Classifications and Ontologies
Technicalities 43(1) Jan/Frb 2023: 11-14
One of the great pleasures of working at UCLA is the fantastic people I get to work with, including staff, students and faculty. Two of them, Jonathan Furner and Rob Montoya, work in the area of classification. Furner was an editor of the Dewey Decimal Classification, and has the broadest knowledge of anyone I have ever met on the topic. Montoya has a new book on classification and bio-diversity out from MIT Press1, and I urge you to read it.
I was really fortunate to take Prof. Furner’s class on classification a coUple of years ago. It gave me a foundation in the topic that I have always been missing. I’ve only taken a few courses in my twenty-five years at UCLA, which is a huge mistake that we professors make. Furner’s class was one of the best classes I have ever taken, and I did not have to do any homework–just the readings.
Classification–both basic classificatory theory and book classification–has always been interesting to me but not something with which I have had a ton of experience. I have had some experience as a student with some of the theory just by reading, but I have never gone through the canon of western philosophy concentrating on the classification components, nor have I really supplemented that reading with important but non-canonical authors who concentrated on the topic, except perhaps those people in library and information science. After all, I think writings oN the concept of the work, like Lubetzky and Svenonius, are on classification, just applied to a particular problem in grouping texts. My real problem is that I have never really gained substantial experience in a classification. I did some copy cataloging as a paraprofessional, so I got exposure to both Dewey and the Library of Congress Classification, but when the time came to actually create or assign class marks as an actual professional librarian, I always seemed to be in a place that did not use standard classifications. At the New York Public Library, we cataloged books by holding them up to a ruler–standard 23mm books were class M, as I recall. And then you would append a year and accession number to that: M20221234. Something like that, I should call and ask if that is an accurate representation of how they still do it at the Schomburg Center. Closed stacks, and the price of Manhattan real estate, along with the fact that you can squeeze 30% more books into a shelf when you organize by size, and I am pretty sure they will be doing it like that for a long time to come.
Why This Classification, And Why Is It Like This?
I was always a little unsettled by classifications beyond the lack of applied experience with them. One reason was because of that lack of professional exposure and practice. When I started teaching at UCLA, I dreaded any question I might get about the LCC–years later, it remains to me an overly complicated and exacting notational system. And when the notation becomes a barrier to understanding the classificatory organization that it is supposed to express, what is its use?
But the other reason I always felt unsettled was because of what I perceived to be arbitrary components in the actual classification. The bias in classification systems has long been noted, and how could anything that originated with the Melvil Dewey not be? Quick perusals of high level perspectives of the DDC, for example, show an almost exclusive preoccupation with Christianity within the religion section; not that other religions are not present, but you have to dive down into the details to find them. But even worse in my experience was the organization of any particular topic like the enumeration of civil rights, or the periodization of Roman history, or what feels like a lack of effort in the representation of literature. To my conspiratorial eyes, those classifications felt capricious, arbitrary and unpredictable. Those are strong accusations in and of themselves. They may also be expressions of bias. Someone organized these topics, and the basic concern is that they view the world differently from not only me, but from many users. Sometimes I felt, perhaps I was looking at the product of expertise that I did not have; other times I felt my perspective just was not being represented.
What I got from that feeling was a conclusion that classifications were not just simply mechanisms for grouping books or helping indexers and searcher find terminological consistency. Instead a classification was a rather complete conceptual model of the world, or alternatively, a series of partitions of conceptual models of the world. By partition, I mean that something like the Dewey Decimal Classification is represented as a single conglomerate classification, but really there are multiple classifications present, one for music, one for poetry, one for legal materials, etc. I know this is literally the case for the Library of Congress Classification and reflects its development as a general organization and notational system that was shared out with various bibliographers and units that comprise the Library of Congress. Relatively little was done to coordinate amongst the units to product a single comprehensive classification to the extent that is possible.
And if a classification is a conceptual model, what I was increasingly feeling was that it was not my conceptual model. As a student, my experience–more felt than expressed–was confusion, that my understanding of the world was incomplete and under development. Later, as I developed expertise, the classifications just felt wrong, along the lines of “that is not how to organize this subject area.” It was only fairly recently that I have come to feel that the classifications expressed a fairly consistent point of view, based on some editorial committee’s expression of what they felt was expert consensus. It could be that at various places that expression was an inelegant compromise, an odd presentation of the consensus, or most interesting but difficult, an alternative conceptual model.
