Brief history of the concepts of ‘personality types’ and ‘personality disorders’

This that holds your eyes is nothing save
The image of yourself reflected back to you.
It comes and waits with you; it has no life;
It will depart if you will only go.
– Ovid, Metamorphoses ‘Narcissus and Echo’

The idea that there are distinct personality or character ‘types’ goes back at least to ancient Greece. The Greek philosopher Theophrastus described 29 character types that he saw as deviations from the norm, while the Greek physician and philosopher Galen developed a theory of personality types based on the concept of the ‘four humours’ proposed by Hippocrates, promoting a theory and typology of human temperaments that was to have an enduring legacy.

Such views lasted into the eighteenth century, when interest shifted from the supposed biologically-based ‘humours’ and ‘temperaments’, to more distinctly psychological concepts of character and ‘self’. In the nineteenth century, the idea of ‘personality’ referred to a person’s conscious awareness of their behaviour, and there was an increasing clinical focus on issues of delusions, hallucinations, and insanity (in a cultural context of Enlightenment notions of rationality, standardisation, and classifcation; see Foucault, Madness and Civilization ; Bentall, Madness Explained), and attempts at clinically delineating and treating these states, as in the pioneering work of Pinel, Prichard, Krafft-Ebing, and Koch.

In the early 20th century, Kraepelin proposed six types of personality disorder (excitable, unstable, eccentric, liar, swindler, and quarrelsome), and three paranoid (or delusional) disorders, resembling later concepts of schizophrenia, delusional disorder, and paranoid personality disorder. But perhaps the most signifcant and enduring contribution to our understanding and treatment of these distinct personality types came from Freud’s psychoanalytic development of the concept of character disorders, which he viewed as enduring psychical problems linked to internal conficts and/or disruptions of normal childhood development, and his theories on neurosis and psychosis. The term ‘borderline’, for example, originally stems from a belief that some individuals were functioning on the edge or ‘border’ of those two categories.

William James also talked in terms of temperaments, and Jung (1921) specifically tried to get away from the notion of personality disorder by outlining the different personality types – thinking, feeling, intuition, and sensation – related to either an introverted or extroverted attitude.

In the 1970s, Otto Kernberg and Heinz Kohut further developed our theoretical understanding of borderline and narcissistic personality types (see Key Figures), and these ideas contributed to the decision by the DSM to include these as ‘personality disorders’ in the DSM-III (1980).

Personality disorders are generally understood today as being characterised by enduring maladaptive patterns of behaviour, cognition, and inner experience (self), exhibited across many contexts and deviating from those accepted by the individual’s culture. This cultural aspect is important in pointing both to the relational and the contextual aspect of these personality traits: the diagnostic term ‘disorder’ denotes ‘differing from social norms’, as well as signifying noticeable individual and/ or relational distress. The DSM characterises the essential features of a personality disorder as impairments in personality (self and interpersonal), functioning, and the presence of pathological personality traits. These clinical personality patterns develop early, are usually enduring and infexible, and are typically associated with signifcant distress, disruption, or disability.

Bollas (2000) talks in terms of character states and character disorder, arguing that we all move between different character states – narcissistic, schizoid, borderline and hysteric – but that when one comes to rigidly predominate he calls it a character disorder.

Prevalence of narcissistic and borderline personality traits

There are currently ten personality disorders listed in the DSM (DSM-V), clustered into three distinct groups (Clusters A, B, and C). Both narcissistic personality disorder (NPD) and borderline personality disorder (BPD) are grouped together in “Cluster B”, together with antisocial and histrionic personality disorders, and are seen as sharing common characteristics and structural similarities. Cluster B personality disorders are characterized by dramatic, overly emotional or unpredictable thinking or behaviour, and manipulative, exploitative interactions with others (see Symptoms).

