In this post, i am revisiting a an essay critiquing an old ‘important book’ from 1961. The warnings in this essay about conceptualizing musical culture, I think, are still relevant today.
[Youngerman, S. (1974). Curt Sachs and his heritage: A critical review of World History of the Dance with a survey of recent studies that perpetuate his ideas. Dance Research Journal, 6(2), 6-19.]
I was recommended a read of Curt Sachs’s Wellsprings of Music a couple of months ago by someone who read it 40 years ago, and remembered it to be interesting, forgetting the environment of discourse surrounding musical-cultural discourse then was completely different from the things we now have come to understand. In an essay, Suzanne Youngerman disseminates (read rips apart) Curt Sach’s approaches in a succinct, wonderful manner. I want to look at some key points from that essay especially with respect to the treatment of music in increasingly computational methods. Youngerman issues warnings for writing about musical cultures in this essay. A lot of the analysis from Sach’s book itself is overtly racially motivated, as seemed to be the theme of the time. Youngerman also brings to light the construction of Sachs’s discourse parallelly with Lomax’s. Our methods to understand ‘musical cultures’ and aesthesis may have changed a lot from the times of cantometrics, but do these changed methods achieve a movement far enough away from the time of eugenics and social darwinism? I think some of this critique can be applied to discrimination-type analyses in today’s music-information related studies.
Youngerman warns: There is one other unfortunate consequence of this study. Lomax
refers to the clusters he uncovers as representing “human subspeciation” (ibid.:232). Whether it was his intention or not, the use of this terminology feeds into racial interpretations of history.
Some fallacies she names:
1. the belief that synchronic data can be used to formulate diachronic theory.
<synchronic: as it exists in one point of time
diachronic: as has evolved over time>
Linearization of historical art-narratives are to a large extent make them sound ‘science-like’ as if creative or cultural production is derived from overt citation, and only one logical pathway survived; although it is as true of evolution as it is of cultural development, that what survives is utterly and completely random.
Third-hand, fourth-hand annotated sources, very broad categories, cultural anachronism are features of computer systems now. Historical treatment of ‘other’ musical cultures is something that most systems spend absolutely no time on, resulting in a lot of glossing over of time periods for all except the western canon.
2. the fallacy of reliance on migration as the explanation for cultural change.
3. the false conceptualization of history as the interaction of a few traditions.
4. the tendency to ignore or to explain away contradictions in lofty theory
(treating data as noise)
Some of this analysis and critique is even more sensitive in today’s time when data paucity on the surface doesn’t appear as a problem, but it actually is.
5. the use of dubious sources of data
Especially when dipping into ‘other cultures’, it is even more vital to ask: WHO in this faraway country is collating data? What are their politics? Do they come from power? In which case are you helping amplify the power discourses that harm many, again, this time with data?
6. the perpetuation of a static conception of culture based on the correlation,
commonly known as correlation is not causation
even neural network based co-occurrence should not be confused with causation, especially when digitally explaining-away cultural practice.
7. poetic obfuscation
We cannot use poetics to cloud analysis when it suits us.
I think this is a handy summary of what to not think when talking about musical ‘culture’ (or even genre).