Relation (history of concept)

From Wikipedia, the free encyclopedia
Jump to navigation Jump to search

The concept of relation as a term used in general philosophy has a long and complicated history. One of the interests for the Greek philosophers lay in the number of ways in which a particular thing might be described, and the establishment of a relation between one thing and another was one of these. A second interest lay in the difference between these relations and the things themselves. This was to culminate in the view that the things in themselves could not be known except through their relations. Debates similar to these continue into modern philosophy and include further investigations into types of relation and whether relations exist only in the mind or the real world or both.

An understanding of types of relation is important to an understanding of relations between many things including those between people, communities and the wider world. Most of these are complex relations but of the simpler, analytical relations out of which they are formed there are generally held to be three types, although opinion on the number may differ. The three types are spatial relations which include geometry and number, relations of cause and effect, and the classificatory relations of similarity and difference that underlie knowledge. Going by different names in the sciences,[1] mathematics,[2] and the arts [3] they can be thought of as three large families and it is the history of these that will be dealt with here.

Ancient Greeks[edit]

Traditionally the history of the concept of relation begins with Aristotle and his concept of relative terms. In Metaphysics he states: "Things are called relative as the double to the half... as that which can act to that which can be acted upon... and as the knowable to knowledge".[4] It has been argued that the content of these three types can be traced back to the Eleatic Dilemmas, a series of puzzles through which the world can be explained in totally opposite ways, for example things can be both one and many, both moving and stationary and both like and unlike one another.[5]

For Aristotle relation was one of ten distinct kinds of predicate (Gk. kategorien) which list the range of things that can be said about any particular subject: "...each signifies either substance or quantity or quality or relation or where or when or being-in-a-position or having or acting or being acted upon".[6] Subjects and predicates were combined together to form simple propositions. These were later redefined as "categorical" propositions in order to distinguish them from two other types of proposition, the disjunctive and the hypothetical, identified a little later by Chrysippus.[7]

An alternative strand of thought at the time was that relation was more than just one of ten equal categories. A fundamental opposition was developing between substance and relation.[8] Plato in Theaetetus had noted that "some say all things are said to be relative" [9] and Speusippus, his nephew and successor at the Academy maintained the view that "... a thing cannot be known apart from the knowledge of other things, for to know what a thing is, we must know how it differs from other things".[10]

Plotinus in third century Alexandria reduced Aristotle's categories to five: substance, relation, quantity, motion and quality.[11]:VI.3.3, VI.3.21 He gave further emphasis to the distinction between substance and relation and stated that there were grounds for the latter three: quantity, motion and quality to be considered as relations. Moreover, these latter three categories were posterior to the Eleatic categories, namely unity/plurality; motion/stability and identity/difference concepts that Plotinus called "the hearth of reality".[11]:V.1.4

Plotinus liked to picture relations as lines linking elements, but in a process of abstraction our minds tend to ignore the lines "and think only of their terminals".[11]:VI.5.5 His pupil and biographer, Porphyry, developed a tree analogy picturing the relations of knowledge as a tree branching from the highest genera down through intermediate species to the individuals themselves.[11]:V.3.10, V.6.1

Scholasticism to the Enlightenment[edit]

The opposition between substance and relation was given a theological perspective in the Christian era. Basil in the Eastern church suggested that an understanding of the Trinity lay more in understanding the types of relation existing between the three members of the Godhead than in the nature of the Persons themselves.[12] Thomas Aquinas in the Western church noted that in God "relations are real",[13]:52 and, echoing Aristotle, claimed that there were indeed three type of relation which give a natural order to the world. These were quantity, as in double and half; activity, as in acting and being acted upon; and understanding, through the qualitative concepts of genus and species.[13][page needed] "Some have said that relation is not a reality but only an idea. But this is plainly seen to be false from the very fact that things themselves have a mutual natural order and relation... There are three conditions that make a relation to be real or logical ..." [14]

The end of the Scholastic period marked the beginning of a decline in the pre-eminence of the classificatory relation as a way of explaining the world. Science was now in the ascendant and with it scientific reason and the relation of cause and effect. In Britain, John Locke, influenced by Isaac Newton and the laws of motion, developed a similar mechanistic view of the human mind. Following Hobbes's notion of "trains of thought"[15] where one idea naturally follows another in the mind, Locke developed further the concept of knowledge as the perception of relations between ideas.[16] These relations included mathematical relations, scientific relations such as co-existence and succession, and the relations of identity and difference.

