An Approach to Systematic Classification of Interpersonal Relationships

Conceived as essential to alternative life styles, social and personal transformation

- / -

[searchable PDF version with more complete tables]


Inter-personal relationships are of the greatest significance to satisfaction with the quality of life in general and day-to-day living in particular. Any discussion of alternative life styles must necessarily consider both their effects on such relationships, and the need to develop a different pattern of relationships to sustain any such alternative life style. Like fish in water, individuals are so deeply embedded in networks of relationships that it appears almost ridiculous to question whether there is a clear understanding of their variety. Their immediacy to perception is so great that any discussion would seem to belabour the obvious.

Yet despite this familiarity, distinctions between relationships are very often ill-defined, leading to considerable misunderstanding and to inter-personal conflict. On the other hand, some types of relationship distinguished carry such a high emotional charge that it is very difficult (fortunately, in some cases) to respond to them or to assess them with any degree of objectivity. This too can lead to considerable suffering and unhappiness. Under such circumstances, subtle modifications to relationship patterns cannot be made in response to alternative life style possibilities. Sublte distinctions are ignored in favour of simplistic use of larger categories - and consequently any such category carries the emotional charge (whether positive or negative) of the most highly charged distinction that could be made within it (if this were admissable). The result is that many people are locked into (if not emprisoned by) simplistic relationship patterns. Appreciation of the more subtle distinctions comes, if at all, with experience often painfully acquired and seldom verbalised (if that is possible) in a form which encourages collective appreciation.

Surprisingly it would appear that there is no systematic classification of inter-personal relationships as such (although much could be inferred from studies in this area). Many have been clearly identified in fiction, but often disappointingly (1). There are many studies of specific kinds of relationship which may develop classification schemes - as is the case with the well-known Indian classic, the Kama Sutra (2). The latter, for example, has played an important role in Indian culture in clarifying understanding of important distinctions in the domain with which it is concerned. It is the non-academic importance of any such scheme which is of significance as a remedy for the problems identified above. But, in another way, this example illustrates the fate of such specialised schemes, since, because of the emotional charge of its subject matter, the work as a whole is labelled pornographic in Western society. In effect what is required is a classification scheme which covers the complete range of inter-personal relationships - of which the physical ones are then only an important subset. Whether or not a relationship is 'socially acceptable' is a qualifier on the nature of the relationship and definitely not a justification for its omission (cf. 'Nihil humanum me alienum puto').

This paper constitutes a tentative step in that direction. Until there is greater collective awareness of the variety of inter-personal relationships and the distinctions between them, the inflexibility associated with the prevalent simplistic distinctions will constitute an important hindrance to smooth social transformation - aside from the significant possibility that such distinctions reinforce an equally simplistic approach to inter-group relationships.


One reason that no systematic classifications have been elaborated is that the question is highly complex, despite (or because of) its very familiarity. A number of dimensions may be selected on which to establish distinctions between relationships - and the result would need a matrix in 4 or more dimensions. This would make it relatively useless as an educational tool. The result must of necessity lend itself to display in tabular form (from which a 'map' could be developed for wider appeal).

The approach here is partly summarised in Annex 1. This shows a comprehensive range of possible 'influences' on a given relationship, or factors in terms of which a relationship could be perceived. A given inter-personal relationship may then be understood as being made up of any combination of these factors (which are distinguished in greater detail in Annex 2). Annex 1 therefore lists the headings of Annex 2, to which it also constitutes a form of index. (The numbered rows in Annex 3 correspond to the numbered items listed in Annex 2. Annexes 2 and 3 should be consulted together.)