Differences in Models
Over time, the idea of conflicts between alternative conceptual models came to be the most generative explanation for understanding classificatory conflict. When people speak of ideology, world views or belief systems generally, I started to imagine those characterizations as networks of concepts. I know it is a generally reductive move but I also am inclined to see characterizations of culture expressed the same way, in part because it gives some semblance of being able to compare and express difference between cultures. That is, a culture can be understood in part as a system and structure of conceptual elements that express the ideas, values and beliefs contained within that culture. Christmas is an annual event in American culture; it can be modeled as a node in a conceptual system, and that system might be modeled as a classification.
Differences in the belief systems between members in a general shared culture can then be expressed as variances or deformities in classifications of the world. For me, Christmas is a secular holiday with origins in Christian religious practice, evolving over time from its origins as a pagan holiday to become this something that can be expressed as a concept with close associations with consumerism and ritual exchange. For others, that concept retains close associations with their religion; an for others yet again, it is a date to go get Chinese food.2 In a classificatory systems, Christmas is a migratory concept, moving from one place in one classification, to another in second classification, and yet another spot in a third classification. The classifications between the three belief systems are mostly shared and overlap, with minor differences, like the DNA between a chimpanzee and a human.3 That is, the members of a culture share a very large portion of a common conceptual system, in part because so much of what we know is tacit, but a relatively small amount of divergence results in apparently large differences in expressed identity.
What I find particularly interesting though is not that Christmas moves from place to place amongst the members of a cultural subpopulation, but that moves systematically with other concepts. By “moving systematically” I mean knowing the subcomponent of your conceptual model on, say, the Second Amendment, I could also predict similar systematic divergences around Abortion, theRelation of the Individual to the State or Freedoms and Liberties. That is certain deflections and rearrangements of the conceptual system move semi-predictably in concert with each other.
At some point, accumulating enough change in one system transforms it into a completely different system, by which we mean it is a representation of a different culture or milieu. Speculating here, a real change from one cultural representation to another is not a systematic rearrangement of concepts, but rather significantly new conceptual content. That is, different cultures have different things in it: new events, different people, different concepts per se.
Enter Ontology, and Some Thorns
So, “conceptual system” is not quite the right name for this, even though that is the phrase that I have used above as a way at getting at a particular implication of classification. At alternative phrase could be “ontology” but something always sticks out as a thorn in the way “ontology” gets used. Ontology is the study of what exists in the world, and even more particularly, as the study of kinds of things exist in the world. If “geography” is a study of the world, then a particular representation of the world is a “map.” “Ontology”, confusingly, refers both to the study of the things, and a representation of those things:
| Field of study Representation | |
| Geography Map | |
| Ontology Ontology |
It just always just confuses me when someone talks about “an ontology.” There is too much of an assumption of truth wrapped up in the notion, as opposed to the notion of an “ontological representation” or “map.” Or the term that I prefer, which is… “classification.” It helps us understand, I believe, in what a classification really is, which is a representation or assertion about what kinds of things are said to exist in a world, and ultimately expresses a shared common undestanding of what those things based on a consensus shared (with some variation) amongst the members of a culture.
One other thing sticks out thorn-like here too is I think we have a mediating concept between the Field of study and the Representation:
|
Field of study |
Commitment |
Representation |
|
Geography Ontology |
Geographical
commitment Ontological commitment |
Map Ontology |
Ultimately, understanding classifications (or ontologies) is that we know now that we are in a pluralistic space of perspective, rather than a singular space of scientifically verified truth. Classifications originate in a cultural context, at best they are theories about the way the world should be modeled. At worst they are simply one relative representation that has no greater hold on truth or accuracy of representation that any other—there is just difference, and no error.
By understanding classification as the product of certain theories about the world (i.e., ontology) and certain commitments about what should appear in a representation, we begin to understand why we might have multiple classifications over the same discursive field. Our problems then become how might we show similarity and difference amongst classifications, and what are the ethics of inclusion and exclusion of our various commitments and representations? And just to make sure this is clear, again, in a column dedicated to the politics of metadata, any time we use one particular classification over another, we should be asking ourselves “whose perspective is present here?” and why? Who is excluded? What other ways of seeing the world are there that we have not included in our classifications?
Works Cited
1. Montoya, R.D. Power of Position: Classification and the Biodiversity Sciences. MIT Press, 2022.
2. Tuchman, Gaye, and Harry Gene Levine. “New York Jews and Chinese Food: The Social Construction of an Ethnic Pattern.” Journal of Contemporary Ethnography 22, no. 3 (1993): 382-407.
3. “Bonobos Join Chimps as Closest Human Relatives.” Science, 2012, accessed Nov. 13, 2022, https://www.science.org/content/article/bonobos-join-chimps-closest-human-relatives#:~:text=Ever%20since%20researchers%20sequenced%20the,them%20our%20closest%20living%20relatives.