Paper: The character–personality distinction: An historical, conceptual, and functional investigation
Between character and personality: Or some hidden difficulties of interdisciplinary study
Interdisciplinary research, as valuable as it certainly is, seems to be a fundamentally ambiguous endeavor. Such ambiguity, importantly, does not consist in the very idea of building bridges between disciplines. Rather, it refers to metatheoretical and methodological issues inevitably involved in connecting two or more fields. The latter are often substantially different and may involve conceptual and theoretical apparatuses which, if not altogether incommensurable, are hard to translate into each other. The fact that fundamental problems of interdisciplinary studies are rarely explicitly explored makes this issue even more troublesome.
Without any attempt at addressing this enormously complex problem as a whole, it may be worthwhile to recall one recent example of interdisciplinary discussion occurring at the intersection of empirical psychology and normative ethics: a discussion of virtuous character. The latter being a paradigmatic subject matter of virtue ethics, at least from Socrates, has recently been reconsidered in the light of experimental results obtained by academic psychology. More specifically, it was related to the criticism of the concept of personality voiced mostly by social psychologists.
The conceptual and theoretical core of personality psychology, both in its scientific and folk versions (Gilbert & Malone, 1995; Ross, 1977), has usually been constructed around the notion of temporally stable and cross-situationally consistent features: so-called global or robust traits. A recent empirical tradition of situationism, however, seems to provide ample evidence not only for the fact that we are all indeed “prone to dispositionism” of this kind, but also that such “dispositionism is false” (Goldie, 2004, p. 63). The researchers from this tradition deny that there are stable and consistent traits or, alternatively, insist that most actual people don’t exhibit traits of this kind. Rather, the large body of empirical evidence (among the research most commonly discussed by virtue ethicists is that by Darley & Batson, 1973; Isen & Levin, 1972; Milgram, 1963; for a more complete review see Doris, 2002) provided shows that it is a situation in which an agent finds him/herself acting, rather than an allegedly context-independent and stable personality, that accounts for the large amount of human behavior.
The experiments conducted by social psychologists were soon generalized into doubts concerning the usefulness of trait concepts for the purposes of scientific explanation and prediction. Understood in such a context, in turn, they attracted the attention of many philosophers. The empirical results mentioned above could, indeed, have been disquieting, especially if one realized that the very center of traditional philosophical moral psychology, especially within so-called virtue ethics, had been founded on the notion of moral character with virtues and vices aspiring to exactly the same stability and cross-situational consistency that was undermined in the case of personality. Among the philosophers it was especially Gilbert Harman (1999, 2000) and John Doris (1998, 2002) who stimulated a fierce debate by stating that situationist literature posed a grave threat against “globalist moral psychologies” (Doris & Stich, 2014), and as undermining the very basis of both ancient and contemporary virtue ethics.
Such a far-reaching claim, obviously, provoked a strong response (for useful reviews see Alfano, 2013; Appiah, 2008; Goldie, 2004; Miller, 2013a). What seems to have been assumed by at least many disputants from both sides of the controversy, however, was a relatively direct applicability of psychological theses concerning personality to philosophical issues referring to character. In brief, it was the interchangeability of the notions of personality and character that had been presumed. Despite the fact that such an implicit assumption has been often made, these two notions are not identical. True, they are often used interchangeably and the difference is vague, if not obscure. Still, however, the notions in question can be distinguished from each other and the effort to draw the distinction is arguably worthwhile because of the latter’s bearing on many particular issues, including the above discussion of situationism.
One possible way of exploring the difference between these two concepts is to compare the typical, or paradigmatic, ways of their application as revealed in their respective original domains. Common language is obviously not very helpful here, as it exhibits the very same confusion that is intended to be clarified. Rather, the context of classical virtue ethics (for character) as well as that of academic personality psychology (for personality), are promising. Such a general clue will be used in the following sections. At first, the concepts of character and personality will be investigated both historically and systematically. Then, in turn, a parallel will be drawn between the pair in question and so-called fact–value distinction and an analysis of the functions played by both concepts conducted. Finally, the outcomes achieved will be placed in the context of some differences between the fact–value distinction and the Humean is–ought dichotomy.
Historical vicissitudes of the notions
As far as the concepts of character
and personality are concerned it is certainly the former that has a longer history. In fact, it was a traditional medium of addressing individual differences at least from the times of the ancient Greeks and Romans until, more or less, the beginning of the 20th century.
The etymology of the contemporary term “character” in many languages, including both English and German, leads to the Greek word charaktêr
, which itself had once undergone a considerable evolution of meaning (see Homiak, 2015
). Initially, it used to refer to a mark impressed upon a coin. Subsequently, it was generalized to mean a distinctive feature of a thing and, finally, took on a recognizable meaning of a set of qualities typical to an individual.
In antiquity the notion of character was inseparably connected with the normative aspect of human conduct and in most contexts amounted to moral qualities of a respective person: to virtues and vices. Such a connection was emphasized in a term alternative to “character”: the Greek word “êthos” (cf. Gill, 1983, p. 472). An evaluative discourse of character can be found in common language and folk psychology (cf. Irwin, 1996), but it is its professional version proper to virtue ethics that is crucial in the present context. The latter philosophical tradition took on its classical form in Socrates and culminated with Aristotle’s (trans. 2000) Nicomachean Ethics, which to this day is a paradigmatic example of a virtue ethical account.
