Abstract
The positive/negative distinction works well in many fields—for example, in mathematics negative numbers hold their own, and in medical pathology negative results are usually celebrated. But in positive psychology negativity should be replaced with positivity for flourishing/optimal functioning to occur. That the designation of the psychological states and processes deemed positive (good/desirable) and negative (bad/undesirable) is made a priori, independent of circumstantial particularity, both intrapersonal and interpersonal, does not seem to bother positive psychologists. But it should, as it results in conceptual muddles and dead ends that cannot be solved within their conceptual framework of positivity and negativity. Especially problematic is an ambiguity I find in positive psychologists’ a priori and a posteriori understandings of positivity and negativity, an ambiguity about constitutive and causal relations that pervades their science and the conclusions drawn from it. By eliminating their a priori dichotomy of positivity and negativity, positive psychologists might well find themselves in a better position to put back together the psychological reality that they have fractured in their ontologically dubious move of carving up psychological reality a priori into positive and negative phenomena. They then might find themselves better placed to “broaden and build” their own science of flourishing.
Keywords
The positive psychology movement, in self-proclaimed distinction to all prior positive mentality movements, famously claims to have founded itself on the rock of science. Of its many scientific findings, none has been trumpeted more loudly than Barbara Fredrickson’s now (in)famous “positivity ratio,” in which 2.9 positive emotions to each negative emotion (which, following Fredrickson, 2009, I will call the 3:1 ratio 1 ) constitutes the “tipping point” that allegedly predicts flourishing over languishing and so is said to validate Fredrickson’s “broaden-and-build theory of positive emotions” (Fredrickson, 2009, 2013; Fredrickson & Losada, 2005). The recent debate between Fredrickson (2013) and Brown, Sokal, and Friedman (2013, 2014), who convincingly challenged Fredrickson and Losada’s (2005, 2013) “nonlinear dynamics model” of positive emotions on methodological grounds, 2 has for positive psychologists tarnished neither the luster of their positivity-ratio pearl nor the motivating positive/negative dichotomy at its core. Although some may counter that conceptions of the positive/negative dichotomy are evolving within the movement—the hedonic concept “happiness” has been eclipsed by the eudaimonic 3 concept “flourishing” (Fredrickson, 2009; Seligman, 2011) 4 —positive psychologists’ conception of flourishing depends logically on positivity and negativity as conceived from the movement’s start (Peterson & Seligman, 2004; Seligman, 2002). And so the positive/negative dichotomy constitutes positive psychology’s movement-defining conceptual foundation.
Does the positive/negative dichotomy give the movement a sound conceptual foundation on which to build its science of flourishing? I make no quarrel with the positive/negative distinction as it appears in mathematics, where negative numbers hold their own, nor in medical pathology, where negative results usually indicate the absence of pathology and so are to be celebrated. In positive psychology, by contrast, negative characteristics should be diminished and replaced with positive characteristics, for flourishing (as well as happiness) to occur. And so we see the polysemantic nature of the terms positive and negative, each of which can mean either good or bad (or neutral) depending on their domain of application. Yet positive psychologists make unwavering a priori designations of positivity (or good/desirable) and negativity (or bad/undesirable) in reference to psychological states, functioning, and processes. These designations have created considerable conceptual muddles and dead ends that defy elimination within positive psychologists’ own foundational conceptual framework of positivity and negativity, muddles that infect their empirical science and the conclusions they draw from it—or so I argue.
In what follows I analyze the nature of the ambiguity built into the very meanings of the terms positive and negative as they are used in positive psychology. If, as I maintain, the psychological characteristics that positive psychologists designate positive and negative do not meet their own criteria for what positivity and negativity are purported by them not only to be (constitutively) but also to do (in causally influencing adaptive and maladaptive functioning), then the very idea of a positive (or negative) psychology is bankrupt, or dubious at best. Put differently, if the foundational, movement-defining dichotomy can be shown to be a form of false dichotomy, one that cannot be cashed out in real kinds that meet positive psychologists’ constitutive and/or causal criteria for the use of the terms positive and negative, then the movement sinks in conceptual and/or empirical quicksand.
This last proposition, if valid, casts Fredrickson’s well-accepted interpretation of her 3:1 positivity ratio in a questionable light. I am, then, challenging nothing less than the validity of the movement’s claims about the beneficial effects of positivity and the detrimental effects of negativity, as the terms positivity and negativity have been defined, designated, and applied by positive psychologists.