In selecting factors for inclusion in Annex 1 the point of departure was to identify the kinds of differences between persons which could determine the nature of their relationship. These have been roughtly grouped under headings. The terms used provisionally in these headings are unsatisfactory as are those used to identify the different factors. This brings up a ma jor problem, namely that if the table is not to be excessively long then aspects of relationships have to be clustered into some more general aspect for which an appropriate term is often lacking - specifically where distinctions are most important. The image that comes to mind is one of a sheet of rubber representing the aspects of possible inter-personal relationship as a field. According to how the rubber is stretched, locally or as a whole (depending on sensitivity to relationships in different domains) new distinctions can usefully be made - but the connotations of available terms are inadequate to the task. Clearly in the light of Annex 1 and 2, further work would lead to the sheet being stretched locally in domains where it has been left unstretched.

Elaborating distinctions

In the development of Annex 2, the differences were explored in four ways:

  1. By assuming a simple 3-level distinction giving the possible degree of difference between the persons as: minor/low (L), medium/middle/significant (M), high/major (H). A simpler approach would have been based on the 2-level distinction (equal, unqueal; or low, high). This was not used in order to maintain the link to the second method (below). A more complex approach could be used (if it was felt useful to 'stretch the rubber' further within a particular category).

  2. Where it seemed appropriate, the degree of difference was made explicit by combining the distinctions of method 1 to give a series: low-low (LL), low-middle (LM), etc. So that, for example, 'low-low' identifies a relationship between two persons, each 'low' on the scale in question. This raises the problem of who determines how a person is assessed on any such scale. Since this scheme would be most valuable if usable by people to guide their own thinking about relationships, such assessment is therefore that of one or both participants in the relationship. Any discrepancy between the two is then itself a factor in their relationship.

  3. Where appropriate, the abstract distinctions of the previous method were replaced by specific relationships (e.g. father-son).

  4. Finally, in the cases where distinctions are more subtle, the degree of relationship was indicated by key words supported by examples. This was used for the relationships influenced by sex-roles.

Types of interaction

The distinctions elaborated above do not of course exhaust the character of the relationship. Some more precise indication is required of the nature of the interaction. The approach employed was to use the series of distinctions refined by E F Haskell on the basis of work in biology and generalised to describe the types of 'coactions' between controller and 'controlled' in social systems and organised systems in general (3).

The interpretation given here to the distinction in the case of irder-personal relationships is that the controller is that person of the two who in some way determines, governs or sets the pattern of the interaction which the other person animates. Unfortunately there are at least two difficulties here.

Although the concept is a very rich and powerful one, it has been formulated within a framework of systems theory and cybernetics and coloured by terminology emphasising mechanistic aspects largely alien to the relationships discussed here. Secondly the terminology needed to counterbalance this emphasis is inadequate. In fact, to illustrate the problem, the corrective influence needs to come from studies of psycho-cultural duality such as that of June Singer on androgyny (4). Both are looking at aspects of a very abstract 2-level distinction (5). Controller and 'controlled' are extremely crude in relation to the richness of the distinctions which incorporate them, such as the Yin-Yang concept from Chinese culture as examined by Singer.

To continue (using the mechanistic terminology) the processes which characterise the controller may, in the interaction with the controlled, be:

  • accelerated or in some way
  • enhanced (indicated by '+' in Annex 3) or
  • unaffected ('O' in Annex 3)
  • decreased ('-' in Annex 3).

The same is true for the controlled. When the combinations of possibilities are cross-tabulated for both, it becomes evident that there are nine and only nine of these qualitatively different 'coactions' (indicated by the columns in Annex 3). Each represents a different manner in which 'rate changes' can be induced by one creator in the other (see Fig. 1). It is tempting to look for an equivalence between positive 'rate changes' and some increment in human development, given that a net positive change constitutes a net permanent benefit from the relationship.

So for an interaction (+, +) both gain and, in the case of two organisms, this may be called symbiosis or mutualism. The relation (+, O) may be illustrated by the case of an older brother (the 'controller') who, without knowing it (O), sets a constructive example (+) to a younger brother (the 'controlled'). This may be called commensalism. (This example is derived from ref. 3, p. 6.)

This approach has the advantage of improving communication in ambiguous situations by the use of operationally constant concepts of coaction. Terms like 'conflict', used commonly and confusingly for (-, +), (-, -), and (+, -) coactions, can be clarified.