Ancient conceptions of character were descriptive and normative with both these features closely intertwined. They involved theories of moral and developmental psychology and, at the same time, a prescription and a detailed instruction of character education and character self-cultivation. And it was, importantly, a ‘life-long learning’ account that was provided: it was a rational adult, rather than a child deprived of genuine rationality, who was regarded by Cicero, Seneca, or Plutarch as able to accomplish “character formation through reasoned reflection and decision” (Gill, 1983, p. 470). The standards for the success of such a process were usually considered objective. In the Aristotelian context, for instance, it was the ability to properly perform human natural functions that provided the ultimate criterion.
The ancient Greek and Roman concept of character turned out to be profoundly influential in the following ages at least, as has been mentioned, until the beginnings of the previous century (for part of the story, see MacIntyre, 2013). Some of the variations on this ancient notion can be found in the Kantian ideal of the ethical personality, the German tradition of Bildung, the 19th-century American model of the balanced character and, last but not least, the Victorian vision of the virtuous character very vivid in the novels from this cultural milieu (Woolfolk, 2002). What is remarkable is that the notion of character, as influential as it used to be, is considerably much less important today. Nowadays, in fact, it seems to be mostly substituted by the concept of personality. And it is the history of the process that led to this state of affairs, of the shift “from a language of ‘character’ to a language of ‘personality’” (Nicholson, 1998, p. 52) that can be very revealing in the present context. Two particularly helpful accounts have been provided by Danziger (1990, 1997) and Brinkmann (2010).3
Danziger begins his account with an important remark that initially the notion of personality carried the meanings which were not psychological, but were theological, legal, or ethical ones. It was only as a result of considerable evolution that it “ended up as a psychological category.” The first important dimension of the process of its coming “down to earth” (1997, p. 124) was the medicalization. Danziger places the latter in 19th-century France, where medical professionals were as skeptical about the earlier theologically or philosophically laden versions of the notion as they were enthusiastic about the promises of its naturalization. It was as a result of their reconceptualization that “personality” began to refer to “a quasi-medical entity subject to disease, disorder and symptomatology” (1997, p. 131). The term understood as such won its place within medical discourse and soon, in 1885, it became possible for Théodule Ribot to publish The Diseases of the Personality without a risk of conceptual confusion. An evolution began which would later lead to the inclusion of the personality disorders into the DSM (cf. Brinkmann, 2010, p. 73).
Among the descendants of the medicalization it is arguably the mental hygiene movement, “an ideological component” (Danziger, 1990, p. 163) of the rise of contemporary research on personality, that was most important at that time. On the basis of the belief that it is an individual maladjustment rooted in early development that is responsible for all kinds of social and interpersonal problems, “a powerful and well-founded social movement” (p. 164) directed at the therapy of the maladjusted as well as at the preventive efforts addressed to the potentially maladjusted (which could include everybody), was initiated. The notion of personality, as noted by Danziger, “played a central role in the ideology” (p. 164) of this movement. More particularly, it was the “personality” of individuals addressed by the latter which was recognized as “the site where the seeds of future individual and social problems were sown and germinated” (Danziger, 1997, p. 127) and, accordingly, established as an object of intervention.
Personality understood as such needed to be scientifically measured on the dimension of its adaptation/maladaptation and it was at this place that the psychologists from the Galtonian tradition of individual differences and mental testing arrived on the scene. In fact, it could easily seem that no one was better equipped than those researchers to perform the task set by the mental hygiene movement and to provide the latter’s ideology with a technical background. Mental testing confined to cognitive abilities or intelligence at roughly the same time, i.e., after World War I, turned out to be insufficient not only as a means of occupational selection but also for its originally intended application, i.e. as a predictor of school success. In effect, there was an increasing recognition of the need for measurement techniques for non-intellectual mental qualities.
And such techniques were indeed soon developed using the very same methodological assumptions that had been previously applied to cognitive abilities. Paper-and-pencil questionnaires measuring non-cognitive individual differences “began to proliferate” (Danziger, 1990, p. 159). Simultaneously, a new field of psychological investigation, “something that could sail under the flag of science” (p. 163), began to emerge. Only one more thing was lacking and it was a label, a name for the new sub-discipline and its subject matter.
The “shortlisted” candidates included the notions of temperament, character, and personality. The first one was rejected due to its then associations with physiological reductionism. Why not “character,” then? Well, that notion in turn was considered inappropriate due to its association with the concept of will being an “anathema to scientifically minded American psychologists” (Danziger, 1997, p. 126) and generally normative connotations. The third candidate, “personality,” as a result, came to the fore.