To be clear, I concede that positive psychologists may validly measure such favorite positives as gratitude, kindness, optimism, forgiveness, courage, hope, awe, engagement, and so on. 5 My complaint, rather, resides in their assumption that it is valid to call these characteristics positives a priori (virtue ethics notwithstanding), if their professed causal relation to flourishing or optimal functioning 6 is in many cases dubious at best—or at least far less certain than is given in Fredrickson’s 3:1 ratio and 11.6:1 ratio of positive to negative emotions, in which flourishing allegedly occurs within those ranges and languishing outside of them. 7 For even if we stipulate that these characteristics are positive because they feel good or pleasant—which is Fredrickson’s (hedonic) a priori criterion for a positivity designation (Fredrickson & Losada, 2005, p. 678)—this does not mean that empirically these characteristics will turn out to be good for enhancing eudaimonic well-being, flourishing, or optimal functioning. Yet it is just this causal claim that “justifies” pushing positivity so aggressively and ubiquitously.
I begin by explicating what I mean by positive psychologists’ a priori designation of positive and negative characteristics, despite their a posteriori appeal to experience/observation in deriving their dichotomy and in promoting its application in clinical contexts and beyond. This I follow with analysis of the conceptual quicksand of positive psychologists’ designations of psychological states, processes, functioning, and interventions as either positive or negative, including their seeming conflation of constitutive and causal relations in their conceptualizations of “positivity” and “negativity”—here I refer to ambiguity and inconsistency in their a priori and a posteriori use of those terms. I then review how these conceptual problems have undermined their empirical science, such that their scientific claims about the effects of positivity and negativity should also be called into question. I consider possible reasons why the many robust findings that contradict their claims about the causal consequences of positive and negative characteristics have eluded them, and how their attempts to integrate positivity and negativity in their science and practice (in their professed quest for “balance” 8 ) cannot help them achieve their goal of enhancing flourishing/optimal functioning. In short, a complete, comprehensive, and accurate science of human flourishing can be better served by a psychology that does not rely on the dubious a priori dualism of positivity and negativity on which positive psychology rests.
Positive Psychology and the A Priori
Standard Definitions of “A Priori”
The term a priori refers to propositional knowledge that is not based on or justified by empirical evidence; the proposition is known “independent of any experience other than the experience of learning the language in which the proposition is expressed” (Baehr, 2003). In contrast a “proposition that is knowable a posteriori is known on the basis of experience. For example, the proposition that all bachelors are unmarried is [known] a priori” (Baehr, 2003), and the proposition that “all bachelors in the U.S. are taxed at a different rate from married men” is known a posteriori (Russell, 2014). Both Baehr (2003) and Russell (2014) emphasized debate about proper meanings of the terms independent and experience in these definitions, discussion of which space constraints preclude. Suffice it to say that a priori knowledge does not require empirical justification, whereas a posteriori knowledge does. Thus, both terms are epistemic concepts; they refer to the basis on which a proposition can be said to be warranted.
We are now in a position to ask, In just what way may positive psychologists’ designations of positive and negative characteristics be characterized as a priori designations? Six universal virtues and 24 signature-strength routes to those virtues were so designated by Seligman (2002) and Peterson and Seligman (2004); they capture a great many if not all the characteristics that positive psychologists deem positive.
9
They were derived and so designated on the basis of both conceptual and empirical considerations, according to Kristjánsson (2010): So how were these virtues and strengths derived? The answer is: through a variety of considerations, some conceptual (such as being measurable, trait-like, distinct, nonexclusive of others), and some historical (being ubiquitous and morally valued across cultures, being recommended by the world’s most influential religious and philosophical traditions . . . see Peterson & Seligman, 2004, pp. 21-28). (p. 305)
To be sure, historical documents are a kind of empirical evidence; whether the propositions they contain are true or false, they themselves constitute observable evidence as documents. Here I make the more relevant point that an empirical/scientific element slips into the derivation of the positive characteristics (at least those that are considered virtues), despite the fact that they are alleged to be conducive to flourishing on a priori grounds. As Kristjánsson (2010) put it, The virtues are at once conducive to and constitutive of eudaimonia; each true virtue represents a stable character state . . . that is intrinsically related to flourishing as a human being. Notably, Aristotle’s eudaimonia is an explicitly moral notion, not conceptually, but empirically: it is in fact impossible to achieve eudaimonia without being morally good—without actualizing the moral virtues. (p. 301)
And herein lies the problem that Kristjánsson glossed over in his otherwise penetrating analysis: If positive characteristics are positive not only because they are constitutive of eudaimonia/flourishing but also because they are conducive to eudaimonia/flourishing when enacted, then they may be rightly said to play a causal role in the attainment/actualization of eudaimonia/flourishing. In that case, there is both a constitutive and a causal/empirical component in the very meaning of positivity in psychology. To be sure, this interpretation hinges on the meaning of the term conducive to, which ambiguously carries both constitutive and causal connotations in the positive psychology literature, as Kristjánsson signaled (2010, pp. 301, 304, 305). But if “conducive to” sometimes means “causes,” then positive characteristics both constitute and cause eudaimonia, even though the causal relation is logically excluded by the constitutive relation. Put differently, if the conception of “positivity” is not logically independent of the conception of “flourishing,” then these causal claims are vacuous; they are tantamount to saying, “having lots of money (which constitutes being rich) is conducive to being rich.” More generally, the distinction between constitutive and causal relations in psychology needs conceptual work (see Held, 2007; Hibberd, 2005; Smedslund, 1991, 1994 10 ), especially since philosophers disagree about what constitutive relations entail (e.g., Evnine, 2011, 2016).
