Paper presented at the 3rd Annual Qualitative Methods Conference: "Touch me I'm sick"
8 & 9 September 1997, University of South Africa Regional Office, Durban

The line in the sand has blown away

Kenneth G. Wilson

Department of Psychology, University of the Witwatersrand


"Self" is arguably one of the most cardinal concepts within the discourses of modern psychology. This paper examines some of the competing discourses of self, and concludes that psychology has historically been tied to an understanding of "self-as-entity". This essentialist and individuocentric view of the self does not hold up to critical scrutiny. The deficiencies of "self-as-entity" are discussed, and the more plausible position of "selves-as-social-constructs" is adopted. Traditional understandings of the self as "within", "intimate" and "bounded" are called into question by the works of authors who can be characterised as broadly "social constructionist" in orientation. The paper ends with a discussion of the implications of viewing "selves-as-social-constructs" for the theory and practice of psychological research and psychotherapy.

Who am we?

The "line" referred to in title of this paper is the line that demarcates us as ourselves. It is often taken as axiomatic that we are who we think we are - autonomous, unitary and rational subjects; and that our selves are really entities which "exist" within us allowing us to be who we are. Furthermore, it is often thought to be desirable and advantageous to be who we really are.

Psychology could perhaps be uncontroversially defined as the study of behaviour and experience. This definition, or variations thereof, are often to be found within first year psychology texts. If one were to ask as to whose behaviour and experience is being studied, the answer could be confidently given that we study the behaviour and experience of individuals or persons. It strikes one as not being incorrect to say that the basic unit of analysis within psychology is the individual. The difficulty arises when one tries to tie this conception of "individual" to a particular ontology. The nature of the individual under investigation or treatment by psychology is always based upon some or other theory of the self.

Psychology often seems to be nothing more than a loosely structured group of theories and practices which has at its heart the project of the self. This project, in its many forms, involves activities and discourses relating to the awareness, actualisation, blame, control, determination, discovery, esteem, fulfilment, help, management and regulation of the self. Hoffman (1992) describes "the self" as being one of "the five sacred cows of modern psychology" - the other four being: objective social research, developmental psychology, the emotions and the idea of levels within human events (p.10). Barglow (1994) states that the goal of all "serious therapy" is the transformation of the self, but that this goal is never accomplished, because it seems that "in our culture there always remains more work to be done on the self, if not by means of therapy, then through a special diet or by visiting a new hair stylist" (p. 183).

Structuralist discourses of self

Using the Cartesian self as a starting point, it is possible to delineate certain key features of the traditional view of self. According to Potter and Wetherell (1987) the key assumption behind all traditional models of self "is that the self is an entity and, like any other entity or natural physical object, it can be described definitively and once and for all" (p.95). This essentialist, unitary, rational and bounded self is what shall be described as the structuralist self. There are of course variations of the structuralist self. These are described as follows.

The Cartesian self: Mind, without a body or a world to live in

The roots of all modern conceptions of the self can be encapsulated in the work of Rene Descartes (1596-1650), the French philosopher and mathematician. Descartes can be thought of as the figurehead of individualistic and atomistic views of the self. These views are based on "the ideal of transparency in subjectivity, where a subject can know itself completely, and reaffirms the disavowal of the body central to western philosophy since Plato, where reason and the mind are posited against unreason and the corporeal" (Bakhurst and Sypnowich, 1995: 4).

The Cartesian self can be identified in the following extract from Meditations on First Philosophy (1641)"But what them am I? A thing that thinks. What is that? A thing that doubts, understands, affirms, denies, is willing, is unwilling, and also imagines and has sensory perceptions" Descrates, 1641 cited in Bakhurst and Sypnowich, 1995:2). From this it can be seen that the Cartesian self is a self without a body, and with no social world in which to live. The Cartesian self is built on the idea that "the individual is, above all else, a thinking thing (Bakhurst and Sypnowich, 1995:2)".As thinking things, we inhabit our own subjective worlds, and mental life exists independently of our interaction with others.

