[r-t] Decisions decisions

[Mark Davies]

Don gives us three main strands of thought in consideration of one-lead
methods:

> I think it may be related to the fact that you basically can't ring a true
> touch of one of them. You sure aren't going to be ringing a royal one
> of these guys, at least not with a whole plain lead!

I'm afraid I don't think this one holds water, Don. You certainly can ring
true touches of these methods, on any number of bells. Here are some popular
cases:

1. A plain course (which may be a peal length, of course).

2. In a multi-extent block, with a plain lead linking different extents. (I
don't see why we can't imagine this on ten bells, or indeed on any number of
bells that ringers ring; the field is finite, after all. But it certainly is
of practical use on six.)

3. In a touch, using calls, perhaps at the half-lead in addition / instead
of the lead end. (We're quite happy to ring other touches where a method is
never rung plained, aren't we?)

These cases seem a wide and useful field to me, and in some instances are
indistinguishable from what we'd ring with "normal" methods.

> Another striking difference, and I think what would seem
> "asymmetric" is that normally the 2nds and last place versions
> of methods have the same number of leads in the plain course.
> That's not the case with these critters.

But, as you point out, it also holds true for methods giving rise to
differentials on stages where n-1 is not prime. What this argument seems to
boil down to is just another feeling that "1" is different from "3" or "9".
I cannot see any underlying, fundamental reason coming out of this argument
that backs up this feeling . What do you think?

> Perhaps the real underlying discomfort is that when you've only
> got one lead, there really is no useful concept of a lead. ...
> With a one lead course a lead offers nothing that a whole course
> doesn't. So in that case we have something fundamentally different.

OK Don, I think we're getting closer to it here. But what, exactly? Yes,
one-lead methods are more limiting, but so are methods which are very false;
is there a difference in kind here?

I think there is, really. To me, single-lead methods are interesting because
they break open a gap between how we might want to define things in
theoretical terms, and how we actually experience methods and touches in the
tower. In some ways single-lead methods look no different from any other
type, the degenerate case obviously; but there are lots of reasons why we
might want to treat them with the same consideration as anything else.

However... there is the problem of classification. If we argue that a 2nd's
place mx method ought to be a method just like all the other nths/2nds
swaps, then we'd presumably want it to be the same *type* of method. Indeed,
if you ring a lead of Bristol Major with a 2nd's place call to bring it
round, it'll feel just like you're ringing a lead of Bristol Major (and of
course xyz would argue it is exactly the same thing, being the same grid).
The treble ringer will think they have trebled, the inside ringers will
think they have rung something very Bristol-ish, albeit only to one lead.
But there is no way we can define this method as a Surprise method with a
treble-bobbing hunt bell. At least, I'm pretty sure there isn't - most
likely, it will look like a differential principle on paper. So already the
chasm opens up between how we might want to describe a method practically,
and how we are able to classify it.

Faced with this, we could throw in the towel and decide not to allow
single-lead methods after all. But is that a good enough reason? Just
because it is difficult to reconcile practice and theory? I think probably
not.

What I imagine there must be, is some underlying concept that makes
single-lead methods, despite all the arguments we can use for them, and the
symmetry they bring to leadhead swaps, useless to us.

Personally I think that underlying concept might be "structure": does not
the very word "method" implies some internal structure? A method is not just
a random round-block of place notation: it must have some internal,
repeating characteristic to be called a method. Otherwise, it has no method.
Perhaps.

I think this is the crucial point, although I am not actually sold on it
either way. But I definitely believe if we could answer that question, we'd
have a much better idea about what a method actually is.

> Yes, we do spend time unpicking the CC decisions, but I disagree
> entirely with Mark's statement. The MC, far from being crap is made up
> of extremely knowledgeable and competent individuals, and the decisions
> are not a load of rubbish.

Yes, agree entirely Graham. I think we should not be wary of disagreeing
with those Decisions, though, at any level. There is a danger, in some of
the technical discussions on this list, that we assume the Decisions are
universal law.

MBD