Not only was it devoid of an unwelcome moralistic background and already popularized by the mental hygiene movement, it also offered a realistic prospect of quantitative empirical research. Already adopted by scientific medicine and understood along the lines of Ribot as an “associated whole” (un tout de coalition) of the variety of forces, personality, rather than holistic character, was a much more promising object for the post-Galtonian methodology (Danziger, 1997, p. 127; cf. Brinkmann, 2010, p. 74). Soon, the newly emerging field “developed loftier ambitions” (Danziger, 1997, p. 128) and became a well-established part of academic psychology4 with its flagship project of discovering basic, independent, and universal personality-related qualities: the traits. And it is actually this tradition that is more or less continued today, with the Big Five model being a default perspective.
Note: I would add, that moralistic social “tradition” did not disappear from “personality theory” – psychology remains a socio-religious “prescriptive and rigid” conception of human behavior, despite the effort to construct “something that could sail under the flag of science”
For the establishment of personality rather than character as a subject matter of the new psychological science, Gordon W. Allport’s importance can hardly be overestimated (Allport, 1921, 1927; cf. Nicholson, 1998). Following an earlier proposal by John B. Watson Allport drew an explicit distinction between normatively neutral personality, “personality devaluated,” and character as “personality evaluated” (Allport, 1927, p. 285). Personality and character, crucially, were regarded by him as conceptually independent. The former, in particular, could be intelligibly grasped without the reference to the latter: “There are no ‘moral traits’ until trends in personality are evaluated” (p. 285). Accordingly, an evaluation was considered as additional and only accidental. As such it was regarded as both relative and connected with inevitable uncertainty (for the cultural background and metaethical position of emotivism lying behind such an approach see MacIntyre, 2013).5
The point which is crucial here is that the recognition of the normative element of the character concept led to its virtual banishment. While listing “basic requirements in procedures for investigating personality,” Allport (1927, p. 292) was quite explicit to enumerate “the exclusion of objective evaluation (character judgments) from purely psychological method” (p. 292). Those psychologists who accept his perspective “have no right, strictly speaking, to include character study in the province of psychology” (Allport, 1921, p. 443).6
The transition from the notion of character to that of personality was a very complex process which reflected some substantial changes in cultural and social milieu. Some insightful remarks about the nature of the latter have been provided by Brinkmann’s (2010) account of the shift between the premodern “culture of character” and essentially modern “culture of personality.”This shift, importantly, was not only a “linguistic trifle.” Rather, it was strictly associated with “the development of a new kind of American self” (Nicholson, 1998, p. 52).
A culture of character, to begin with, was essentially connected with moral and religious perspectives, which provided the specification of human télos. And it was in relation to the latter that the pursuit of moral character was placed. In the paradigmatic Aristotelian account, for instance, the notion of the virtuous character was essentially functional in the same way in which the concept of a good watch is (MacIntyre, 2013). The criteria of success and failure, accordingly, were defined in terms of one’s ability to perform the natural functions of the human creature. And the latter were not “something for individuals to subjectively decide” (Brinkmann, 2010, p. 70). Rather, they were predetermined by a broader cosmic order of naturalistic or theological bent.
The goal of adjusting one’s character to suit the requirements of human nature was institutionalized in social practices of moral education and character formation. According to Brinkmann, it was especially moral treatment or moral therapy that embodied the default approach “to the formation and correction of human subjects” (2010, p. 71). This endeavor was subsequently carried on in the very same spirit, though in an essentially different cultural milieu, by William Tuke and Phillipe Pinel and it was no earlier than with Sigmund Freud that a new form of therapy, properly modern and deprived of an explicit normative background, emerged.
Note: And yet, in American psychology, it is precisely this “imaginary normal” that continues to be the default assumption against which pathology and defect are assigned.
The ancient virtue ethical approach embodied in a culture of character was taken over by the Middle Ages, with an emphasis shifted considerably towards theological accounts of human goals. A thoroughly new perspective proper to a culture of personality appeared much later with the emergence of the scientific revolution, which seriously undermined the belief in objective normative order. The earlier cosmic frameworks started to be supplanted by psychological perspectives with romanticism and modernism being, according to Brinkmann (2010, p. 72), two forces behind them.
One of the main running threads of romanticism is the idea that “each human being has a unique personality that must be expressed as fully as possible” (Brinkmann, 2010, p. 73). Before romanticism, the final purpose had been specified in terms external to a particular individual. It was related to generic norms of humanity as such or to those determined by God. (Today, “generic norms” are determined by a “new” God: the psych industry) Now the goal to be pursued started to be understood as properly individual and unique
Note: I don’t think that Americans understand how pervasive “the shift” away from the individual being “a unique personality that must be expressed as fully as possible” to a totalitarian demand for conformity as dictated by a “new religious” tide of psycho-social tyranny was accomplished in a few decades. It is not surprising that Liberalism is every bit as religious as the Christian Right in its goal to “restore” the extreme religious aims (and hatred of humanity) of Colonial America; a continuation of the religious wars that raged in Europe for centuries.
This difference is evident when one compares Augustine’s and Rousseau’s confessional writings. The former “tells the story of a man’s journey towards God,” whereas the latter “is about a man’s journey towards himself, towards an expression of his own personality” (Brinkmann, 2010, p. 73). (Not allowed anymore!)
