Concept in Knowledge Organizatio

Concepts are important in KO because concepts are what is organized by thesauri (or any other kind of semantic tool). Different understandings of the nature have implications for how to construe, evaluate and use semantic tools.

"Concept" is, however, a difficult concept. The literature about concepts is enormous, it is muddled, scattered over many disciplines and connected to deep theoretical and philosophical problems. In such a situation it is important always to remember why we find it necessary to use that concept: What kind of work it has to do for us, and in what way different ways of understanding this concept may facilitate or counteract what we want to accomplish. When we are working with terminology and semantic tools like dictionaries, classification systems and thesauri the concept of concept seems unavoidable and progress in the understanding of concepts may influence the very way we do work in terminology, in Library and Information Science (LIS) and many related fields. One expression of the importance of concepts is given in the following quote:

The biggest challenge in information retrieval is concept identification in a specific domain of interest!

(Soergel et al., 2004).

The study of concepts is closely related to the study of meaning and semantics. Two words with the same 'meaning' is said to represent the same 'concept', and semantics is defined as the study of meaning, why these three terms must refer to the same thing. In spite of this they tend to be discussed separately in different literatures. Robert Fugmann, however, warns not just to consider concepts as word meanings:

"The Idea of Concept
Concept, as the core entity in the information field, is currently widely understood as the meaning of a word, a view that has been adopted from linguistics. But in the information field we must recognize a concept entirely independent of whether or not it happens to have been assigned a lexical expression in natural or technical language. In other words, in our view a concept exists independent of whether natural language has (already) coined for it a lexical unit. In its initial stage of emergence a concept must even be presented in the nonlexical, paraphrasing mode of expression. Many concepts never become lexicalized in natural language and only in classifications through the notations assigned to them." (Fugmann, 2004)

Concepts may or may not be expressed by signs, e.g. linguistic signs such as words. The same concept may however be expressed by different signs (synonymy) and the same sign may be used for expressing different concepts (homonymy or polysemy). The word "school" expresses different concepts such as [place for learning] and [school of thought]. The concept [school of thought] may also be expressed by other words such as "paradigm" or "perspective". Concepts are equal to classes in classification systems. Class 150 in the DDC thus correspond to this systems definition of the concept [psychology].

One important understanding of concepts is expressed in the following quote from computer scientist John F. Sowa:

"Concepts are inventions of the human mind used to construct a model of the world. They package reality into discrete units for further processing, they support powerful mechanisms for doing logic, and they are indispensable for precise, extended chains of reasoning. But concepts and percepts cannot form a perfect model of the world, -- they are abstractions that select features that are important for one purpose, but they ignore details and complexities that may be just as important for some other purpose. " (Sowa, 1984, p. 344).

The importance of Sowa's understanding is the emphasis on purpose: Concepts developed to serve certain kinds of goals and interests may not be suited to serve other goals and interests. This is the pragmatic understanding of concepts and it is opposed to a rationalist understanding as reflecting universally valid models of the world. In other words: According to the pragmatic understanding are concepts relative to perspectives, world views and theories, while this is not the case according to the rationalist understanding.

Studies of how a term has been used cannot help us to decide how we should define it. When we use language and terms, we perform some kind of act, with the intention of accomplishing something. The different meanings of the terms we use are more or less efficient tools to help us accomplish what we want to accomplish. In this way, according to pragmatic philosophers such as Charles Sanders Peirce (1905), the meaning of a term is determined by not just the past, but also the future.

Smith; Ceusters & Temmerman (2005) found that four distinct views of concepts can be distinguished:

They do not consider, however, that these four meanings of concepts need not contradict each other. Concepts may very well be abstractions of kinds, expressed by terms, understood by people and representing units of knowledge. The differences in the understanding of concepts is probably in particular related to different theories of knowledge, which may cross all four fields mentioned above.