The idea of people as thinking things is based on Descartes' mind-body dualism. If the mind and body are separate entities, then it follows that the we, as thinking things, are somehow separate from the world in which we live. The individual-society dichotomy sets us up as fundamentally distinct from the social character of the world. The Cartesian self is thus simultaneously asocial and disembodied.

The self as a "personality"

Trait theory has been incredibly influential in psychology, particularly within the fields of personality and social psychology (Mischel, 1968; Potter and Wetherell, 1987). Within trait theory the self is presented as a "personality". Behaviour is thought to be influenced by the different combinations of traits that people possess. Personality, although being a multi-discursive term generally refers to the distinctive and relatively stable qualities and characteristics of people. The trait theories of psychologists such as Allport (1937), Cattel (1966) and Eysenck (1953) portray the self as a "personality" comprised of traits, which "outweigh the influence of the immediate situation or the context surrounding the person" (Potter and Wetherell, 1987: 96). Trait theory and the view of personality which it presents has come under fire from numerous critics on the grounds that it is asocial and ignores inconsistency in behaviour (cf. Burkitt, 1991; Burr, 1995; Henriques et al, 1984; Leonard, 1984; Mischel, 1968; Potter and Wetherell, 1987; Seve, 1975). The traditional view of personality as existing within people is really just an extension of the Cartesian self. This extension is an attempt to attain the scientistic respectability which is believed to go hand-in-hand with the empiricist projects of personality testing and measurement.

In spite of the fact that theorists such as Burkitt (1991), Leonard (1984) and Seve (1975) have attempted to rehabilitate personality as a useful term, by focussing on the its social formation, it is difficult to imagine the term independently of its essentialist nature. It is for this reason that Burr (1995) argues that personality is not a meaningful way of understanding ourselves (p.30). The abandoning of personality is not exclusively the preserve of social constructionism, as Burr (1995) is aware, noting that the behaviourists and social learning theorists did exactly this some time ago. Burr (1995) proposes "identity" as a term which avoids "the essentialist connotations of personality, and is also an implicitly social term" (p.30).

The self as "role"

Roles are defined sets of qualities and styles of behaviour that are associated with social positions (Potter and Wetherell, 1987: 98). People are viewed as "role-players" - social positions, which are independent of individual subjects, are thus involved in the creation of the self. The "self as role" presents us as social performers whose identity is tied to the social positions which we occupy. This is clearly a social understanding of self, but portrays people as insincere, trying on different roles at will, and possessing a set of potentially discordant identities.

The self as "real/false"

Humanistic views of the self position our "false" selves, the social selves of role theory, against the backdrop of the "true" self. In the same way that the actor plays a part yet has a real self, it is proposed that people have a "double consciousness" whereby they play roles underlain, yet detached, from who they really are. The self as real/false can be clearly identified in the work of Maslow (1968) and Rogers (1961), and informs much of the Humanist psychology.

The relational self

According to Barglow (1994) "the social character of the self has been an axiom of twentieth-century sociology" (p.184). The American philosopher and social psychologist George Herbert Mead (1863-1931) is notable as one of the few people at the start of this century who were concerned with the social formation of personality. Mead viewed mind and self as being formed as a result of the communicative activity of group members ( cf. Burkitt, 1991: 28).

The recognition of humans as "essentially relational creatures...(whose) identity consists not simply in their separateness from others, but in the myriad affiliations that link them with a shared lifeworld" (Barglow, 1994: 184 brackets added). The relational self has a long history, with not least of which, Marx (1845) stating that the essence of each individual is the "ensemble of social relations" (p.29). The relational self, may be either structuralist or post-structuralist in nature.

The structuralist self: obituary

Potter and Wetherell (1987) describe traditional, structuralist views of the self as being untenable because "it is taken for granted that this self, the object to be discovered, is the centre of experience, an initiator of action, a coherent whole, separate from other distinct selves" (p.101). Furthermore each of the outlined structuralist positions of self assume they are dealing with the "real" objects, and do not entertain the possibility that they are in fact "constructions or discursive articulations" (ibid.). These essentialist views of self make it possible for the individual-society dualism, to function as if the individual and society were somehow "natural pairs in a balanced dichotomy" (ibid.).