It was left to the Scottish philosopher David Hume to reduce these kinds of mental association to three: "To me there appears to be only three principles of connexion among ideas namely Resemblance, Contiguity in time or place, and Cause or Effect".[17]

The problem which troubled Hume of being able to establish the reality of relations from experience, in particular the relation of cause and effect, was solved in another way by Immanuel Kant who took the view that our knowledge is only partly derived from the external world. Part of our knowledge he argued must be due to the modifying nature of our own minds which imposes on perception not only the forms of space and time but also the categories of relation which he understood to be a priori concepts contained within the understanding. Of these he famously said: "Everything in our knowledge... contains nothing but mere relations".[18]:87

Kant took a more analytical view of the concept of relation and his categories of relation were three namely, community, causality and inherence.[18]:113 These can be compared with Hume's three kinds of association in that, firstly, community depicts elements conjoined in time and space, secondly causality compares directly with cause and effect, and thirdly inherence implies the relation of a quality to its subject and plays an essential part in any consideration of the concept of resemblance. Preceding the table of categories in the Critique of Pure Reason is the table of judgements and here, under the heading of relation, are the three types of syllogism namely the disjunctive, the hypothetical and the categorical,[18]:107, 113 developed as we have seen through Aristotle, Chryssipus and the logicians of the Middle Ages.[19] Schopenhauer raised objections to the term Community and the term disjunction, as a relation, can be usefully substituted for the more complex concept of community.[20] G.W.F.Hegel also referred to three types of proposition but in Hegel the categories of relation which for Kant were "subjective mental processes" have now become "objective ontological entities".[21]

Modern developments in philosophy[edit]

C.S. Peirce in America recorded that his own categories of relation grew originally out of a study of Kant. He introduced three metaphysical categories which pervaded his philosophy, and these were ordered through a consideration of the development of our mental processes:

  • Firstness. "The first is predominant in feeling... the whole content of consciousness is made up of qualities of feeling as truly as the whole of space is made up of points or the whole of time by instants".[22]:149–159 Consciousness in a sense arises through the gradual disjunction of what was once whole. Elements appear to be monadic in character and are represented as points in space and time.
  • Secondness. The idea of secondness "is predominant in the ideas of causation" coming into being as "an action and reaction" between ourselves and some other, or between ourselves and a stimulus.[22]:159–163 It is essentially dyadic in character and in some versions of symbolic logic is represented by an arrow.[23]
  • Thirdness. "Ideas in which thirdness predominates include the idea of a sign or representation... For example, a picture signifies by similarity".[24] This type of relation is essentially triadic in nature and is represented in Peirce's logic as a brace or bracket.[25]

These categories of relation appeared in Peirce's logic of relatives and followed earlier work undertaken by the mathematician Augustus De Morgan at Cambridge who had introduced the notion of relation into formal logic in 1849. Among the philosophers who followed may be mentioned T.H Green in England who took the view that all reality lies in relations and William James in America who, emphasising the concept of relation, pictured the world as a "concatenated unity" with some parts joined and other parts disjoined.[26]

Bertrand Russell, in 1921, reinforced James's view that "... the raw material out of which the world is built up, is not of two sorts, one matter and one mind but that it is designed in different patterns by its interrelations, and that some arrangements may be called mental, while others may be called physical".[27] Wittgenstein, also in 1921, saw the same kinds of relation structuring both the material world and the mental world. While the real world consisted of objects and their relations which combined together to form facts, the mental world consisted of similar subjects and predicates which pictured or described the real world.[28] For Wittgenstein there were three kinds of description (enumeration, function and law) which themselves bear a notable if distant "family resemblance" to the three kinds of relation whose history we have been following.[29]

Also of note at the beginning of the twentieth century were arguments associated with G.E.Moore among others concerning the concept of internal and external relations whereby relations could be seen as either contingent or accidental parts of the definition of a thing.[30]

Logical relations and database theory[edit]

Around the same time that George Boole was developing his system of algebra (1847)[31] so important to computer science Augustus de Morgan was finalising his own logic of relations in a paper entitled Formal Logic published in 1849.[32] The syllogisms of subject-predicate logic were to be replaced in his system by what he was to call transitive relations. Ernst Schröder in The Algebra of Logic consolidated and advanced the work of Boole and Peirce introducing the concept of the reflexive relation,[33] and Alfred Tarski, adhering to Schröder’s notation, introduced further types of relation in his paper On the Calculus of Relations published in 1941 where he said “We may for instance distinguish certain important categories of relation such as symmetric relations, transitive relations, ordering relations, one-many relations… and one-one relations…”.[34] Most contemporary authorities[who?] now recognise three main types of relation in symbolic logic:[35]

