Ring the bells, sort of.

Stephanie Strickland is a wonder of compelling poetic investigations. Experiencing her works–try “slippingglimpse” for a quick fix–is only slightly less exciting than having coffee with her.

In either setting, she’ll offer a series of interconnections between things that appear to have no interconnection, so that rising from the table after the event places you thick into edges of a spider web, one that has been woven around your table and one that has now trapped you, sticky and glued, to its gossamer edges.

And so, here, Strickland delights us by doing the opposite of what she normally does. Here, the web is woven, bright and clear, but it catches only itself in its word glue.

Sure, this #Authorfail is three times the length of the guidelines, but if anyone deserves the right to fail for so long, it’s Strickland.

See you next week, and the weeks of this column are rapidly waning.

***

**Proto-Collaboration with Potential Collaborator: A Digital Poem In Prospect **

**Stephanie:** I want to make a work of e-lit based on the mathematics of jingling (bell change-ringing).

**Potential Collaborator (Jeremy Douglass): ** I’m unclear right now on whether you are interested in authoring a specific work of e-literature based on the mathematics of change-ringing, or looking to produce a kind of change-ringing media player within which many works might be authored.

**Stephanie:** I want to make a specific work of e-lit, but my sense is that programming the system for it would be in some sense equivalent to programming the “machine” or media player for it. I want to explore various kinds of historic “changes,” to see which would really work as a *literary* project. That’s not something I can know ahead of time without playing with it first. I find it is not so useful to explicitly define a whole project from the top, and so there is always a kind of negotiation going on with the programming, which in the best cases is a kind of back-and-forth. I envision a textual instrument on which many works might be authored and played.

** Notices of the AMS:** The topic for this month’s cover was taken from the book

*The Mathematics of Juggling.*Chapter 6 in the book is about bell ringing. The object of this principally English pastime is to cycle though all the permutations of a certain number of bells, following strict rules that effectively force the ringers to trace a Hamiltonian cycle in a Cayley graph associated to the permutation group S

_{n}. For 4 bells the Cayley graph can be drawn on the edges of a truncated octahedron. The figure at the right is the score of the method, telling each of the bell ringers what his timing is in each change. I have left out the last change, which is the same as the first. The names of sequences of changes are extremely attractive, I suppose going back for centuries. In doing the cover, I was torn between ‘Plain Bob’, ‘Canterbury’, ‘St. Nicholas’, or ‘Single Court’. The version of Plain Bob illustrated on the cover is more correctly known as Plain Bob Minimus, to distinguish it from analogous sequences with more bells. For more on the mathematics of bell ringing, look at the article

*Ringing the Changes*(

*Math. Proc. Cambridge Phil. Soc*. 94, 1983) by Arthur T. White and also this article. My thanks to Alexander Holroyd for expert help. —Bill Casselman, Covers Editor

**Jeremy:** Thanks for forwarding me this material. I believe I understand the tablature now. I was not able however to understand while examining the cover how that mapped onto the surface of the polyhedron, and reading the cover description didn’t clear this up for me. What does each point and each edge on the polyhedron represent, is there a time component, and why are there 4 bells in the tablature but only 3 edge-types colored on the solid?

Have you looked at the Wikipedia article on Change Ringing Software? I’ve only glanced over it, but it seems like some parts of the authoring and testing tools you want might already be available—and if we were looking to implement a display system for digital art / e-lit presentation, some of these tools might have an open source code base that could be borrowed / repurposed to drive the logic, rather than starting from first principles.

**Stephanie:** Being able to switch from geometric to graphical to sonic output within the tool would be very helpful—and the sonic need not mirror the graphic (which could be a line of text, whereas the sonic could actually be a bell or a whale sound or whatever). I also like the idea of walking geometric edges to explore the site-swaps or transition-paths. I think all of these interfaces give a lot of room to play with site-design.

Digital lit seems particularly suited to explore not the vertices, so to speak, but swaps=transitions=walks. It is hard to find a language between math, programming, poetry, and working vocabulary where everything means the same to everyone, for the moment, at least!