The demand for the “journey towards himself” can be connected with a disenchantment of the world, which had left an individual in a universe devoid of meaning and value. If not discovered in the world, the latter needed to be invented by humans. One had to “turn inwards” in order to find the purpose of life and this entailed, significantly, the rejection of external and social forces as potentially corrupting the genuine inborn self. The idea of “an individual in relative isolation from larger social and cosmological contexts” began to prosper and it “paved the way for the modern preoccupation with personality” (Brinkmann, 2010, pp. 67, 73) defined in fully atomistic or non-relational terms.
The second major force behind a culture of personality was modernism, which, in alliance with the modern idea of science, entailed an “ambition of knowing, measuring [emphasis added], and possibly improving [emphasis added] the properties of individuals” (Brinkmann, 2010, p. 73), which proved to have a considerable bearing on the newly emerging notion of personality. The latter concept had been deeply influenced by the logic of standardization and quantification characteristic of the whole of modernity; not only of its industry, but also of education, bureaucracy, and the prevailing ways of thinking. This logic found its direct counterpart in trait-based thinking about personality with the idea that the latter can “be measured with reference to fixed parameters” and that individuals “vary within the parameters, but the units of measurement are universal” (Brinkmann, 2010, p. 75). (This assumption that “opinions that arise from a social agenda” can be quantified is disastrous.)
The romantic and modernist branches of a culture of personality, with all the differences between what they laid emphasis on, were connected by the common atomistic account of the self and a plea for the development of unique qualities of the individual. And it is this “core element” of their influence which is still in place today,8 even though some authors, Brinkmann included, have announced the appearance of a new cultural formation, a culture of identity.
The character–personality distinction
The complexity of the processes which helped the contemporary notion of personality to develop from its ancient predecessor can explain both semantic similarities and the subtlety, if not obscurity, of the differences existing between them. In order to clarify this vague conceptual area an attempt will be made to distinguish between the concept of character as paradigmatically present in virtue ethical discourse and the concept of personality as represented, in an arguably clearest form, by academic personality psychology.
Some valuable efforts to draw a distinction between these two notions, importantly, have already been made in a variety of contexts (Gill, 1983
; Goldie, 2004
; Kristjansson, 2010
; Kupperman, 1991
; Miller, 2013a
; cf. Kristjansson, 2012
; McKinnon, 1999
). The ensuing exploration is intended as a systematization and synthesis of these proposals.
The relationship between two notions in question can be elucidated by, first, indicating their common features (genus proximum) and, then, by specifying the ways in which they differ from each other (differentia specifica). As far as the former is concerned, both “character” and “personality” can be regarded as constructs belonging to the discourse of individual differences.9 Both notions are analyzable, even if not reductively analyzable, in terms of some lower-level terms such as virtues and vices or, respectively, traits. These lower-lever concepts are usually understood as dispositional. A personality trait, for instance, can be defined as a “disposition to form beliefs and/or desires of a certain sort and (in many cases) to act in a certain way, when in conditions relevant to that disposition” (Miller, 2013a, p. 6). The higher level notions of character and personality, accordingly, are also dispositional.
The formal features indicated above are common to the notions of character and personality.10 And it is on the basis of this “common denominator” that one can attempt to clarify the difference between them. A good place to begin with is a brief remark made by Goldie (2004) who claimed that “character traits are, in some sense, deeper than personality traits, and … are concerned with a person’s moral worth” (p. 27). It is a dimension of depth and morality, then, which can provide one with a useful clue. (Note that both “traits” and moral rules are subjective, culturally defined and NOT quantifiable objects: that is, this remains a religious discussion.)
As far as the depth of the notion of character is concerned, the concept of personality is often associated with a considerable superficiality and the shallowness of mere appearances (Goldie, 2004, pp. 4–5; Kristjansson, 2010, p. 27). The fact that people care about character, accordingly, is often connected with their attempt to go beyond the “surface,” beyond “the mask or veneer of mere personality” (Goldie, 2004, p. 50; cf. Gaita, 1998, pp. 101–102).11 Even the very etymology of the term “personality” suggests superficiality by its relation to the Latin concept of persona: “a mask of the kind that used to be worn by actors.” Character as deeper “emerges when the mask is removed” (Goldie, 2004, p. 13; cf. the Jungian meaning of persona).
The reference to the depth of character, as helpful as it may be, is certainly insufficient due to its purely formal nature. What still remains to be determined, is a substantive issue of the dimension on which character is deeper than personality. As far as Goldie’s distinction is concerned such a specification is provided in what follows: “someone’s personality traits are only good [emphasis added] conditionally upon that person also having good character traits … On the other hand, the converse isn’t true: the goodness [emphasis added] of someone’s character trait is not good [emphasis added] conditionally on his having good personality traits” (2004, p. 32). It is depth referring to ethical dimension, then, which distinguishes between character and personality.12 One’s virtue of honesty, for instance, can still be valued even if the person in question is extremely shy (introvert, as the psychologist would say). (Both introversion and “honesty” are labeled symptoms of “developmental disorder” in the ASD / Asperger diagnosis)
It does not work the other way around, though. An outgoing and charming personality, when connected with considerably bad character, is in a sense polluted. A criminal who is charming can be even more dangerous, because he/she can use the charm for wicked purposes.13 Such a difference, importantly, should not be taken as implying that personality cannot be evaluated at all. It can, with a reservation that such an evaluation will be made in terms of non-moral criteria or preferences. An extraverted person, for instance, can still be considered as a “better” or more preferable candidate for the position of talk show host (cf. Goldie, 2004, p. 47; McKinnon, 1999, pp. 61–62).