The historically and culturally relative nature of the structuralist self/ self-as-entity has been discussed by numerous authors. Barglow (1994), Danziger (1994), Henriques et al. (1984), Potter and Wetherell (1987) and Rose (1989, 1996) provide detailed discussion of how the structuralist self/self-as-entity is in fact a social construction, in and of itself.

Post-structuralist discourses of self

The structuralist self (self-as-entity) which offers up to us the unitary rational subject as the object of investigation of psychology can be juxtaposed against a non-essentialist post-structuralist self (selves-as-social constructs). The post-structuralist self is based upon the idea that "psychological models of the self are inevitably culturally and historically contingent, dependent on certain kinds of social practices" (Potter and Wetherell, 1987: 102).

Selves-as-social constructs

If we return to the question of whose behaviour and experience is to be studied by psychology, the structuralist self tells us that it is the individual. This conception of self becomes the origin of society and knowledge thereof (cf. Henriques et al., 1984). It is assumed that the self is a unitary and bounded entity, a source and explanation of that which interests psychologists. The structuralist self comes to define that which should be investigated by psychologists. Potter and Wetherell (1987) refer to all "traditional" views of self as "self-as-entity" (p.102). This is the same sense which I wish to capture when I make use of the term "the structuralist self". The "discursive turn" which heralded the "arrival" of post-structuralism strongly challenges the idea of self as "within", "intimate" and "bounded". Self-as-entity can be thought of as a line in the sand - a line that has been drawn as the result of historically and culturally contingent discursive practices, a line that has nothing more to do with who we "really" are than the terrain upon which it is drawn. The line in the sand has blown away, and suddenly we are "multiple", "social", and "distributed". We are selves-as-social constructs.

This is not a model of self, in the same way that I discussed the "relational self" or the "Cartesian self". Selves-as-social constructs implies being positioned in relation to particular discursive formations. Harre states that "to be a self is not to be a certain kind of being, but to be in possession of a certain kind of theory" (1985, cited in Potter and Wetherell, 1987: 102). Similarly Potter and Wetherell (1987) claim that it is "meaningless to ask which of these models is the correct description of the self, they are simply equivalent ways of making sense of the self with their own context, relative advantages and disadvantages" (p.102).

Selves-as-social constructs presents a new approach to the self. The major strands of this new approach are summarised below:

* Psychological models of the self are constructed socially, and are historically and culturally contingent.

* New prescriptions for future study are to be set up - there is a shift in emphasis from studying "behaviour" and "attitudes" to examining the language practices and discourses prevalent in different contexts (from Potter and Wetherell, 1987: 103-104).

Selves-as-social constructs: Implications for psychology

* Social psychologists abandon attempts to describe once and for all human acts and their meanings, and instead attempt to describe the linguistic practices which produce an act as a certain type of event (Potter and Wetherell, 1987: 104)

* The way you understand selves-as-social constructs, and attendant linguistic practices "have vital consequences for the positioning of people in society; they are not neutral or without impact, they produce senses of the self which may be negative, destructive, oppressive, as well as senses which might change or liberate" (Potter and Wetherell, 1987: 104).

* Your own subjectivity and how you construct yourself and others is not inconsequential in determining the outcome of psychotherapy and psychological research.

* The acknowledgement of self as multiple and distributed is not necessarily negative (cf Barglow, 1994; Turkle, 1995).

* The implication of the taking seriously of selves-as-social constructs allows us to start thinking about viable alternatives to both "personality" and the problems associated with it under the guise of "personality disorders" and other forms of "psychopathology" (cf. Hacking, 1995; Parker et al., 1995).