The inconsistent use of constitutive and causal talk pervades positive psychology. For example, early on Peterson and Seligman (2004) made their constitutive case decisively: Well-being is not a consequence of virtuous action but rather an inherent aspect of such action. . . . Outcomes of the strengths (fulfillments) do not show up at a later point in time, caused . . . by the strength, but instead are part and parcel of the actions that manifest the strength. . . . [Doing a favor] does not cause you to become satisfied . . . ; being satisfied is an inherent aspect of being helpful. (p. 18)
And even earlier Seligman (2002) said, “A strength is valued in its own right. . . . Although strengths and virtues do produce . . . desirable outcomes, we value a strength for its own sake, even in the absence of obvious beneficial outcomes” (p. 137). Yet the causal relation is touted vociferously. Thus, however positive psychologists derived their designations of positive and negative characteristics, they have justified or validated those designations on the basis of their empirical findings of positivity’s causal effects on flourishing/optimal functioning. This was seen from the movement’s start, in demonstrations of how positivity causes subjective/hedonic well-being, which became increasingly linked to the ever more important psychological/eudaimonic well-being, flourishing, or optimal functioning. Indeed, if positivity referred only to characteristics that inherently feel good/pleasant and/or that cause us to feel good, then positive psychology would sink back into the hedonic well-being (“happiology”) that attention to eudaimonic well-being is intended to transcend (Kristjánsson, 2010, p. 304).
This transcendence motivates the movement’s unwavering push for positivity: Positivity not only feels good, it is good (in the constitutive appeal to virtues ethics) and good for you (because it causes flourishing/optimal functioning). And once so “validated” empirically, the a priori designation of positive characteristics becomes reinforced; those characteristics are more likely to be treated as justified in their a priori use—by this I mean that designations of positive and negative characteristics are more likely to be made independently of the particularities of the persons to whom those terms are applied. It is this meaning of “a priori” that I use in my exposition of the a priori problem in positive psychology, in which circumstantial particularity, knowable only a posteriori in any given case, can enjoy no empirical role in modifying the a priori designations of positive and negative characteristics—those a priori designations have become entrenched, with supposed a posteriori warrant. Thus, if positive psychologists believe that their a posteriori case is closed (owing to their own causal findings), then their a priori designations cannot be challenged by the extensive contradictory a posteriori evidence/knowledge that they have blissfully ignored. In this I propose negative implications of the conflated, ambiguous, and/or inconsistent use of constitutive and causal relations, of a priori and a posteriori knowledge/justification, which I find in positive psychology and which, if operative as theorized, undermines the touted scientific justification for positive psychologists’ claims about the nature and application of their positive/negative dichotomy.
Hedonic Versus Eudaimonic Well-Being/Flourishing
In his book Flourish, Seligman (2011) officially shifted the mission of positive psychology from his original authentic happiness theory of hedonic well-being to a eudaimonic well-being theory of flourishing or optimal functioning (see Notes 3 and 4): The goal of positive psychology in authentic happiness theory is . . . to increase the amount of happiness in your own life and on the planet. The goal of positive psychology in well-being theory is . . . to increase the amount of flourishing in your own life and on the planet. (p. 26)
Seligman here defined “flourishing” by way of five constitutive elements that form his PERMA acronym: “positive emotion, engagement, meaning, positive relationships, and accomplishment” (p. 16). Happiness and flourishing are said to be related. But whereas for Seligman positive emotion is a constitutive element of flourishing, for Fredrickson positive emotions stand in causal relation to flourishing: Thus, we see that flourishing is conceptualized not only differently but also incompatibly, by two major theories (PERMA and broaden and build) within the movement.