The above-given specification of the distinction can be enriched by some remarks by Gill (1983, p. 470), who notices that “character” and “personality” are not only distinguishable as two concepts but also as “two perspectives on human psychology” for which they are, respectively, central. The character-viewpoint, to begin with, “presents the world as one of … performers of deliberate actions” (Gill, 1986, p. 271). Human individuals, in particular, are considered as more or less rational and internally consistent moral agents possessing stable dispositions (virtues and vices) and performing actions which are susceptible to moral evaluation and responsibility ascription. The evaluation of their acts, importantly, is believed to be objective: to be made along the lines of some definite “human or divine standards” (p. 271). No “special account,” accordingly, is taken “of the particular point of view or perspective of the individuals concerned” (Gill, 1990, p. 4).
The personality-viewpoint, on the other hand, is not associated with any explicitly normative framework. Rather, it is colored by “the sense that we see things ‘as they really are’ … and people, as they really are” (Gill, 1986, p. 271). The purposes are psychological, rather than evaluative: to understand, empathize with, or to explain. Also the default view of the individuals in question is considerably shifted. Their personality is recognized as being “of interest in its own right” (Gill, 1983, p. 472) and their agency as considerably weakened: “The person is not typically regarded as a self-determining agent,” but rather as a “relatively passive” (p. 471) individual often at the mercy of the forces acting beyond conscious choice and intention. The unpredictability and irrationality entailed by such a view is substantial.
To sum up the points made above, it may be said that while both “character” and “personality” belong to the discourse of individual differences, only the former is involved in the normative discourse of person’s moral worth and responsibility. The thesis that the notion of character, but not that of personality, belongs to the discourse of responsibility should be taken here as conceptual. What is claimed, in particular, is that linguistic schemes involving the former notion usually involve the notion of responsibility as well and allow us to meaningfully hold somebody responsible for his/her character. Language games involving both concepts, in other words, make it a permissible, and actually quite a common, “move” to be made. Whether and, if yes, under what circumstances such a “move” is metaphysically and ethically justified is a logically separate issue, which won’t be addressed here.
In those accounts in which the connection between character and responsibility is considered stronger, i.e., as making responsibility claims not only conceptually possible but also justified, a separate account of responsibility is needed (e.g., Miller, 2013a, p. 13). One possible ground on which such an account can be developed is the relationship between character and reasons (as opposed to mere causes). Goldie (2004), for instance, emphasizes the reason-responsiveness of character traits: the fact that they are dispositions “to respond to certain kind of reasons” (p. 43). Actually, he even defines a virtue as “a trait that is reliably responsive to good reasons, to reasons that reveal values” (p. 43, emphasis removed; cf. the definition by Miller, 2013b, p. 24). A vice, accordingly, would be a disposition responsive to bad reasons.
Whether all personality traits are devoid of reason-responsiveness is not altogether clear (cf. Goldie, 2004, p. 13). For the notion of personality proper to academic psychology the answer would probably depend on a particular theoretical model employed. There would be a substantial difference, for instance, between personality understood, along the behavioristic lines, as a disposition to behavior and more full-fledged accounts allowing emotional and, especially, cognitive dispositions. What seems to be clear is the importance of reason-responsiveness for character traits.
Personality, character, and the fact–value distinction
Please note that none of the following has any connection to “scientific activity”
As soon as the notions of character and personality have been both historically and systematically explored some further and more detailed analysis can be conducted. The first issue to be addressed arises from the fact that the distinction between character and personality notions “looks sharpest when the word good is introduced”
, p. 8). One may argue, in particular, that all the dimensions enumerated in the above section including depth, ethical nature, relationship to the discourse of personal worth and responsibility, as well as reason-responsiveness are strictly connected with what has come to be called the fact–value distinction. The latter distinction, more specifically, seems to be in a sense parallel to the dichotomy between personality and character, respectively.
The fact–value distinction is usually derived from some remarks in David Hume’s (1738/2014, p. 302) Treatise of Human Nature, in which the idea of the logical distinctiveness of the language of description (is) and the one of evaluation (ought) was expressed. A relatively concise passage by Hume soon became very influential and gave birth not only to a distinction, but actually to a strict dichotomy between facts and values (cf. Putnam, 2002). A methodological prescription “that no valid argument can move from entirely factual premises to any moral or evaluative conclusion” (MacIntyre, 2013, p. 67) was its direct consequence.
