Many important books actually cross disciplinary boundaries, and are assigned to e.g. philosophy or literature just by convention. Even worse, books can cover several disciplines at the same time: for example, essays on bridges often include aspects of architecture, engineering, physics and mathematics. When a librarian assigns a shelf mark to it, she has to decide which of these disciplines prevails in the
book content, but in this way she is also failing to represent the aspects of the remaining disciplines. Is this the only option? Actually, the alternative is to directly name the phenomena (that is, any entities and their properties) the book is about, like bridges, and ignore the various disciplines by which they are approached. To this purpose, we need a phenomenon-based classification, listing all kinds of
phenomena such as stars, plants and bridges instead of disciplines such as astronomy, botany and engineering. This way was explored in 1906 by J.D. Brown in his Subject Classification, and studied in more detail in the 1960s by members of the Classification Research Group. Research in this direction has resumed in the 2000s, leading to the early versions of the Basic Concepts Classification (BCC) and the Integrative Levels Classification (ILC).

In these systems, main classes are phenomena, often ordered by levels of increasing organization such as atoms, chemical substances, cells, organisms, minds, societies, institutions, artworks and so on. Each class is subdivided into a common hierarchical tree (organisms are divided into amoebae, fungi, plants, animals, etc.; animals in turn are divided into molluscs, annellids, arthropods,
and so on). Classes can have their own facets (e.g. female, male) and can freely combine with any other classes by means of relationships. Of special interest to researchers often are causal relationships, e.g. bridges being affected by animals, or animals being enabled to cross highways by bridges. Disciplinary classifications do not allow such free combinations, as you are either within zoology or within engineering.

Although minoritarian, phenomenon-based classifications are gaining increasing interest, as they allow users to search for the interactions of a phenomenon with any other phenomena in a newly interdisciplinary perspective. They are promising tools for education programs and school libraries, as young people more spontaneously think in terms of “things” than in terms of disciplines. They agree with
the open perspective of the Semantic Web, as data labeled by phenomena could be exchanged without the barriers of domain-specific schemes.

All these advantages suggest that research on phenomenon-based classification, development of such knowledge organization systems and their comparison with traditional schemes should be encouraged.

See Claudio Gnoli – Richard Smiraglia – Rick Szostak, Phenomenon-based classification: An Annual Review of Information Science and Technology (ARIST) paper, Journal of the Association for Information Science and Technology, https://asistdl.onlinelibrary.wiley.com/doi/pdf/10.1002/asi.24865