No one has done more to promote a science of flourishing than Fredrickson. In her “broaden-and-build theory of positive emotions,” Fredrickson asserts that positive emotions cause various forms of well-being, all of which she considers forms of flourishing (eudaimonic well-being) or optimal functioning (Fredrickson, 2009, pp. 14, 181; Fredrickson & Losada, 2005, p. 685; see Note 6). Garland, Fredrickson, Kring, Johnson, Meyer, and Penn (2010) made that causal relation clear: Negative emotions have long been held to narrow the scope of people’s attention and thinking. . . . The broaden-and-build theory . . . holds that positive emotions broaden individuals’ thought–action repertoires, enabling them to draw flexibly on higher-level connections and wider-than-usual ranges of percepts, ideas, and action urges; broadened cognition in turn creates behavioral flexibility that over time builds personal resources, such as mindfulness, resilience, social closeness, and even physical health. (p. 850)
Earlier, Fredrickson and Joiner (2002) claimed that positive emotions not only cause/initiate optimal functioning (here called broad-minded coping) but also that such coping in turn enhances positive emotions; this bidirectional causality constitutes her “upward spiral of positivity,” in which hedonic well-being happily enhances eudaimonic well-being—and vice versa: Positive emotions broaden the scopes of attention and cognition, and, by consequence, initiate upward spirals toward increasing emotional well-being. . . . Initial positive affect, but not negative affect, predicted improved broad-minded coping, and initial broad-minded coping predicted increased positive affect, but not reductions in negative affect. Further mediational analyses showed that positive affect and broad-minded coping serially enhanced one another. (Fredrickson & Joiner, 2002, Abstract, p. 172)
And so we see that, by way of broad-minded coping, positive emotions ultimately increase positive emotions—seemingly without their spiraling out of control or anyone’s worrying about that entailment of the theory. Yet Fredrickson’s notion of positivity extends beyond positive emotions. Thus we may now ask, What are the states and processes that her positivity concept picks out, and on what basis they are so selected?
Positive Psychology’s Positive Characteristics
Fredrickson and Losada (2005) expressly defined the states and processes to which positivity refers: We use affect to represent [a] spectrum of valenced feeling states and attitudes, with positive affect and positivity interchangeably representing the pleasant end (e.g., feeling grateful, upbeat; expressing appreciation, liking) and negative affect and negativity representing the unpleasant end (e.g., feeling contemptuous, irritable; expressing disdain, disliking). (p. 678)
Positivity, then, is constituted by both pleasant emotions/moods and pleasant evaluative sentiments/attitudes—this is the hedonic feature of positivity, which is presumably known a priori because positive characteristics are constituted by their inherent pleasantness. 11 And positivity so constituted allegedly causes flourishing—this is the eudaimonic effect of positivity, which is known a posteriori. In this hedonic well-being happily meets eudaimonic well-being/flourishing. It is nontrivial that positive psychologists lay claim to strong evidence for the causal relation between hedonic well-being and eudaimonic well-being/flourishing: If positive characteristics were nothing more than feel-good states that cause hedonic well-being (or increased feel-good states 12 ), then positive psychologists’ determination to recast “happiology” in a supporting role to flourishing’s top billing would be thwarted. Moreover, if the causal hedonic–eudaimonic relationship obtains as claimed, then positive psychologists could justify their mission to enhance positivity as defined a priori (hedonically) by them—that is, defined prior to or independently of anyone’s circumstantial particularity.
Fredrickson (2009) listed the 10 “forms of positivity” or “positive emotions” that are said to be the “most common” forms of positivity (p. 39): joy, gratitude, serenity, interest, hope, pride, amusement, inspiration, awe, love. These top 10 allegedly instantiate positivity most representatively; they also overlap considerably with Seligman’s 24 signature-strength routes to his 6 ubiquitous virtues (see Note 9). These pleasant feeling/attitudinal states, then, constitute the common denominator that drives a considerable body of research on positivity. Do researchers outside of positive psychology accept a hedonic definition of positivity? If not, what are the “boundary criteria” for the proper use of the terms positivity and negativity in picking out mental kinds/characteristics? As it turns out, there is vast disagreement about what makes a characteristic positive; the conceptual quicksand is deep indeed.
The Conceptual Quicksand
What Makes a Positive Emotion Positive?