In order to refer the above dichotomy to the notions of character and personality, it may be helpful to remember Allport’s (1921) idea of character being “the personality evaluated according to prevailing standards of conduct” (p. 443). A crucial point to be made here is that the act of evaluation is considered as an addition of a new element to an earlier phenomenon of personality, which can be comprehended without any reference to normativeness. The latter notion, in other words, is itself morally neutral: “There are no ‘moral traits’ until trends in personality are evaluated” (Allport, 1927, p. 285).
The thesis that personality can be specified independently of character or more generally, without any application of normative terms, is of considerable importance because it illustrates the fact that the character–personality distinction logically implies the fact–value one. The validity and the strictness of the former, in consequence, rely on the same features of the latter. Character and personality, in brief, can be separated only as long as it is possible to isolate personality-related facts from character-related values.
Such dependence should necessarily be referred to contemporary criticism of the fact–value distinction (e.g., MacIntyre, 2013; Putnam, 2002; cf. Brinkmann, 2005, 2009; Davydova & Sharrock, 2003). This criticism has been voiced from different perspectives and involves at least several logically distinct claims. For the present purposes, however, it is an argument appealing to so-called thick ethical concepts14 and the fact–value entanglement that is of most direct significance.
The distinction between thick and thin ethical concepts was first introduced (in writing) by Bernard Williams (1985/2006)15 and subsequently subjected to intense discussion (for useful introductions see Kirchin, 2013; Roberts, 2013; applications for moral psychology can be found in Fitzgerald & Goldie, 2012). What is common to both kinds of concepts is that they are evaluative: they “indicate some pro or con evaluation” (Kirchin, 2013, p. 5). Thick concepts, furthermore, are supposed to provide some information about the object to which they refer (information, which thin concepts do not provide). They have, in other words, “both evaluative conceptual content … and descriptive conceptual content … are both evaluative and descriptive” (Kirchin, 2013, pp. 1–2). If I inform somebody, for instance, that person A is good and person B is courageous, it is obvious that my evaluation of both A and B is positive. At the same time, however, the person informed doesn’t seem to know much about a good (thin concept) person A, whereas he/she knows quite a bit about a courageous (thick concept) person B.
The significance of thick concepts for philosophical discussion is usually connected with some “various distinctive powers” they supposedly possess. More specifically, when they are interpreted along the lines of a so-called non-reductive view they seem to have “the power to undermine the distinction between fact and value” (Roberts, 2013, p. 677).16 The non-reductive position is usually introduced as a criticism of the reductive idea that thick concepts “can be split into separable and independently intelligible elements” (Kirchin, 2013, p. 8; cf. the idea of dividing character into two parts mentioned above) or, more specifically, explained away as a combination of (supposedly pure) description and thin evaluation. If such a reduction was successful thick concepts would turn out to be derivative and lacking philosophical importance.
Many authors, however, including notably Williams (1985/2006), McDowell (1981), and Putnam (2002), claim that no such reductive analysis can be conducted due to the fact–value entanglement characteristic of thick concepts. The latter, as is argued, are not only simultaneously descriptive and evaluative, but also “seem to express a union of fact and value” (Williams, 1985/2006, p. 129). The fact–value entanglement proper to thick concepts becomes apparent if one realizes that any attempt to provide a set of purely descriptive rules governing their application seems to be a hopeless endeavor. One cannot, for instance, develop a list of necessary and jointly sufficient factual criteria of cruelty.17 It is obviously possible “to describe the pure physical movements of a torturer without including the moral qualities” (Brinkmann, 2005, p. 759), but it would yield a specification which comes dangerously close to the description of some, especially unsuccessful, surgical operations. In order to convey the meaning of the word “cruelty” (and to differentiate it from the phrase “pain-inflicting”) one needs to refer to values and reasons (rather than facts and causes only). An evaluative perspective from which particular actions are recognized as cruel, accordingly, must be at least imaginatively taken in order to grasp the rationale for applying the term in some cases, but not in others. Communication using thick concepts, as a result, turns out to be value-laden through and through.
The above-given features assigned to thick concepts by the non-reductionists are crucial due to the fact that they cannot be accounted for within the framework of the fact–value distinction. As such they are often believed to “wreak havoc” (Roberts, 2013, p. 678) with the latter or, more precisely, to undermine “the whole idea of an omnipresent and all-important gulf between value judgments and so-called statements of fact” (Putnam, 2002, p. 8).
The undermining of the sharp and universal dichotomy between facts and values has a very direct bearing on the character–personality distinction being, as emphasized above, dependent on the former. A crucial point has been made by Brinkmann who noticed that almost “all our words used to describe human action are thick ethical concepts” (2005, p. 759; cf. Fitzgerald & Goldie, 2012, p. 220). And the same applies to the language of character which, contrary to Allport’s expectations, cannot be neatly separated into the factual core of personality and the normative addition. The distinction between the notions of character and personality, in consequence, even though often applicable and helpful, cannot be inflated into a sharp dichotomy.
The functional dimension of the notions in question
Having analyzed the reliance of the character–personality distinction on the dichotomy between, respectively, value and fact, it becomes possible to carry out the second detailed investigation devoted to the functions played by the two concepts scrutinized. A good starting point for this exploration may be a remark made by Goldie (2004) who, while discussing the omnipresence of the discourse of personality and character, noticed that it is “everywhere largely because it serves a purpose: or rather, because it serves several purposes [emphasis added]” (p. 3). These functions merit some closer attention because they can help to further specify the difference between the concepts investigated.
