non-standard call. A 123456 single reverses the coursing order in one step, so 65432 becomes 23456 and 3-4 are now to the left of 5-6. Using this idea gives the following composition.

5376 (5152) Superlative Surprise Major
Simon J Gay

M  W  H  (25364)
-         35462
s  s      65423
   -* #   26453
8 part, calling s for -* in parts 2, 4, 6, 8,
and calling 123456 at # in parts 4 and 8.
Snap start and finish.
For handbells: 3-4 and 5-6 course throughout.
For 5152, omit one of the ss resulting from a call substitution. 

If you prefer to stick to standard calls, Don Morrison has a neat composition that uses half-lead bobs.

5376 (5152) Superlative Surprise Major
Donald F Morrison (no. 1694)

(25364)  M  3½  W
 64532   -  [-  -]      
 23546   s      2*     
 24536   ss     2*
Repeat three times, omitting bracketed bobs from alternate parts.
Snap start and finish.
2* = s -. 
For 5152 omit one ss.
For handbells: both 3-4 and 5-6 course throughout, except for one half-lead each.

What happens at the first half-lead bob (which is a 58 place notation) is that the coursing order is 65342 and the bob produces 63452; a bob at 3½ has the same effect as a home. Then the bob wrong, at the next lead end, produces 34652. At the half-lead bob, the 4th makes 5ths as the 3rd does 3-4 down, which puts 3-4 into the 5-6 position at the point where they do 3-4 places and 5-6 places in the second half of the lead. At the lead-end bob, the 3rd makes 4ths and the 4th does 5-6 down, putting them back into coursing. The second time a half-lead call is used, 3-4 and 5-6 are in each other's positions, so it's 5-6 that go into the 5-6 position for half a lead.

In practice, with our band, I don't think we really need to keep both 3-4 and 5-6 coursing. However, the Don Morrison composition has such a neat structure that I'm tempted to call it anyway.

Another composition by Don Morrison uses the same combination of a half-lead bob and a standard call at the next lead end, as a way of moving between coursing orders 8765xxx and 6587xxx. This produces lots of good combinations of 5678, with a very small amount of split tenors (as 7-8 do the work described for 3-4 in the previous composition).  

5184 Superlative Surprise Major
Donald F Morrison (no. 1930)

(25364)  M   W       H
 24536  [-]      a
 36452       ss  b
 26453       s   a   s
Repeat five times, omitting [-] from alternate parts.
Snap start and finish.
a = 4½,s5; b = 5½,6. 
Contains all 24 each 56s, 65s, 5678s off the front, 6578s off the front, 8765s off the front and 8756s off the front, 18 each 8765s and 8756s, and all 7 near misses.

Handbell Compositions: 5056 Rutland Surprise Major by Don Morrison

We're doing well with the "Edinburgh crossover band" - we rang a peal of Lincolnshire last month, using one of the rotated William Barton compositions. Last week we rang Rutland. What was the composition? Rutland isn't a particularly popular method on handbells (in fact last week's peal was my first) and unlike Yorkshire, there aren't a lot of compositions advertising themselves as handbell-friendly.

The Henry Dains composition that we rang for London is also true to Rutland, and it would also be possible to use a handbell-friendly composition of Yorkshire, such as this one.

5,376 Yorkshire Surprise Major
Roger Bailey

(32456) M  W  H
 34256        s
 52436     2  2
 36425  -  -  s
 45263  - [-  2]
(35462) 2
Twice repeated. Snap start and finish. 3–4 ring 18 courses of 7-8
position. True to all Bc. 
Reduce to 5152 by replacing [...] by WHWH and omitting it in one part.

I have called this composition for Yorkshire, but last time we tried it we came unstuck because I overlooked the fact that omitting the WHWH results in 3-4 ringing part of the 3-4 course that they don't get in the other two parts. I was ringing 3-4 myself, and I was so perturbed by ringing a new combination of place bells in the last part that I fired it out. The worst of it was that exactly the same thing had happened in the previous attempt, which I then forgot about because of the simultaneous and even greater distraction of gathering darkness because of not switching the lights on before starting... peal-ringing adventures, eh?

Anyway, I wanted to find a different composition for Rutland, and picked this one by Don Morrison.

5,056 Rutland Surprise Major
Donald F Morrison (no. 1)

23456    M  B  W  H
54632    -     -
43265       2     -
24365       5     -
32465       5     -

It's a simple 2-part, and 5-6 are coursing 73% of the time and never ring the 3-4 position. In each part, one of the blocks of 5 befores has 3-4 coursing in 4 of the courses. In that respect, both parts are equally good for 3-4, even though the part end isn't 12436578 which would usually be considered the most desirable for a 2-part handbell-friendly composition.

Many compositions of Rutland have a before in most of the courses, and this one is no exception. In this case, the first course of each part is the only one in which 7-8 ring 2nd and 3rd place bells. It's OK at the beginning of the peal, but the ringer of 7-8 has to stay alert ready for the beginning of the second part.

When calling this kind of composition, it's best not to try to count up to 5 in the blocks of befores. Here, 5-6 are in the 5-6 position at the beginning and end of each block, so it's easiest to follow the coursing order and just call bobs until 5-6 come back home.

We rang it quite well, and it's encouraging that we have been able to ring each of Cambridge, Lincolnshire and Rutland at the first attempt. Next we might try Superlative.