For me, the difference between playing a tune (line of melody, line of poem perhaps) and playing ‘the changes’ seems like scaling up a notch—in other words you are playing the permutations, but in such a manner that sensory reception is still very possible (sonic, as in the ancient jingling practice or visual reception of the graph, or braid, of the score). Note: I’d have to go back to the articles to answer your questions about the polygons.

**Jeremy:** I’ve also read the articles on the mathematics of change ringing… I now understand the relationship between “site swap” transitions and the way that the edges of solids (e.g. the truncated octahedron) forms a representation of the transition path.

If each vertex represents a unique order of bells, and each edge represents a single “site swap” transition operation, then a successful “true” ringing of the changes is a walk along the edges that touches each vertex once without ever repeating.

At this point . . . I don’t understand the math well enough to try to *solve* for such true walks in arbitrary cases, but I could certainly already implement representations (geometric, graphical, or sonic) of extant examples such as Plain Bob, Reverse Canterbury, etc. It looks like there is some specific literature on a general solution at least for cases of 4x bells (4 8 12 16 etc).

I agree that this permutation system (and its attendant aesthetic notions, such as “trueness”) seems promising for computational poetry. I’d be interested in hearing more about how you envision mapping language into this space—for example, would the moment of each bell note map against a word, a line, a lexia? Also, given that the juggling or change-ringing metaphor is fixed and progressive, would the visualization of such a piece be linear and immutable (like a musical score or a film), or would there be some interactive or configurative component?”

**Stephanie:** My thought was to have a system I/we/collaborators/others could use to explore poem-space or word-space or syllable-space or language-space. When you say you can already implement *representations*, does that equate to a little tool one can use to explore ______? That blank should be tone at a minimum, so we could hear the Plain Bob, Reverse Canterbury, and so on as rung on bells. But beyond that, the question would become what is a ‘bell’? And what is a ‘vertex’ if it is a given order of the n bells? Does a bell need to be ‘bigger’ than a ‘word’? If we could do 12 or 16 bells, maybe ‘word’ would work. Reverse Canterbury as a new sort of ‘triolet’ is the way I kind of intuit it, though, in general ‘lines’ or ‘word cluster/sequences’ might be the more appropriate ‘bell’. I would want to be able to play any of the traditional changes by setting parameters (at the interface), but also be able to attempt new patterns similarly.

Once the system was encoded (or so I was thinking), one could explore exactly how many bells/changes made what kind of language mapping interesting. Is there a ‘golden’ number that seems to work with natural language? Just for a start there seems to be a kind of sestina mentality in it all—

And, to be clear then, a vertex represents say 641325, and walking to another vertex gets you to 461325 and so on, yes? And you walk around the edges till you would get back to the first vertex, right? [It does seem to raise the question of whether there is such a walk in all cases, but in the historic named cases we know there is, right?]

I’m not sure we need a *general* solution. I would be happy to explore the present patterns, the historically given and rung patterns as being pre-vetted, so to speak; being of both bodily and mathematical abstract interest.

However, a recent useful chapter on change-ringing makes it clear that we don’t want to handle more than 7 bells (with perhaps an 8th held constant), as 7! =5040 permutations. The chapter clearly explains the three meanings of peal—want we want are peals, perhaps touches (abbreviated peals); not tunes or melodies. The term peal seems to work as a search term, whereas specific peal names don’t seem to (Grandsire Triples). By the way, to play 5040 permutations takes 3 hours.

*CODA: We abandoned the project, not for lack of interest, but for lack **of world and time. And, now, one can search Grandsire Triples successfully. *

###

**Stephanie Strickland’s** recent book, * **Zone : Zero** *(book + CD), includes two interactive digital poems. Her poetry generator written with Nick Montfort, * **Sea and Spar Between**, *appeared in *Dear Navigator, *a new journal of electronic literature from the School of the Art Institute of Chicago. Two of her collaborative digital pieces appear in the recently published Electronic Literature Collection/2.

Last week #AuthorFail 11: Roxane Gay

Next week #AuthorFail 13: Debra Di Blasi

Submit to this space.

It’s great to see the planning stage of a digital poem here…what an AuthorFail!