The purposes served by the discourse of individual differences have been briefly summarized by the abovementioned author when he said that we use it “to describe people, to judge them, to enable us to predict what they will think, feel and do, and to enable us to explain their thoughts, feelings and actions” (and to control, manipulate and abuse them) (Goldie, 2004, pp. 3–4; cf. an analogous list provided by Miller, 2013b, pp. 12–13). Some of these functions are common to the notions of character and personality. Some others, however, are proper to the concept of character only.
The first of the common functions is description. The language of character and personality can serve as a kind of shorthand for the longer accounts of the actions taken. When asked about the performance of a new employee, for instance, a shift manager can say that he/she is certainly efficient and hard-working (rather than mention all particular tasks that have been handled). Similarly, if we say that A is neurotic, B is extraverted, C is just, and D is cruel, we do convey some non-trivial information about A, B, C, and D, respectively (even though our utterances may include something more than a mere description).
The second of the purposes that can be served by both concepts is prediction. We may anticipate, for example, that neurotic A will experience anxiety in new social situations. Despite the fact that such a prediction will be inevitably imprecise and fallible, it does enable us to narrow down “the range of possible choices and actions” (Goldie, 2004, p. 67) we can expect from a particular agent.
In fact, predictions regarding human behavior are notoriously inaccurate “guesses” – note the inability of the Psych Industry to identify mass shooters before they act.)
The notions of character and personality, furthermore, can be employed as a means of judgment. At this point, however, an important qualification needs to be made. If this function is to be assigned to both concepts it can be understood only in a weak sense of judging as providing an instrumental assessment. The ascription of personality traits of neuroticism and extraversion to A and B, respectively, can be used to say that A would not make a good candidate for an assertiveness coach, whereas B may deserve a try in team work. It falls short, however, of any moral judgment, which can be made only by means of character-related notions.
The concepts of personality and character, finally, can both be used to provide explanation. We can, for instance, say that C was chosen as a team leader because he/she is just and expected to deal fairly with potential conflicts. Having assigned an explanatory role to “character” and “personality,” however, one should necessarily remain fully aware of the experimental results reported in the first section. An appeal to “character” and “personality” as explanatory constructs does not have to mean that they provide the whole explanation. Situational factors still count and, as a matter of fact, one may need to acknowledge that in a “great many cases … [they] will be nearly the entire explanation” (Kupperman, 1991, p. 59).
One other reservation concerns the kind of explanation conveyed by the personality- or character-trait ascription. Human behavior, in particular, can be explained in at least two distinct ways (e.g., Gill, 1990, p. 4). Explanation, to begin with, can be made along the lines of the deductive-nomological model and refer to causes and natural laws. In such cases it is not substantially different from explanations of natural facts (like an earthquake) offered by the sciences. And it is this kind of explanation that is provided when non-reason-responsive features of personality are appealed to (cf. Goldie, 2004, p. 66).
Human action, however, can be also made comprehensible by the reference to reasons behind it. If we know what a person in question “values or cares for,” in particular, we can “make sense [emphasis added] of the action, or make the action intelligible, understandable or rational [emphasis added]” (Goldie, 2004, p. 65). Such an “explanation” can be given by the indication of only those traits, which are reason-responsive and, strictly speaking, is much closer to Dilthey’s (1894/2010) understanding (Verstehen) than to naturalistically understood explanation (Erklären).
The functions of description, prediction, instrumental assessment, and explanation (at least as far as the latter is understood in terms of causes) are common to both concepts of “personality” and “character.” The latter notion, however, can serve some additional purposes, which give it a kind of functional autonomy. Among the character-specific functions, to begin with, there is moral judgment. When we say that C is just and D is cruel we don’t make an instrumental and task-relative assessment. Rather, we simply evaluate that C is a morally better person than D (other things being equal). With this, the function of imposing moral responsibility is often connected. The issue of the validity of such an imposition is very complex and controversial. Still, it does remain a discursive fact that the claim that D is cruel is usually associated with holding D, at least to some extent, responsible for his/her cruelty.
Note that “pathological, disordered, mentally ill, socially defective” are labels every bit as “moral / judgmental” in the “real social environment” as “sinful, perverted, possessed by demons, Godless atheist, or an “agent of Satan”
The functions of moral judgment and moral responsibility ascription are not typically served by the scientific notion of personality. They may, however, become formally similar to description, explanation, and prediction if they are, as is often the case, applied within mostly third-personal language (as judging others and imposing responsibility on others). Apart from these functions, however, the notion of character can fulfill some essentially first-personal kind of purposes. And it is the latter that seems to be its most specific feature.