There is a weakness in Annex 3 as presented, namely that where the two roles are identified (e.g. father-son), there are two possibilities for each column. in the case of (+, -) for example:

  • (a) father may benefit as controller and son suffer as controlled
  • (b) son may benefit as controller and father suffer as controlled

The (b) case is not covered by the (-, +) case unless the focus is on the benefit or harm to each, and the question of who is controller or controllee is ignored. This is also true of two persons having an equitable (i.e. ++, --, 00) relationship in which neither maintains the dominating role for any length of time.

The columns of Annex 3 can be filled with examples of specific relationships to give meaning to the apparently abstract scheme.

Other relationship qualifiers

Unfortunately, Annex 3 does not complete the task of distinguishing relationships. as might be expected.

At least three additional qualifiers are required on any relationship identified in a column in Annex 3.

  1. Drive, intensity, force activating the relationship. This may also, if required, be identified in terms of whether it is the same intensity for both participants, or different, and to what degree.

  2. Duration of the relationship. There is a difference in the quality of a relationship between one lasting hours or days and one lasting months or years.

  3. Re-activation of the relationship. Again an interaction between two persons may (a) not be repeated, or (b) be self-re-activating with an appropriate periodicity until other factors intervene, or (c) be unstable, namely repeated at increasingly longer (or shorter) intervals until it ceases (or engenders a different relationship), or (d) have ceased, but be influential as an 'old relationship'.

Ideally the examples cited in the column of Annex 3 should indicate this variety wherever appropriate.

It is interesting that the Kama Sutra (3) uses the first two of these qualifiers, together with a third, as the basis for a scheme identifying 729 types of physical interaction (each qualifier being itself broken down into three types for each sex, of which the combination gives 9 in each case). The third qualifier it discerns with great elegance may be euphemistically described as 'configuration'. When generalised to non-physical interactions this may usefully distinguish what is so admirably described in French by such phrases as 'petite nature' and 'grande nature'. Whether it is worth elaborating this notion of configuration or whether its consequence is effectively covered by the columns of Annex 3 merits further consideration.

Influence on relationship of attitude to relationship

A further series of qualifiers of a somewhat different kind is identified in the final group in Annex 2. They include degree of: exclusivity (or possessiveness); voluntary acceptance; vicariousness; and social acceptability. Also included here are relationships characterised by cycles or games, as identified by Eric Berne, author of Games People Play. (It would be interesting to categorise the games he identifies in terms of the nine coaction columns of Annex 3.)

Miscellaneous concluding points (for later amplification)

  1. Each relationship aspect distinguished in Annex 3 may usefully be conceived as a 'window' through which the relationship as a whole may be perceived. The point is that a relationship is partly characterised by whichever aspects are manifest between two persons but it is the totality of such aspects which meaningfully delineates their relationship in its full reality. How the aspects interact with one another in the web which makes up such a whole is a very important question. One way of exploring this is by using tensegrity structures as a guide (6).

  2. The complexification of understanding of relationships implicit in this approach takes the emphasis off any particular aspect of a relationship. This is important as is discussed elsewhere (6). The 1-aspect relationship can usefully be developed into a 3 or more aspect relationship. And only when sufficient aspects are in play will the resultant totality have qualities of permanence and self-reactivation.

  3. The number of sub-categories admissable for distinguishing relationship aspects is a matter of personal preference and sensibility. But the number finally selected in a particular case may have unexpected side-effects in structuring thought about them (7).

  4. The question of whether some of the relationship aspects converge at the more intimate levels or necessarily entrain one another should be explored (cf. the concept of 'frequency entrainments.

  5. It is interesting to see the scheme in terms of which relationship aspects are considered necessary or sufficient for marriage and which are considered as incidental to any such formal bond.

  6. The cultural bias built into the scheme or its use should be examined. Clearly this would specially apply to any notion of social acceptability.