In The Positive Side of Negative Emotions, editor Gerrod Parrott (2014) surveys the many meaning-makers of the terms positive emotions and negative emotions. He concedes that there is historical agreement about which emotions are positive and which are negative (Colombetti, 2005),
13
but finds no such agreement about “what makes an emotion positive or negative” (p. x). Regarding the boundary criteria or application conditions for the proper use of these two terms, Parrott cites Solomon and Stone (2002), who listed 18 ways in which positive and negative emotions have been distinguished. Some [are] ethical: virtue versus vice, right versus wrong, socially approved versus socially unacceptable. Others [are] emotions’ effects: healthy or unhealthy, calming or upsetting, strengthening or weakening, satisfying or dissatisfying, motivating approach or motivating avoidance. . . . Yet other interpretations focus on the various appraisals and judgments that are attached to the emotions: is the situation in accord with one’s wishes or not. . . . [And still others] refer to qualities of phenomenal experience: perhaps [positive] emotions are pleasant whereas [negative emotions] are painful. (Parrott, 2014, p. x)
The last criterion in the list, phenomenal character—pleasant or painful, is of course Fredrickson’s a priori basis for designation. And the second, an emotion’s effects, constitutes her empirical/a posteriori finding (her 3:1 positivity ratio), which, I contend, may seem to justify the a priori designations made by her and other positive psychologists. Parrott (2014) rightly adds, “One commonality is that both positive and negative emotions can be either functional or dysfunctional.” And so he is keen to study the “factors that help determine when an emotion will work adaptively in a particular context” (p. xiii).
Parrott does not reject the historical designation of positive and negative emotions, thereby perpetuating the dubious dichotomy. However, he and his contributors do not deploy it a priori but rather with an a posteriori eye toward circumstantial demands, thus allowing negative emotions (as defined historically) to have positive causal consequences, at least in principle. The retention of this non-a priori use of positivity and negativity can be seen as a corrective step along the way to a more complete and accurate psychological science of flourishing; it certainly paves the way for reevaluation of the a priori designation of any characteristic as positive or negative. In addressing what psychologists might mean when they expressly categorize emotions as positive or negative, Parrott says that the aforementioned 18 senses are problematic because they “do not categorize the same emotion consistently. For example, anger can be painful or pleasurable or both, depending on the circumstances, so phenomenal experience does not explain why anger is considered a negative emotion” (p. x). Thus, pleasantness cannot serve as the criterion for designating positive characteristics a priori, independent of circumstantial particularity, as positive psychologists suppose.
A bit earlier, Kristjánsson (2010) made a similar point: Among emotion researchers in psychology, the term “positive emotion” is most commonly used to denote pleasant emotions . . . and here the positive psychologists follow suit. . . . [Yet] most emotions are not wholly pleasant or wholly painful, but rather a mixture of the two (take love or anger). (p. 304)
What is of course needed is a “consistent basis for justifying why each emotion is classified as positive or negative” (Parrott, 2014, p. xi), either on a priori or a posteriori grounds. And Parrott finds “the most useful criterion [to be] the situation’s perceived compatibility with a person’s needs, goals, and values: negative emotions generally involve interpreting something as being against one’s wishes” (p. xi). The compatibility of a situation with each individual’s needs, goals, and values is a far different basis (from pleasantness) by which to designate an emotion as positive (compatible) or negative (incompatible) than we find in Fredrickson’s a priori universal designations: Whether any one emotion is positive or negative will, in virtue of Parrott’s compatibility criterion, vary both within and between persons, depending on the goals in play in relation to a particular circumstance in any given moment. For example, if one seeks to end a relationship experienced as destructive, then anger may be consistent with that evaluation and thus aid in attaining that goal (and so may feel pleasant). Therefore, we cannot say whether any emotion is positive or negative prior to knowing a person’s particular circumstances, especially the challenges that are faced in relation to needs, goals, and values in any given life situation.
Kristjánsson (2010), like Parrott, enumerated the many “promiscuous” meanings of the term positive emotion: In addition to . . . (a) appropriateness and (b) pleasantness, “positive emotion” is sometimes used as a shortening for (c) “positively evaluating emotion” . . . and even for (d) “an emotion conducive to health.” . . . These meanings do not always coincide. Compassion, for instance, is typically (a) a morally positive (appropriate) emotion, but it is always (b) painful, and (c) negatively evaluating; moreover, the question of when and to what extent it is (d) conducive to health must be open to debate. . . . These promiscuities of meaning have led many theorists to conclude that the very term “positive emotion” obscures rather than illuminates, is detrimental to serious emotion-research, and would best be expunged from academic vocabularies (Colombetti, 2005; Kristjánsson, 2006, pp. 23-35; Lazarus, 2003, p. 99; Solomon & Stone, 2002). (p. 304)
And consistent with Parrott, Kristjánsson raises the conceptual question of what makes an emotion positive: “If . . . positive emotions simply are pleasant emotions, then positive psychology seems to have collapsed into the very hedonistic theory (mere ‘happiology’) that its leaders take such great pride in renouncing” (p. 304).