Among the first-personal functions of “character,” identification is fundamental, both psychologically and conceptually. When a person identifies with a character trait or, more holistically, with a complete character ideal, she begins to consider such a trait or character as a part of her identity (cf. Goldie, 2004, pp. 69–70): as something she “decides to be or, at least, to see herself as being” (Kupperman, 1991, p. 50). Such an identification, if serious, is very rich in consequences: it establishes “the experienced structure of the world of direct experience as a field of reasons, demands, invitations, threats, promises, opportunities, and so on” (Webber, 2013, p. 240) and helps one to achieve a narrative unity of one’s life (cf. Goldie, 2004; Kuppermann, 1991; McKinnon, 1999).
First-personal functions of the character notion, additionally, enable the agent to undertake more specific self-formative acts such as evaluating oneself against the idealized self, structuring moral progress, or providing motivation needed to cope with the difficulties of moral development. The notion of character employed in such a way becomes a kind of an internalized regulative ideal with a considerable emotional, imaginative, and narrative dimension. Its specific purposes are self-evaluative, self-prescriptive, and self-creative (rather than descriptive, predictive, and explanatory). The criteria of its assessment, accordingly, should be at least partially independent from those proper to strictly scientific constructs.
The latter fact, as may be worthwhile to mention, has a direct bearing on the challenge of situationism mentioned at the beginning of these analyses. The arguments in favor of this disquieting position have typically referred to experiments indicating that situational variables possess much bigger explanatory and predictive value than those related to personality and concluded that the usefulness of the personality concept needs to be seriously questioned. The doubts concerning the notion of character usually followed without further ado. No special attention, in particular, was paid to the assumption that the concepts of character and personality fulfill the same functions of description, explanation, and prediction. Accordingly, it was usually taken for granted that the failure of the latter concept automatically entails the uselessness of the former.18 As far as it is admitted that such an approach is at least partially erroneous, it may be worthwhile to refocus the debate towards the specific, first-personal, and normative functions of the notion of character. Do we need the latter to perform them and, if so, does this notion really serve us well, even though it is scientifically weak?
Some final remarks
As soon as the notions of character and personality have been scrutinized on the historical, conceptual, and functional dimensions it may be worthwhile to put forward some suggestions concerning the ways in which the results obtained can bear on the recent interdisciplinary discussions. The analysis of the logical dependence of the character–personality distinction on the strictness of the dichotomy between value and fact has already shown that debates concerning the latter should be carefully followed by any researcher who studies character and personality in their mutual relationship.
An important clarification, however, that needs to be made here is that any skepticism concerning the fact–value dichotomy suggested by some features of thick concepts should not be conceived by the psychologists as a call to develop a prescriptive and moralistic science of character and, thus, to become “like priests” (too late: this is where American Psychology stands today) (Charland, 2008, p. 16). A false impression that it is the case might result from the conflation between the full-fledged version of the fact–value distinction and the original, and relatively modest, Humean dictum that “no valid argument can move from entirely factual premises to any moral or evaluative conclusion” (MacIntyre, 2013, p. 67).19
That it is the latter that most psychologists care about can be clearly seen in two recent papers by Kendler (1999, 2002), who issues a stern warning that any psychological project developed along the lines of what he calls “the enchanted science”20 and motivated by the belief that psychology itself can discover moral truths can lead not only to Gestalt psychology’s holism or humanistic psychology, but also to the quasi-scientific justification of “Nazi and Communist ideology” (1999, p. 828). And it is in order to prevent these kinds of abuses that Kendler (1999) refers to what he calls “the fact/value dichotomy” or “an unbridgeable chasm between fact and values” (p. 829). By this, however, he does not seem to mean anything more than that “empirical evidence can validate factual truth but not moral truth” (p. 829). An example he provides considers the possibility of obtaining reliable empirical data supporting the thesis that bilingual education is advantageous for ethnic identification, but disadvantageous for academic development. Such data, as he rightly insists, would still leave it to the society to decide which value, ethnic identification or academic progress, should be given priority.
All of this, however, does not need to lead one to the acceptance of the fact–value dichotomy in the strong version that has been criticized by Putnam, McDowell, and others. Rather, it is the is–ought dichotomy which seems to be sufficient. The subtle differences between these two distinctions have been clarified by Dodd and Stern-Gillet (1995) who argue that the Humean dictum can be best understood as a general logical principle without any substantive metaphysical dimension of the kind usually connected with the fact–value dichotomy. That the is–ought gap is narrower and weaker is also illustrated by the fact that it is confined to “ought” statements with a considerable amount of other evaluative statements left aside. The examples provided by the authors are the aesthetic language of art and, importantly, the virtue ethical discourse of character. And as the ascription of beauty to a painting does not automatically entail any particular prescription,21 so does the assignment of courage or foolishness to a person. Even though such a feature of the characterological language has often happened to be conceived as a weakness within metaethical contexts, it can be arguably beneficial to all those psychologists who want to study the complexities of character without making an impression that any particular normative position can be derived from purely scientific study. A substantial amount of normativity, as shown by the example of thick concepts, will obviously remain inevitable, but it is certainly worthwhile to emphasize that it is mostly placed before empirical research as an evaluative framework taken from elsewhere and, thus, subjected to criteria and authorities of a non-empirical nature.
This paper has been written during a visit to Oxford University’s Faculty of Philosophy. I am greatly indebted to Edward Harcourt for all his help and support.