No internet connection
  1. Home
  2. Philosofolly

General Grammar Lessons from Motivation, Language, Music, and Graphics

By thechickenman
    2021-03-22 03:36:03.007Z

    xpost from my blog that includes footnotes for those interested

    Ken Burke’s A Grammar of Motives is the fourth “grammar” I’ve engaged with that explicitly identified itself as such. The first two were English and French. The third was the grammar of graphics, largely through the R ggplot2 package’s implementation. It only just clicked why the concept of grammar itself is so powerful and why a person might be motivated to learn (or make) one in any creative context they regularly work in. Burke’s “dramatic pentad” for analyzing motives includes an “agent,” in a “scene,” with an “agency,” who “acts,” for a “purpose.” If I was to apply the pentad to my motivation for writing this, I (agent) am in a pandemic (scene) with access to research materials (agency) writing about grammar (act) to better understand how we might teach why it’s useful generally (purpose).

    The purpose of a society-as-agent acting to set up schools is to purify educational experiences in a safe nurturing environment. Further committing to Burke’s language, with school we are sculpting the scenes young agents find themselves in and sanction an agency of specific actions hoping that they come to understand and realize a higher possible self. Structuring school experiences is a double-edged sword. We expose people to ideas and materials they otherwise may not have discovered on their own, but at the cost of asking them to set aside more spontaneous interests. One of the most painful things a learner can say is what are we ever going to use this for, which I admit I may have done when being asked to identify nouns and verbs or diagram subjects and predicates in sentences for the first time. Often in schools understanding the later real-world purpose is murky at best. A grade easily becomes the immediate purpose as a surrogate and gets baked into our formal incentive (or motivational) structures that have social consequences.

    I realize now the first set of rules that might be classified as a “grammar” I ever actually learned never directly identified itself as such. It was embedded in music theory, but it was clear from the beginning what I could use those ideas for. My 8-year-old self was more than willing to sit with books about pitches and their durations, how they were notated, how their harmonies could be analyzed, and how to use that general knowledge to reproduce the Star Wars theme or Für Elise at a piano. Read the notes, press the keys, and there’s an immediate reward if you did it right. A mom and a few grandparents who played a little demonstrated where I might be later in life, and recordings demonstrated just how wide the musical world was. The encouraging scene I found myself in, the tools I had available, and my purpose were all clear. Now I’m in that position later in my life where I can theoretically play anything for enjoyment and I can compose new musical ideas.

    Clarity of purpose seems to be the potential rocket-fuel of motivation. When you know where you’re going and why, there’s a mental focus able to dart between technical problems in material being learned with little unnecessary stress so long as progress is being made. This should not be taken as a panacea for coercing people into learning specific things on an imposed schedule, but there are four approaches to clarifying purpose I can see:

    A teacher is well equipped to explain the heart of how a concept will be useful for future actions

    A student is well equipped to seek out the heart of how a concept will be useful for future actions

    Meaningful future actions using the concept being learned are ideally only a few seconds away

    Teachers and students look to already familiar concepts to relate to what is being learned

    There is also the case where a purpose was already clear coming in. A reason I’ve pointed so many aspiring data scientists to R for Data Science as an introductory book is because data visualization is the first substantive material covered. It’s ideal for learning programming because of how immediately obvious changes in input then map to meaningful output a few seconds later. If software installation is taken care of, even most teenagers are able to work their way through some powerful concepts in a few hours facilitated by a well thought out grammar for working with graphics. The grammar eases the cognitive burden of picking up additional possible aesthetic mappings and carrying them over to additional kinds of plots.