Smith; Ceusters & Temmerman (2005) also found that a particular view of concepts - that of the terminologist Eugen Wüster - represents something like "International Standard Bad Philosophy", which is not facilitating the work that needs to be done:

"Wüster’s ideas – which see terminology work as being focused on the classification of concepts in people’s minds – and we argue that they serve still as the basis for a series of influential confusions. We argue further that an ontology based unambiguously, not on concepts, but on the classification of entities in reality can, by removing these confusions" (Smith; Ceusters & Temmerman, 2005).

Smith; Ceusters & Temmerman (2005) exemplify by saying that for example diseases are not just imagined, why they conclude that the concept of diseases is not just a psychological phenomenon (and similar with other concepts). They find that:

"The application of a sound realist ontology to the domain of healthcare can make coding systems both logically more coherent and also more closely compatible with our commonsensical intuitions about the medically salient objects and processes in reality. It can thus not only help in detecting errors in existing coding systems but also, by allowing the formulation of intuitive principles for the creation and maintenance of such systems, help in avoiding similar errors in the future". (Smith; Ceusters & Temmerman, 2005)

While this emphasis on realism is important are the kinds of understandings of concepts that they list not necessarily conflicting. When they say that concepts may be understood as either:

This view is not, without problems. There exist different psychological views of concepts, different linguistic views, different epistemological views and different ontological views. These different views are interrelated in a way that one psychological view match one linguist view which match one epistemological view which match one ontological or metaphysical view. If this was not the case, no of the theories could be reasonable consistent.

The authors discuss whether concepts are inborn or rooted in the experiences of the individual. The view of concepts as inborn is related to cognitivism in psychology, to the linguist theory of Noam Chomsky, to rationalism in epistemology and to Platonism in metaphysics. Correspondently, the theory of concepts as rooted in the experiences of the individual is related to behaviorism in psychology and linguistics, to empiricism in epistemology and to subjective idealism in ontology.

In other words: It is important to base a theory of concept on a realist ontology. Such a project is, however, related to a realist theory of both psychology, linguistics, epistemology and ontology. In order to develop a consistent theory all four aspects must be included. (For a paper on a realist theory of concepts from the psychological point of view, see Mammen, 1994, unfortunately only available in Danish).

The pragmatic view of concepts may be understood as follows: To be in order to pursue goals and to plan future actions must intelligent systems – humans or machines – be able to classify some objects, behaviors or events as equivalent for achieving given goals. Concepts are ways of classifying the world. Any concept/classification may be more or less suited or unsuited helping the system achieving a particular goal. If a human culture, for example, do not conceptualize animals as eatable, that culture have fewer objective possibilities to feet its members compared to other cultures. An intelligent system must be able to form classes of objects that are equivalent in relation to a certain task. A common goal or a common use is the principle that unites the classes and that forms the concepts. This view of concepts is basically the pragmatic view. It is explicitly shared by some researchers in cognitive science and artificial intelligence (AI), such as Michalski (1992, p. 249), but people associated with machine learning and AI have, however, to simplify the nature of concepts, why they in reality may operationalize concepts differently.

Scientific concepts are based on scientific theories and views. The concept of "gold", for example, is seen as a chemical element belonging to the group of coinage metals, based on atomic theory. Whales are seen as mammals, not as fish, based on evolutionary theory. Such scientific concepts may diffuse to the broader society and is mostly used as the basis for classification, thesauri and other semantic tools in LIS. Concepts are in other words generally not based on psychological studies or user studies but on domain specific theories and discoveries. They are related to "paradigms". Isaac Newton changed the meaning of fundamental concepts in physics such as the concept of "mass". This aspect is neglected in much theory about concepts and may form the basis of the critique that Smith; Ceusters & Temmerman (2005) raises against what they call the psychological view.