  • Symmetric relations: where a relation that holds between a and b also holds between b and a, for example the spatial relation “is parallel to”.
  • Transitive relations: where a premise based on a relation that holds between a and b and between b and c leads to the conclusion that the same relation holds between a and c, for example the causal relation “is an ancestor to”.
  • Reflexive relations: where a relation holds between any object and itself, or where all the members of a set or sub-set share the same stated relation {(a,a), (b,b), (c,c)}, for example the inherent relation “is identical with”.

Compound or complex relations can be built on these three. For example, a relation is called an equivalence relation if it is symmetric, transitive and reflexive.[35]

In database and information retrieval theory the distinction between concepts and relations and the classification of relations into different types was also developing at this time. The “relation operators” for example of J. Farradane published in papers dating back to 1966 included such notions as space/time factors, dimensional relations, causation (or functional dependence) and association.[36]:26 An attempt was made by Perry, Kent and Berry to break down subjects into semantic factors (reminiscent of earlier work by J.Kaiser (1915)[37] and S. R. Ranganathan (1933)[38] in their work on library classification systems) but was abandoned due to the immense complexity of relations involved even for the simplest subjects.[36]:368–369 The advent of computer technology was of revolutionary importance in allowing these general ideas to be taken forward. The earliest models in electronic database managements systems (DBMS), the hierarchical and network models, were superseded in 1970 when E.F.Codd proposed the “relational model”[39] which was then developed and transformed into the database programming language SQL (Structured Query Language). Eugen Wüster, in what he called terminology science, also drew on the important distinction between concepts and relations and, borrowing ideas from David Hume (see above) regarding types of relation, set out his findings in his General Theory of Terminology (1974).[40] Further work on Wüster’s concept systems and concept relations was published in 1994 by Anita Nuopponen[41] and taken up in the preparation of ISO 12620 Data Categories and in the further development of data categories in 1999 by researchers at Brigham Young and Kent Universities in the USA.[42] Relations used in these works include (using Peirce’s headings to group related terms together):

  • Relations of Firstness including Disjunctive, Partitive, Contiguous, Spatial, Geometric and Quantitative relations
  • Relations of Secondness including Causal, Sequential, Operational, Systemic, Interactional, and Functional relations
  • Relations of Thirdness including Inherent, Generic, Identity, Hierarchical, Associative and Qualitative relations

Meanwhile, research into the working of the brain, as an extreme example of a relational database, continues. Roger Sperry in looking at split-brain phenomena in 1972 observed that the right cerebral hemisphere deals with, amongst other things, spatial concepts and the left cerebral hemisphere deals with linguistic concepts.[43] Later research has mapped out functional divisions within the brain, identifying sensory areas, motor cortex areas and associative areas, and has investigated connectivity patterns both laterally between the two hemispheres and from front to back (to the hindbrain and brainstem) to demonstrate how the brain deals with general bodily functions.

Relational mathematics[edit]

Recently another approach to relations and relational methods emerged. It may very roughly be compared with numerical mathematics. Relations are typed and may be used for computation in several application fields. This is best documented in the conference series on Relational Methods in Computer Science, now Relational and Algebraic Methods in Computer Science, as well as in the books