Whilst dividing personalities into such discrete subdivisions is diagnostically useful, it should be noted that in actual life many of the boundaries tend to blur into one another (indeed, the diagnostic division into ‘clusters’ of shared traits is intended in part to refect this tendency). For example, NPD frequently co-occurs with other personality disorders, as well as with other forms of mental distress such as depression. There is also an ongoing debate about whether NPD should be classifed as a distinct ‘disorder’ at all: the latest edition of the ICD does not specifcally distinguish or define the characteristics of NPD, but rather categories it under “Other Specifc Personality Disorders” (see Controversies).

It is considered that 40-60% of all psychiatric patients have some form of personality disorder (i.e., pervasive and infexible behaviours), making them among the most frequent of psychiatric diagnoses. Up to 10% of the general population may have some form of personality disorder.

With regard to narcissistic personality disorders, it is estimated that between 0.5 and 1% of the population may have NPD at some point in their life. It is more frequently found in males; it affects young people more than older people; and it usually develops in adolescence or early adulthood (Klonsky et al.).

The prevalence of BPD is believed to be 1-2% of the general population. Where narcissistic personality traits are generally more commonly found in males, borderline characteristics have traditionally been thought to occur more often in females, although there have only been a few large-scale, population-based epidemiological studies, and more research in this area is needed (see Oldham for a recent overview). One difficulty in ascertaining accurate statistics concerning prevalence and possible gender bias is the extensive comorbidity of BPD with other disorders; alcohol and substance abuse and eating disorders are also commonly co-diagnosed.

The concepts of ‘borderline’ and ‘narcissism’

The clinical term ‘borderline‘ is in some ways a rather problematic one, refecting the early historical confusion concerning this complex personality type. The term was first brought into psychiatric terminology in 1938 by American psychoanalyst Adolph Stern, who used it to describe patients who were, as he put it, “on the border” of psychosis and neurosis – for example, who displayed particular symptoms under stress but then soon became relatively functional again (‘Psychoanalytic investigation of and Therapy in the Border Line Group of Neuroses’, in The Psychoanalytic Quarterly 7.4).

Because of the difficulties with the term, there is currently a lively debate about re-naming the disorder, both because of its historical clinical confusion and because it may inadvertently stigmatise those with it, suggesting that they perhaps ‘hover’ between sanity and insanity, or exist in some sort of liminal state or hinterland (see Controversies section). In psychiatric circles it has now been renamed as emotionally unstable personality disorder (EUPD).

Despite its comparatively recent clinical definition (BPD was only listed as a distinct personality disorder in the 1980 edition of DSM), some of the personality traits associated with this disorder seem to have relatively ancient origins, as in the vivid descriptions of individuals with apparently ‘manic-depressive’ forms of personality (the co-existence of intense, divergent moods within an individual) such as is found in the literature of Homer, Hippocrates, and Aretaeus.

Interestingly, ‘narcissism‘ also has its etymological roots in Greek culture, the word deriving from the celebrated myth of Narcissus, a beautiful male youth who fell in love with his own image. As Jeremy Holmes notes in his talk, the original story of Narcissus strikingly captures and embodies many of the key diagnostic features of what today is classified as NPD, in its compelling presentation of self-love, the crucial role that visual image, refection, and gaze plays in this; the themes of gender, self-knowing, and suicidal behaviour. Even our modern understanding of the relational dynamic between ‘thick-skinned’ and ‘thin-skinned” aspects of narcissism is poignantly present in the doomed relationship between the self-absorbed, grandiose Narcissus and the vulnerable, ‘thin-skinned’ Echo, who is fatally attached to him and eventually disappears into just an echo.