In this we see that Fredrickson’s a priori basis for classifying an emotion as positive puts her (and other positive psychologists) in danger of descending into a hedonic hole. But Fredrickson’s positivity ratio seemingly circumvents this danger, as three positive emotions to every negative emotion not only are alleged to cause hedonic well-being but also, more important, to cause eudaimonic well-being/flourishing. Kristjánsson summarizes Fredrickson’s answer to the question “of how positive psychology can avoid collapsing into ‘happiology’ if it one-sidedly valorizes positive affectivity,” which answer, he says, resides in the so-called broaden-and-build thesis that underlies positive psychology’s endorsement of pleasant emotions. It is argued that such emotions, first, broaden a person’s thought-action repertoire (attention, working memory, verbal fluency, openness to information) and, second, that such a broadened repertoire builds enduring personal resources. These effects gradually transform people, forming an upward spiral of increased activity and engagement which is conducive to happiness.
14
(p. 304)
Empirical support for Fredrickson’s “upward spiral” has been convincingly challenged in another debate between Fredrickson and her collaborators on the one side (Kok et al., 2013; Kok & Fredrickson, 2015) and Brown and his collaborators on the other (Heathers, Brown, Coyne, & Friedman, 2015). Whatever the outcome of that controversy, there is no consensus among emotion researchers about what makes a mental state positive or negative. These researchers include the many contributors to Parrott (2014), who find important positive causal consequences of negative emotions. Thus, even if all psychologists agreed that “positivity” denotes phenomenologically pleasant characteristics, and also agreed about which characteristics are phenomenologically pleasant, that agreement in itself would not justify positive psychologists’ many claims about (a) positivity’s beneficial causal consequences for everyone and so (b) their use of interventions designed to increase positivity and decrease negativity in every case that comes their way.
Positive Versus Negative Interventions
Positive psychology is sold not least on its positive interventions, both within and beyond clinical settings. If the a priori designation of positive and negative characteristics is problematic, then it follows that the a priori designation of positive and negative interventions, which distinction depends logically on the positive/negative characteristic distinction, is also problematic. And yet, the a priori positive/negative dualism/dichotomy holds for psychological interventions too (see Held, 2016; A. M. Wood & Tarrier, 2010).
Seligman (Seligman, Rashid, & Parks, 2006) led the way for positive psychologists to distinguish a priori negative interventions from positive interventions: Negative interventions directly target the decrease of negative/undesirable and/or pathological states and functioning, and positive interventions directly target the increase of positive/desirable and/or healthy states and functioning. And so we move with apparent a priori ease from positive versus negative states, processes, and functioning to positive versus negative interventions, despite this shift in the domains of application of the positive/negative distinction.
In promoting their positive clinical psychology, A. M. Wood and Tarrier (2010) said, “It is not logical to study either negative or positive functioning in isolation [from each other], as . . . this prevents interventions being designed to both decrease the negative and promote the positive” (p. 827; see A. M. Wood & Johnson, 2016). Here we catch a glimpse of the desire for a complete science by means of integrating positivity and negativity. But such “integration” is limited at best, because negativity remains devoid of constructive value: It should be diminished as much as possible in the service of enhancing positivity.
Positive interventions and negative interventions, then, are cast a priori as two distinct kinds of interventions, each of which expressly targets two distinct kinds of states and two distinct kinds of functioning, one desirable, the other undesirable. But it is at least logically conceivable that interventions designed to increase certain negative states under certain circumstances may enhance positive functioning. If that could be demonstrated empirically, the inviolate positive/negative divide would begin to blur.
Owing to a conflation (or inconsistent and ambiguous use) of a priori and a posteriori understandings of positivity and negativity, I prefer the terms constructive and destructive (or adaptive and maladaptive) because they at least hint at the necessity of empirical evidence to determine whether any one state or process is helpful or harmful in the pursuit of a particular goal by a particular person facing particular circumstantial demands. And interventions designed to (a) build up what we deem adaptive and (b) tear down what we deem maladaptive are both not only circumstance-dependent but also constructive, despite “deconstruction” in the latter case. This way of thinking entails obstacles for Fredrickson’s “broaden-and-build theory,” which severely limits the value of negative emotions in constructive endeavors (see Note 7). This limiting is also a profound limitation; it flies in the face of extensive empirical evidence that supports the constructive value of negativity, a value that evidently defies Fredrickson’s 3:1 to 11.6:1 positivity-ratio range. The issue, then, is whether a priori designations of positive and negative states and processes map onto the states and processes that causally contribute to adaptive and maladaptive functioning, respectively, as positive psychologists claim without regard for circumstantial particularity.