  7. The question of the contextual or environmental influence on relationships also merits discussion.

  8. The question of relationship of a person to a significant object, perhaps representing a person needs consideration.

  9. Haskell uses his coaction cardioid as a basis for a very interesting categorisation of socio-cultural coactions (ref. 3). This could (with a number of reservations) provide clues to render the classification of inter-personal relationships more systematic. Of special interest would be the implications for changes in relationships which he explores in detail for other types of interaction.

  10. It is important to note that, whilst an interaction labelled with a '-' in Annex 3 signals some form of loss or harm to one partner in the relationship, in a larger context this may in fact serve some positive function. 'Parasitism', for example, may be viewed as unsatisfactory but in the larger scheme of things elimination of such interactions would break significant chains in ecosystem food webs. Again Haskell explores the significance of such points (ref. 3). It may well be the case that the sequence of interaction types in Annex 3 (as portrayed in Fig 1) constitutes some form of cycle. For individual to have an appropriate 'relationship balance', possibly he or she should have relationships of all types of interaction

  11. Further to the last point, there would seem to be a way of interrelating the interaction types with the tensegrity structure mentioned in point 1.

  12. Also on the tensegrity possibility, it is interesting that some of the relationships noted in Annex 2 have sub-categories of increasing inter-personal interaction. But the achievement of such 'intimate' levels of relationship, despite the attractive tension they exert, is opposed by various constraints. This detail can be very appropriately represented and explored in tensegrity structures (6).

  13. The whole approach here has focussed on 2-person relationships. Whether and how it could be extended to 3 (or more) person relationships needs consideration.

  14. With regard to the use of any such classification scheme and reasons for encouraging its elaboration, it is perhaps not too naive to suppose that intergovernmental agencies concerned with the population explosion or the quality-of-life in human settlements, should be aware of the rich variety of such relationships. How, for example, can the UN Fund for Population Activities fulfil its mandate without exploring the varieties of human relationships in order to discover how the overwhelming popular interest in the (unmentionable) one which is causing all their problems can be partially transferred to other forms of relationship? These would hopefully be equally captivating to the human being, although of a different nature. More generally a similar argument could be made in relation to any international programmes for human and social development.

  15. By the same token, the link between human relationships and basic human needs should also be explored. The inter-personal needs should be reflected in the possible relationship patterns.


1. Johann Wolfgang von Goethe. Elective Affinities. 1809. [summary]

2. The Kama Sutra of Vatsyayana. Panther, 1963 (translated by Sir Richard Burton, 1883 from a Sanskrit version written between 100-300A.D.). As an illustration of its continuing social significance, Dom Moraes in his introduction writes 'The definite fact is that, once written, the Kama Sutra became an essential part of the reading of thousands of Indians. It was less a textbook on sex than a textbook on conduct: and Kalidasa, the greatest of the Sanskrit poets, quoted from it as such. But then Kalidasa lived in the 5th century AD, and I can remember that young brides in India, only ten years ago, were taught from the Kama Sutra before their weddings'. (p. xv) This may be contrasted with the simplistic operational concepts of the traditional 'brown-paper covered' book made available to teenagers in Western society.

3. E F Haskell. Co-action cardioid. In: Full Circle. Gordon and Breach, 1972, pp. 5-7, 33; 45 and 53 (Current Topics of Contemporary Thought, vol. 8). Originally mentioned in Science, 3 Sept 1948, p. 264

4. June Singer. Androgyny; toward a new theory of sexuality. New York Doubleday, 1978

5. The two approaches reflect the right hemisphere/left hemisphere perspectives (e.g. arts/sciences, yin/yang, etc)

6. Anthony Judge. Transcending duality through tensional integrity. Transnational Associations, 30, 1978, 5. pp. 248-265 [text]

7. Anthony Judge. Representation, comprehension and communication of sets; the role of number. International Classification 5, 1978-9 [text]

Tables in annexes to this document have not yet been scanned.

creative commons license
this work is licenced under a creative commons licence.