* The therapeutic implications of viewing selves-as-social constructs can be seen in the work of White and Epston (1990) & Epston et al. (1992)


The aim of this paper has been to call into question the often uncontroversial idea that the self is something that is private, intimate, asocial and unitary - the essence of who we are. This view of self-as-entity has been successfully challenged, on the grounds that it incorrectly portrays persons as that which they only sometimes are. The alternative to this structuralist view of self-as-entity was presented as the post-structuralist view of selves-as-social constructs. The question that now arises is if the line in the sand has indeed blown away, and the structuralist self is untenable, then how can it be that it is still the prevailing and dominant discourse of the self. The answer to this could be that the line is blown away in much the same way as it is drawn: slowly. The line in the sand is being blown away... The ubiquity of the idea of self-as-entity also speaks of the myriad political, economic, discursive and ideological practices which cluster around the self, and the management thereof.


Allport, G. (1937) Personality: A psychological interpretation. London: Constable.

Bakhurst, D. & Sypnowich, C. (1995) Introduction: Problems of the self. In D. Bakhurst and C. Sypnowich (eds), The social self. London: Sage.

Barglow, R. (1994) The crisis of self in the age of information: Computers, dolphins and dreams. London: Routledge

Breakwell, G.M. (1986) Coping with threatened identities. London: Methuen.

Burkitt, I. (1991) Social selves: Theories of the social formation of personality. London: Sage.

Burr, V. (1995) An introduction to social constructionism. London: Methuen.

Cattel, R.B. (1966) The scientifc analysis of personality. Chicago: Aldine.

Danziger, K. (1990) Constructing the subject: Historical origins of psychological research. Cambridge: Cambridge University Press.

Epston, D., White, M. and Murray, K. (1992) A proposal for re-authoring therapy: Rose's revisioning of her life and a commentary. In K.J. Gergen & S. McNamee (eds) Therapy as social construction. London: Sage.

Eysenck, H.J. (1953) The structure of human personality. New York: Wiley.

Gergen, K.J. (1982) From self to science: What is there to know? In J. Suls (ed) Psychological perspectives on the self, volume one. Hillsdale, New Jersey: Lawrence Erlbaum and associates.

Hacking, I. (1995) Why multiple personality tells us nothing about the self/mind/person/subject/soul/consciousness. In D. Bakhurst and C. Sypnowich (eds), The social self. London: Sage.

Hoffman, L. (1992) A reflexive stance for family therapy. In K.J. Gergen & S. McNamee (eds) Therapy as social construction. London: Sage.

Henriques, J., Hollway, W., Urwin, C., Venn, C. & Walkerdine, V. (1984) Changing the subject. London: Methuen.

Hollway, W (1989) Subjectivity and method in psychology: Gender meaning and science. London: Sage

Kitzinger, C. (1992) The individuated self-concept: a critical analysis of social constructionist writing on individualism. In G.M. Breakwell (ed), Social psychology of identity and self concept. London: Academic press.

Leonard, P (1984) Personality and ideology: Towards a materialist understanding of the individual. London: Macmillan.

Maslow, A.H. (1968) Toward a psychology of being (second edition). Princeton: Van Nostrand.

Marx, K. (1845) Theses on Feuerbach. In K. Marx and F. Engels Selected works volume one. London: Lawrence and Wishart, 1968.

Mischel, W. (1968) Personality and assessment. New York: Wiley.

Parker, I., Georgaca, E., Harper, D., McLaughlin, T. and Stowell-Smith, M. (1995) Deconstructing psychopathology. London: Sage.

Potter, J. & Wetherell. M. (1987) Discourse and social psychology: Beyond attitudes and behaviour. London: Sage.

Rogers, C. (1961) On becoming a person. Boston: Houghton Mifflin.

Rose, N. (1989) Governing the soul: The shaping of the private self. London: Routledge.

Rose, N. (1996) Inventing our selves: Psychology, power and personhood. Cambridge: Cambridge University Press.

Seve, L. (1975) Marxism and the theory of human personality. London: Lawrence and Wishart.

Turkle, S. (1997) Life on the screen: Identity in the age of the internet. London: Phoenix.

White, M. & Epston, D. (1990) Narrative means to therapeutic ends. New York: Norton.

Paper presented at the 3rd Annual Qualitative Methods Conference: "Touch me I'm sick"
8 & 9 September 1997, University of South Africa Regional Office, Durban
critical methods society - -