    So is trying to explain the importance of grammar as a general concept early on doable? Can pointing to a pattern in notation and organization strategies across contexts excite as much as the content in any one? Imagine a 13-year-old who is identifying nouns and verbs in a language class, creating data visualizations in a mathematics class (because it’s the resource-rich future and we’ve moved on from single-purpose calculators as the auxiliary tool of choice, so it isn’t as much of a reach as we would imagine it now), analyzing motivations in a history class, and making music in a band class. How coherent of a school experience might be possible if purpose started regularly clicking and motivation was high? Could a well positioned science or philosophy teacher link those things together to light some kind of fire? Music notation was the first graphical display of real world “data” with time on the x-axis and pitch on the y-axis6, long before William Playfair created conventional statistics graphics in the 18th century. He put time on the x-axis and trade balances on the y-axis in one early chart. Isn’t that interesting? Making sentence generators may be a fun use of grammar concepts in the language class, which could turn into silly lyrics for a choir session. The specifics of any one potential cross-over aren’t as important as realizing fruit exists in every case, and most certainly do not require computers or programming. A teacher ideally might investigate where a student’s spontaneous motivations are and would enthusiastically relate them to potential paths of learning (and be granted the agency to provide ad-hoc rails to their pupils with their expertise for taking student motivation into account). A student would ideally recognize how the experiences of going deep into one area they enjoy spontaneously could be used to investigate what makes another area enjoyable for others. They would recognize where they might use the new ideas socially or to create.

    Of course, our modern institutional philosophy of education is impoverished. Bureaucratic paranoia has led to a narrow conception of what school can be and who is trusted with degrees of free agency. In our efforts to squish learning into legible standard experiences that can be represented as data and visualized using a grammar of graphics, a grammar of motives can literarily articulate how that paradigm might be invisibly backfiring at a scale unknown outside the practical lens of its surveillance systems and scope of scientific methods. Our higher purpose in education is to develop a deep understanding, or acquiring equipment for living as Burke would say. In addition to what are we ever going to use this for, the other canary in the coal mine might be will this be on the test. We measure actions (skills or demonstrations of understanding) with tests under the assumption a short run score is a safe indicator we are facilitating skill development and deeper understanding in the long run. When a necessarily limited measure of an action of an agent in a scene surreptitiously transforms also into the psychologically dominant purpose of the action because of the social importance placed on that surrogate measure, we might assign the label surrogation to this grammatical faux-pas. We would want to look for and correct this in the same way we spot double-negatives in some language contexts, parallel 5ths in some musical systems of harmonic motion, and dual axis charts in data visualization. We would have to judge, not calculate, whether some part of our pentad of acts, purposes, scenes, agencies, and agents have become corrupted and consciously reform those elements. We would have to recognize the dangers of optimizing for a few limited measures through institutional or social pressure, which is difficult given our cultural anxieties and quantophrenic inclinations.

    A grammar’s usefulness in a general sense then is to help one adapt gracefully to new challenges and identify common errors in a specific creative context. Being grammatically correct does not guarantee a useful result, but grammar does help us compose ourselves so that we may more precisely articulate problems and compose more adequate solutions. Or beautiful solutions. It is beneficial or beautiful then to learn multiple grammars or dialects with similar purposes inside a single language’s domain (and then other languages). A grammar’s ultimate purpose is to allow us to more consciously or freely spend our limited cognitive resources on higher purposes. As examples:

    To represent an idea in composing a message

    To define musical motions in composing a song

    To see underlying patterns in a data set in composing a graphic

    Motivation can then be thought of as composed in a way conceptually similar to an essay, a song, or a data visualization: roughly, to realize a possibility in composing a motivation. A motivation can be composed for the self or suggested to others. In my interpretation, for either an individual or a society:

    An agent asks who they are and who they want to be

    The scene challenges who they are and suggests who they need to be

    An agency is the structure of what they can do or means available for acting

    A purpose is why they do what they do in pursuit of a possible future

    An act is what they do to affect reality

    Given our dramatically different imposed scene from normal and relatively stable higher purpose, pandemic isolation may have been the ideal time to grant people in education environments the agency to locally experiment with roles and responsibilities16. We should have made it possible to go deep on a passion or new skill if the motivation was there despite the chaos. Our school structures that require juggling prescribed demands may even be limiting; our scientific methods have no way of predicting what would happen in the long run if periods of independent, self-motivated, but institutionally supported deep learning were given a chance to get going on occasion in all of our childhoods and young adulthoods. Extended concentration with minimal distractions in solitude may even be what’s necessary to reach higher orders of progress on a learning curve. To borrow terms from another context, we might conceive of a velocity of learning, an acceleration of learning, and when pressing on long enough while motivated a sudden jerk of learning. Jerk would be where entirely new gears are found to be brought back to other contexts, like what happens when the general power of grammar shifts into place.

    • 5 replies
    1. hazard
        2021-03-23 16:32:37.043Z

        Thinking about grammars makes me think of comments @snav had at some point about "generators of difference". It also makes me think of ontology, but it makes me think of it in a more dynamic way. Less "here are the basic things" and more "here's the way they all interrelate and interact with each other". This feels more conducive to asking better questions.

        There aren't to many grammars I've learned. Though I have noticed a trend of trying to soak up the "type systems" of diff domains. There's lots of chunks of math like this, were I've read a bunch of wikipedia pages and skimmed a few books, and haven't really gotten an understanding I can use, but have gotten a sketch of the type system that let's me at least make well formed questions.

        Also, I love me some fast feedback loops! One reason I did a lot of web dev projects the other month was because I noticed how much I enjoyed being able to see the results of my code right after writing it.

        1. In reply tothechickenman:
          suspendedreason
            2021-03-23 22:08:48.096Z

            I still feel like it'd be useful for us to get a good operational definition of "grammar." I asked Chickenman what how he'd define it, and he said he wasn't sure how to explicitly factor it, but its purpose is "to help one adapt gracefully to new challenges," and to "identify common errors" (perhaps we can extrapolate this to "commonalities" generically?)

            I've also found so much of benefit from musical grammars. It's interesting that schoolhouse grammar still feels mostly pointless to me—maybe because English grammar we pick up naturally, in a more functional way, already? And therefore the formal school instruction is giving either redundant or vestigial teachings.

            Coding grammar is an interesting example because of its rigidity. OTOH, humans are a bit like computers—it doesn't matter what you think you're telling them, what matters is their schematic affordances for interpretation. (In other words, "pragmatic reader-response")

            1. T
              In reply tothechickenman:
              thechickenman
                2021-03-24 10:17:25.492Z2021-03-24 10:25:04.416Z

                Was definitely in the weeds overthinking it. A dictionary definition is fine: "the whole system and structure of a language or of languages in general, usually taken as consisting of syntax and morphology (including inflections) and sometimes also phonology and semantics." Grammar is like a typology of what you'll find in a language game. It's powerful because, when going into a new language system in any domain, if you realize what you're looking for you can relate lessons from previous games.

                When I was first learning the grammar of graphics, I didn't really understand why that approach was going to be so powerful. My purpose was "I want to create date visualizations". How I understood the options available to me as an agent, the utility of the output of base R or Stata or Python were roughly equivalent to what you could get with the grammar implemented with ggplot2. Each tool appeared equivalent. For many they still do because the primary concern is whether you can get an immediate job done, not how it's going to shape your long-term cognitive strategies. But you don't know what you really have with any one until you've been around I suppose.

                1. In reply tothechickenman:
                  suspendedreason
                    2021-03-29 00:53:05.501Z

                    Potentially relevant... Conversation at The Wire about the musical score—the conversation is pretty critical/skeptical, thinks notation has stifled expressive range through bad categories. One of the most controversial pieces the mag has run lately, comment section over-brimming.

                    1. Tthechickenman
                        2021-03-29 05:21:17.551Z

                        I would not be surprised if they had done some plotting of their own to generate conversation and clicks specifically, not make any substantive point about the merit of various music cultures.