If we accept that whales should be conceptualized and classified as mammals rather than fish, then the following conditions are all met:

According to realist philosophy the listing should be reversed. First came the whales, then came their reflection in human concepts:

This example demonstrates that what Smith; Ceusters & Temmerman (2005) regard as conflicting views of concepts may be in harmony if viewed from a realist view of both psychology, linguistics, epistemology and ontology. There is, however, an important aspect of their criticism of what they term the psychological view, but which should rather be termed a view related to mentalism, psychologism, subjective idealism and anti-realism: The view that concepts are determined by individual acts of cognition rather than through scientific discoveries.

Concepts are ad hoc: they are defined for specific purposes; they may be generalized beyond their original purposes, but they soon come into conflict with other concepts defined for other purposes. This point is important and not merely a speculative puzzle, but also a major problem in the design of any kind of Knowledge organization system. Concepts are very important for knowledge organization. This importance can be understood from the fact that concepts are often regarded the basic elements in knowledge, implying that knowledge organization basically is organization of concepts. Thesauri, for example are basically displaying semantic relations between a selected set of concepts.

In August of 2006 the International Astronomical Union redefined the term "planet", and classified Pluto along with some asteroids as dwarf planet. In this case, the species concept "planet" was redefined, and consequently the classification system used to classify the heavenly bodies. (We may also say that the concept of "planet" was changed). Not everybody agreed in this change of a species concept, but the example illustrates how species concepts are constructed scientifically in order to serve our understanding of phenomena. A specific way of defining a concept or a species may be more or less well argued and accepted, but the definition cannot be regarded as anything but a part of the scientific construction of representations.

"We think that the compositionality and analyticity objections only have force for those who are in the grips of two dogmas. Concepts do not compose most of the time, and concepts are not perfectly shared. Perhaps Frege is to blame for these dogmas. Twentieth century pragmatism departed from Frege in various respects but, we think, it did not go far enough. The notion of immutable senses in the third realm must be replaced by senses that are sufficiently pliable to guide us through the terrestrial realm." (Prinz & Clark, 2004, p. 62).

Andersen, H., Barker, P. & Chen, X. (1996). Kuhn’s Mature Philosophy of Science and Cognitive Psychology, Philosophical Psychology, 9, 347-363.

Fugmann, R. (2004). Learning the Lessons of the Past. IN: The History and Heritage of Scientific and Technical Information Systems: Proceedings of the 2002 Conference, Chemical Heritage Foundation, eds., W. Boyd Rayward and Mary Ellen Bowden. Medford, NJ: Information Today, 168-181. Available at: http://www.chemheritage.org/events/asist2002/14-fugmann.pdf

Mammen, J. (1994). En realistisk begrebsteori: Om forholdet mellem virksomhedsteorien og den økologiske kognitive psykologi. Pp. 43-58 IN: Jens Mammen & Mariane Hedegaard (Eds.: Virksomhedsteori i udvikling. Århus: Psykologisk Institut, Aarhus Universitet. http://www.db.dk/jni/lifeboat/Concepts/Mammen_realistisk%20begrebsteori_1994.PDF

Michalski, R. S. (1992). Concept learning. IN: Encyclopedia of Artificial Intelligence, Vol. 1-2. Ed by S. C. Shapiro, New York: John Wiley & Sons. (Vol. 1, pp. 249-259).

Smith, B. (2004). Beyond Concepts: Ontology as Reality Representation. Achille Varzi and Laure Vieu (eds.), Formal Ontology and Information Systems. Proceedings of the Third International Conference (FOIS 2004), Amsterdam: IOS Press, 73–84. http://ontology.buffalo.edu/bfo/BeyondConcepts.pdf

Soergel, D., Lauser, B., Liang, A., Fisseha, F., Keizer, J., & Katz, S. (2004). Reenginnering thesauri for new application: the AGROVOC example. Journal of Digital Information,4(4). http://jodi.tamu.edu/Articles/v04/i04/Soergel/

Sowa, J. F. (1984). Conceptual structures. Information Processing in Mind and Machine. Reading, MA.: Addison-Wesley.