Ben Leeds Carson, with C++ programming by Ian Saxton | Versions 3.0 - 3.4
Abstract [2013]
This software is grounded, in part, in concepts emerging from my perception research published in the Journal of New Music Research (36/4, December 2007). That research yielded data about pulse confidence under a variety of non-metrical experiences of rhythm. If you’d like to learn more, visit my resource page for this paper (and one other).
The software, in turn, visualizes that data—and hypothetical transformations of it—as though each perceptual possibility were a particle or a celestial body, in some gravitational relationship to the others. Finally, user-generated theories of pulse confidence contribute the composition of rhythms, which can be reshaped in a uniquely flexible multi-track MIDI editor.
In advance of presentations, I often ask seminarians to read articles like Klemp, K., Ray McDermott, Jason Raley, Matthew Thibeault, Kimberly Powell, & Daniel Levitin, “Plans, Takes, and Mis-takes” in Critical Social Studies, vol. 1/2008 (4-21); and/or Bipsham, John, “Rhythm in Music: What is it? Who has it? And Why?” in Music Perception 24/2, December 2006 (125-134).
Unpulser: an abbreviated manual
You’re reading about an application designed to help us think about variables in the study of additive rhythm, and about the streams, layers, and other types of groups that can complicate an experience of additive rhythm. Additive rhythm is rhythm whose timespans are made of multiples of a basic (usually small) unit of time, that serves as a micropulse or a beat subdivision.
(So, imagine a piece of music in 6|8 time. If it has notes* falling only on the meter’s 8th-note beats, it will be additive in the sense that its timespans—i.e. all the distances between those noeses—must be products of an indivisible 8th note.)
Additive rhythms can be studied in terms of how ratios formed by timespan pairs help listeners to infer a pulse, or reinforce their confidence in future time intervals confirming that pulse. For example, a rhythm made mostly of quarter notes and half notes will sound out an abundance of homogeneity in timespan-pair ratios—even if the pattern of alternation between the two values is unpredictable. Simple ratios—2:1 and 1:1—will characterize the relationships between successive timespans.
However, basic mathematical assessments of simplicity fail to account for the way we experience some rhythms made of those pulses. First: in rhythms that eschew regular accent patterns, and avoid reënforcing periodicities, one pair of (successive) timespans and another might be mathematically dissimilar, but produce similar temporal effects. A 12:4 succession (forming ratio of 3:1 in rhythm) seems simple, normative, and suggestive of a pulse structure. An 11:4 succession, meanwhile, feels unfamiliar and doesn’t evoke any class of common dancing or elocution styles. But in circumstances where the underlying pulse is not reinforced by other means, a listener might struggle to put 11:4 and 12:4 in separate camps. This is not a trivial conflation in reference to compositional practices that attempt to imagine and deploy rhythmic experience outside the constraints of meter.
Second: rhythm is complicated by identity; to know whether a rhythm has one quality or another, we first have to choose which events are part of a succession, and which are not. (Even a ticking clock is not “like clockwork” until we are sure that the tics and tocs are more similar to each other than they are to any other sounds in the room.) In this software, group identity (and particularly the way we connect streams of events) is heard and understood in contexts of multi-dimensional streaming—the separation of musical event-collections into streams on the basis of different and independent dimensions.
(So, in the previous imaginary example, there might be 12 notes. Notes 3, 7, and 11 might be accented, and form a distinct stream based on the relative intensity afforded to them. But all 12 notes could be distinguished between a low and a high group. Low and high identities might form streams that include members of the accented stream, as well as non-members. Multiple dimensions—including both pitch-height and of accent—therefore potentially form streams in the music, and therefore also potentially form separate perceived rhythms.)
And that is where the work of the software begins.
To get started with the software, you need Macintosh OS 10.2 or later. (Sorry, no version yet for MSWindows.) The software doesn’t need to be installed; just unstuff it, drag it to your desktop or applications folder, and get started. If you’re ready to start, just download it below. You can also follow along some of this dialogue without using the software.
DOWNLOAD UNPULSER 3.4 [7.3 MB zip download]
***
About Unpulser
This program has two main areas, which make two kinds of files: grammars, and compositions.
The grammar window lets you define rhythmic complexity, using a variety of variables. It then invites you to set boundary conditions for them, so that you can later generate strings of notes that are either complex or simple, by your definition.
The composition area includes a string generator that lets you build rhythms from your grammar, place them in compositions, and then develop, transform, repeat, and vary them. This second area also allows you to look at any composition through three differently functioning windows, allowing you to control and filter the layers or dimensions in a composition from contrasting perspectives. The composition area can be tool for viewing an in-progress composition, or for viewing a MIDI file that you import from another source.
Why these two areas?
UNPULSER’s two main areas—address the pulsedness of rhythm and the issue of layers, streams, or voices—don’t appear closely related on first glance. Why pair the issue of “streaming” with this issue of “rhythmic complexity”?
These two features are put together to accommodate what many composers discover when their goal is rhythmic complexity or rhythmic freedom: rhythm doesn’t exist until you first decide how streams of events to connect (or don’t connect) through time. Consider that all rhythms are lines within textures, and lines are only meaningfully distinguished from their surroundings by some combination of dimensional attributes. (Think of our “clockwork” example above.)
Finally, consider that some of the most exciting rhythms, in your favorite music, are actually found in motoric, repetitive streams of eighth- or sixteenth-notes. You could describe the entire rhythm of the Agitato of Paganini’s Capriccio No. 5, or Bach’s C major Prelude No. 1 from the Well-tempered Clavier, with the string {1:1:1:1:1:1…}, where “1” is the value of the smallest subdivision. But in many other ways these two pieces are not really similar, rhythmically. One factor that distinguishes these works and their senses of rhythm is the differing ways that high notes and low notes, or loud notes and soft ones, separate into subsets, or streams. While it is possible to imagine a rhythm that has no layers or divergent streams, real-life musical experiences of rhythm almost always hinge on the way we attend special, and continuous, threads of events within a whole texture.
[ Proceed to Getting Started : Task #1 — grammar, or Task #2 — rhythm generation. ]