Integrative Levels Classification project scheme monograph references

« Concepts

Division of concepts

A concept usually identifies a class of phenomena sharing some characteristics, like the class of earthquakes, the class of even numbers, or the class of cats. Thus the process of classing consists in identifying a particular object of knowledge as a member of one among these classes, eg identifying the event occurred in San Francisco on April 18, 1906 as an earthquake, or identifying the animal walking in the room where I am writing as a cat.

In this way, we produce a list of concepts, which is the first step of a classification. We may claim that the world is made of water, air, earth, and fire, which will be the main classes of our system. At the beginning, they can be listed in no particular order, ie can be an unordered set. However, especially in case their number is not small, it is very useful to list them in some systematic order, eg from bottom up: water, earth, fire, air; or by increasing weight: fire, air, water, earth. This ordered set (sequence or tuple in mathematical terms) is called a "helpful sequence" by Ranganathan [Ran67], who sees it as the essential feature of classification. Ordered lists of classes resulting from the subdivision of a bigger class (in this case the whole world) are also known as arrays.

A systematic sorting, as opposed to a random one, is very useful to browse lists of concepts. Still, as we collect more and more concepts, and their list gets very long, it can become uncomfortable to be managed by our cognitive abilities. Indeed, some studies [Bla80] find that people tend to browse lists of documents only until about the 30th item, a futility point after which they give up or look for a different search strategy. It is perhaps not by chance that alphabets are formed by a number of signs not very different from 30, as this is a natural size limit for our grouping faculties.

On the other hand, once we have collected a large number of concepts, we will probably notice some affinities between some of them. This means that we can identify more general classes, each divisible into more specific subclasses. Cats, elephants, lizards, and trouts can be grouped into the general class of animals; water can be divided into the subclasses of fresh water and salt water.

By proceeding in this way, we obtain the most classical kind of classification, having the form of a hierarchical tree. Each of its branchings is the division of a class into more specific subclasses. As we move to smaller and smaller branches, we reach concepts of increasing specificity — from animals to domestic cats. We can then distinguish several degrees of specificity, like the order, the family, the genus and the species of animals (we will call them degrees rather than levels, as often done in common talking, in order to reserve levels for another important meaning [Gno10]). In this tree-like structure, we can also consider a series of concepts of increasing specificity, like: animals, mammals, carnivores, felids, cats, domestic cats; in technical terms this is called a chain.

Morphology and phylogeny »

References cited in this section

Bla80: Searching biases in large interactive document retrieval systems / DC Blair = Journal of the American society for information science. 17 : 1980. 3. P 271-277

Ran67: Prolegomena to library classification. Part F # ed 3 / SR Ranganathan ; assisted by MA Gopinath – Sarada Ranganathan endowment for library science : Bangalore : 1967 || (DLIST) – <> : 2007-


Integrative Levels Classification. The ILC project. Division of concepts / Claudio Gnoli – ISKO Italy : <> : 2008.11.07 - 2011.07.29 -

  Integrative Levels Classification project scheme monograph references