  1. ^ Perreault J. Categories and Relators (International Classification, vol 21, No 4, Frankfurt 1994) pp.189ff where Perreault talks of ordinal relations (e.g. dimensions), determinative relations (e.g. activities) and subsumptive relations (e.g. types and kinds)
  2. ^ Piaget J. Structuralism (tr. Maschler C., Routledge and Kegan Paul, London 1971) pp 24-25 where Jean Piaget refers to the mathematicians, Nicolas Bourbaki, using topological structures (derived from geometrical relations), order structures (from predecessor-successor relations) and algebraic structures (including sets and sub-sets)
  3. ^ Norberg-Schulz C., Intentions in Architecture (Allen & Unwin Ltd., London 1963) Besides such headings as Form (geometrical and spatial relations), Building Task (functional relations) and Semantics (relations of meaning) the book refers to Charles W. Morris's division of semiology into syntactics (formal relation of signs), pragmatics (use and effect of signs) and semantics (the relation between the sign and reality)
  4. ^ Aristotle Metaphysics 1020b; cf De Generatione 333a
  5. ^ Plato Parmenides 129, cf 136
  6. ^ Aristotle Categories in Aristotle's Categories and De Interpretatione (tr. Ackrill J.L., Clarendon Press, Oxford, 1963) Ch.4
  7. ^ Long A. & Sedley D. The Hellenistic Philosophers (Cambridge University Press, Cambridge 1987) p.206
  8. ^ Aristotle Metaphysics 998b ff; cf 1029a & 1070a; Physics 185a; Analytica Posteriora 83b
  9. ^ Long & Sedley, op.cit. p.480
  10. ^ Ex Brit cv Speusippus
  11. ^ a b c d Plotinus Enneads
  12. ^ Thomas Aquinas Summa Theologiae (Blackfriars, 1967) p.30 (note); cf St.Augustine The Trinity (Catholic University of America Press, 1963) p.180
  13. ^ a b Gilby T. St. Thomas Aquinas: Theological Texts (Oxford University Press)
  14. ^ Pegis A.C. Introduction to St. Thomas Aquinas (Random House, 1948) S.T. Q13.A7
  15. ^ Hobbes T. Leviathan (Blackwell, Oxford, 1946) p.13
  16. ^ Locke J. Essay Concerning Human Understanding (J.F.Dove, London, 1828) pp. 367-372; 4.1.2-4.1.7
  17. ^ Hume D. An Enquiry Concerning Human Understanding (Cohen R. (ed.) Bantam, London, 1965) p.57
  18. ^ a b c Kant I. Critique of Pure Reason (tr. Smith N.K., Macmillan, London, 1968)
  19. ^ e.g. Peter of Spain and William of Sherwood
  20. ^ Schopenhauer A. The World as Will and Representation (tr. Payne E., Dover Publications, New York, 1966) p.459
  21. ^ Stace W.T. The Philosophy of Hegel (Macmillan & Co., London, 1924) p.70
  22. ^ a b Peirce C.S. Collected Papers of Charles Sanders Peirce (Hartshorne C. & Weiss P. (eds.) Harvard University Press, 1931) Vol I
  23. ^ Wittgenstein op.cit. pp.23 & 61
  24. ^ Peirce op.cit. pp. 170-176
  25. ^ Peirce op.cit. Vol II, p.159; cf Wittgenstein op.cit. p.123
  26. ^ Passmore J. A Hundred Years of Philosophy, 1957 (Penguin, 1968) pp.58,108, ; cf Peirce op.cit. Vol II, p.267
  27. ^ Russell B. Analysis of Mind (George Allen & Unwin, London 1921) pp.10 & 23
  28. ^ Wittgenstein L. Tractatus Logico Philosophicus, 1921 (Routledge & Kegan Paul, London, 1961) p.37
  29. ^ Wittgenstein L. Philosophical Investigations, 1953 (tr. Anscombe G., Blackwell, Oxford, 1978) p.32 ff
  30. ^ Passmore op.cit. p.207
  31. ^ The Mathematical Analysis of Logic (1847) cf Passmore op.cit. pp.125ff
  32. ^ Passmore op.cit. p.123
  33. ^ Schröder E. Lectures on the Analysis of Logic (1890-1905) cf Passmore op.cit. p.144
  34. ^ Tarski A. On the Calculus of Relations (Journal of Symbolic Logic 6, 1941, p.87) cf Passmore op.cit. p.398
  35. ^ a b e.g. Encyclopædia Britannica (15th edition.) s.v. Logic
  36. ^ a b Foskett A.C. The Subject Approach to Information (Clive Bingley, London, 1971)
  37. ^ Kaiser J. Systematic Indexing (1911)
  38. ^ Ranganathan S.R. Dictionary Catalogue Code (1945) and various editions of his Colon Classification
  39. ^ Codd E.F. The Relational Model for Database Management (Addison-Wesley, Boston, 1990)
  40. ^ Wüster E. Introduction to the General Theory of Terminology (Springer, Vienna, 1979)
  41. ^ Nuopponen A. Concept System for Terminological Analysis (University of Vaasa, 1994)
  42. ^ Melby A.K. & Wright S.E. The CLS Framework (BYU Translation Research Group, BYU, 2000)
  43. ^ Encyclopædia Britannica (15th edition.) s.v. Sperry, Roger
  44. ^ "Archived copy". Archived from the original on 2013-05-12. Retrieved 2015-01-06. 
  45. ^ "Relational mathematics - Logic, categories and sets". Cambridge University Press. 

External links[edit]