The crucial corollary that seems to get lost in the rush to positivity is this: Improving coping/functioning is not always compatible with feeling happy, with positive affect, and thus with interventions that are positive in virtue of aiming to increase positive or pleasant emotional (and cognitive/attitudinal) states directly. This corollary flies in the face of Fredrickson’s ratios and so must be supported by strong a posteriori evidence.
The Relentless Empirical Evidence
Parrott (2014) compiled 12 chapters written by psychologists who, collectively, have conducted three decades of research in which they demonstrate robustly how various negative emotions—including sadness, anxiety, embarrassment, shame, guilt, jealousy (and negative coping styles such as defensive pessimism)—can contribute causally to optimal functioning. 15 Given positive psychologists’ mission to vanquish negative emotions, one might think that they would take note of these movement-defying findings. Yet that has not been the case. And so we are moved to ask, What happens to positive psychologists’ a priori referents for positivity and negativity when those so-designated positive characteristics not only do not aid optimal functioning/flourishing, as they claim, but also can under many circumstances actually hinder it? Are they still considered positive characteristics then?
McNulty and Fincham (2012) have persistently challenged the a priori positive/negative labeling of traits and processes altogether, on conceptual as well as empirical (circumstantial) grounds. Regarding the conceptual, they state, “By defining the field of positive psychology as numerous psychological characteristics, Seligman and Csikszentmihalyi (2000) implied that the listed characteristics are inherently positive. But are they?” (p. 101). McNulty and Fincham deny that any such stipulations can be made independent of circumstance. And so they appeal to extensive empirical findings: “The psychological characteristics that benefit people experiencing optimal circumstances may not only fail to help people experiencing suboptimal circumstances, but may harm them” (p. 106). They demonstrate empirically how forgiveness, optimism, benevolent attributions, and kindness—favorite “positives” of positive psychologists—can diminish adaptive functioning.
16
They conclude, “An understanding of the complete human condition requires recognizing that psychological traits and processes are not inherently positive or negative—whether they have positive or negative implications depends on the context in which they operate” (McNulty & Fincham, 2012, pp. 107-108). And that context has much variability built into it: Psychologists need to move beyond examining the main effects of traits and processes that may promote well-being on average to study the factors that determine when, for whom, and to what extent those factors are associated with well-being. Failing to do so will result in an incomplete understanding of the contextual nature of psychological characteristics that could have harmful implications. (p. 106)
Here we see that a posteriori knowledge challenges or competes with a priori designations. Hedonics may not always harmonize with eudaimonics.
To conclude, the a priori designation of positive and negative characteristics arguably constitutes a form of false dichotomy. Contrary to positive psychologists’ designations, there are no characteristics to which the terms positive and negative refer consistently, if those terms mean the characteristics that feel pleasant versus unpleasant and/or cause adaptive versus maladaptive functioning, respectively. In fact, in many cases adaptive functioning is causally related to unpleasant feelings, and maladaptive functioning is causally related to pleasant feelings.
When “Negativity” Causes Positive Functioning: The Ignored Case of Defensive Pessimism
Having just considered empirical research in which positive characteristics have contributed to harmful outcomes, we are now in a better position to question their a priori classification as positive. I refer here to the wealth of research that demonstrates how negative coping strategies, including the negative emotions and thinking that constitute such strategies, can be highly advantageous to adaptive/constructive functioning. How have positive psychologists responded to these robust findings? The coping strategy of defensive pessimism, studied for some three decades by personality psychologist Julie Norem, provides the most prominent case in point. The title of her popular book, The Positive Power of Negative Thinking (2001), captures her message succinctly. Yet scant attention is paid in the positive psychology literature to Norem’s extensive scientific findings about the value of defensive pessimism.
Norem (2008) defined defensive pessimism as a “motivated cognitive strategy” that entails “setting low expectations (being pessimistic) and then thinking through, in concrete and vivid detail, all the things that might go wrong as one prepares for an upcoming situation or task” (p. 123). This “helps people manage their anxiety and pursue their goals” (Abstract, p. 121), by formulating contingency plans for possible disruptions in the execution of an anxiety-inducing task, which plans diminish debilitating anxiety. She emphasizes the (intra- and interpersonal) context-dependent nature of the strategy: Understanding how and why defensive pessimism works requires an integrated understanding [of] the role of traits, motivations, and self structures within the individual, the resultant goals toward which strategies are directed, and the particular constraints of different situations and cultural contexts. (Abstract, p. 121)
Norem explains how defensive pessimism works by means of specific kinds of negativity, kinds which she calls “the right tool for the job” (Norem, 2014). Thus, negative affect and thinking “function as positive motivation for defensive pessimists,” in that, unlike the negativity of rumination and catastrophizing, “negative reflections are directed toward the future, and focus on potential negative scenarios that are directly relevant to the [desired] situation or goal” (Norem, 2008, p. 126). What begins as a seemingly “negative” process in fact functions as a “positive” process: “The defensive pessimist is able to shift emphasis from anxious feelings to thoughts about possible specific problems, and then to actions to prevent those problems from derailing progress” (Norem 2008, p. 126; see Norem & Cantor, 1986).