The concept of excessive self-admiration has frequently been culturally linked in history and literature with the idea of hubris (a state of arrogance and aloofness often portrayed as being out of touch with reality), and as such has been extensively discussed by various Western philosophers and writers throughout history. However, it was only relatively recently that the notion of narcissism as a clinical disorder became a subject of scientifc interest in the feld of psychology. In 1898 Havelock Ellis, an English physician and sexologist, used the term “narcissus-like” in reference to excessive masturbation, in which the person becomes his or her own sex object. A year later, Paul Näche frst used the term “narcissism” in his study of sexual perversions, and in 1911 Austrian psychoanalyst Otto Rank published one of the earliest psychoanalytic descriptions of narcissism, in which he connected it to self-admiration and vanity. But it was Sigmund Freud who investigated the deeper psychoanalytic processes behind and within ‘narcissism’ in his landmark 1914 paper ‘On Narcissism: An Introduction’, suggesting that narcissism was developmentally connected to whether one’s libido is directed inward toward one’s self, or outward toward others. Importantly, Freud distinguished between Primary and Secondary narcissism: the former category has since been challenged by some infant studies (see Stern, 1985), whilst the latter is more generally accepted.

As Phil Mollon has noted, “narcissism has been a deep theme in psychoanalysis for a long time”, and in the 1960s and 70s, psychoanalysts Otto Kernberg and Heinz Kohut helped spark further interest through their pioneering psychoanalytic research into this distinctive personality type. The term “narcissistic personality structure” was introduced by Kernberg in his 1975 work ‘Borderline Conditions and Pathological Narcissism’, in which he developed a theory of narcissism that suggested three major forms: normal adult narcissism, normal infantile narcissism, and pathological narcissism.

A year later, Heinz Kohut put forward the concept of “narcissistic personality disorder” in ‘The Psychoanalytic Treatment of Narcissistic Personality Disorders’. Kohut developed a different theoretical understanding of narcissism from Kernberg, and challenged a number of Freud’s earlier observations about narcissism, for instance in postulating separate lines of development for narcissism (primitive narcissism, mature narcissism). The concept was to play a key role in Kohut’s theory of ‘self-psychology’, which posited the notion of ‘healthy’ narcissism, suggesting that narcissism might be an essential aspect of normal development, and that it was difficulties with early “self-object” relationships that lead to challenges in maintaining an adequate sense of self-esteem later in life, thus contributing to narcissistic disorders. This discussion between so-called ‘healthy’ narcissism and its more malign or ‘pathological’ variant, is one of the most discussed and elaborated elements of subsequent theoretical formulations (see Controversies).

In 1980, narcissistic personality disorder was officially recognized in the third edition of the DSM and criteria were established for its diagnosis. There has been extensive debate about how to classify personality disorders in the more recent edition of the DSM (DSM-5), but in the end narcissistic and other personality disorders have remained relatively unchanged in their diagnostic criteria from the previous edition.

Herbert Rosenfeld also contributed significantly to this area in his classic book Impasse and Interpretation(1987) with his work on narcissism, describing narcissistic omnipotent object relations and how these can manifest in analytic work. Rosenfeld also introduced the concepts of thick- and thin-skinned narcissism, conceptualisations that have been taken up and developed further by Ron Britton – see Sex, Death, and the Superego (which includes two chapters on narcissism).

From 1989 borderline personality disorder has been increasingly linked with trauma in a detailed way (viz. Herman, Perry and van der Kolk 1989) and since that time there has been an increasing recognition of the role of trauma in psychiatric diagnoses as reflected in each subsequent edition of the DSM.

Peter Fonagy’s groundbreaking 1991 paper on work with borderline patients (‘Thinking about thinking‘, Int. J. Psycho-Anal, 72) began to introduce his integration of attachment perspectives into psychoanalytic thinking, as well as his conceptualisation of mentalisation. In this paper he describes how the borderline patient cannot bear to think about the hatred or murderousness that is in the caregiver’s mind.

This recognition of the detailed effects of early relational trauma and real world experience has been increasingly integrated into an analytic understanding of work with personality disorders and has been particularly taken up the school of relational psychoanalysis. Alongside this has been the integration (and rehabilitation) of Bowlby’s work on attachment. See in particular Fonagy (see References section), Bromberg (2011), Davies and Frawley (1992), Benjamin, Donnel Stern, the Boston Change Process Study Group (2011), and West (2016).