Norem (2008) also explains how the use of any particular coping strategy must vary as a function of personality structure, for adaptive functioning to occur. For example, compared to her so-named “strategic optimists”—who are not anxious, have high expectations, and don’t worry about their performance—defensive pessimists report greater degrees of trait anxiety, neuroticism, lower self-esteem, and negative affect in general; they also “generate more negative potential outcomes and plans” (p. 124).
Those results do little to demonstrate that defensive pessimism is more than a generally negative view of self and the world; [we should then ask] why those using the strategy cannot just “lighten up,” especially given that they typically perform as well as the strategic optimists. . . . Yet, just as saying “hey, relax” to an anxious person rarely helps, the research evidence makes clear that simply trying to be more optimistic will not work for defensive pessimists. (p. 125)
Relatedly, Norem (2008) delineates the effects of positive versus negative experimental manipulations (i.e., interventions) with defensive pessimists: “Attempts to disrupt or make more optimistic any component of their strategy seem to interfere with the defensive pessimists’ performance, and lower their satisfaction after the fact. . . . Further analyses showed that anxiety indeed mediated these results” (p. 125).
And so it should come as no surprise that Norem’s extensive findings invoke a familiar punch line, namely, one size does not fit all: “Both defensive pessimists and strategic optimists perform best when allowed to pursue (or avoid) mental simulation according to their preferences” (p. 126). Yet that circumstance-dependent message is lost on those committed to a priori positivity stipulations. Referring to mood, Norem (2008) warns of the dangers of “cheering up” defensive pessimists, which can be accomplished but at considerable cost to adaptive functioning: “While it is possible to put defensive pessimists in a better mood, doing so leads to poorer performance (Norem & Illingworth, 2004; Sanna, 1998)” (p. 126).
Thus, even so-called positive interventions (those designed to increase positive states directly) can cause negative/deterioration effects. 17 For example, J. V. Wood, Perunovic, and Lee (2009) found that repeating positive self-statements (e.g., “I accept myself completely”) caused those with low self-esteem to feel even worse. They speculate that these may backfire for those with low self-esteem because the unsuccessful attempt to “avoid negative thoughts . . . may have signified that the positive self-statement was not true of them . . . the very people they are designed for” (p. 865).
With their determination to enhance adaptive functioning/flourishing, one might think that positive psychologists would happily embrace Norem’s extensive “positive” findings about defensive pessimism. Instead, her findings are most commonly ignored by positive psychologists, even in literature devoted to interventions/practice (e.g., Linley & Joseph, 2004; Parks & Schueller, 2014) and in positive psychology’s canonical texts, including the Handbook of Positive Psychology (Snyder & Lopez, 2002) and the Encyclopedia of Positive Psychology (Lopez, 2009). And when her findings are mentioned, they tend to be dismissed as “negative.” 18
For example, prior to the emergence of the positive psychology movement, Scheier and Carver 19 (1993) conceded that “defensive pessimism does seem to work,” in that defensive pessimists perform better than “real [i.e., dispositional] pessimists.” But they also claimed that “people who use defensive pessimism in the short run report more psychological symptoms and a lower quality of life in the long run than do optimists. Such findings call into serious question the adaptive value of defensive pessimism” (p. 29). And in Character Strengths and Virtues, Peterson and Seligman (2004) weighed in critically: “We do not deny that defensive pessimism can prove useful in some circumstances, but the relevant research also shows that defensive pessimists annoy others” (p. 528). They evidently miss the point of defensive pessimism, which is to harness debilitating anxiety so as to improve functioning. And need it be said that optimists can be annoying?
In any case, the best evidence does not support positive psychologists’ dismissive claims. Norem and Chang (2002), in comparing anxious people who used defensive pessimism to those who did not, reported that defensive pessimists show significant increases in self-esteem and satisfaction over time, perform better academically, form more supportive friendship networks, and make more progress on their personal goals than equally anxious students who do not use defensive pessimism. . . . Taking away their defensive pessimism is not the way to help anxious individuals. (p. 997)
Norem readily concedes precise benefits and costs of both strategic optimism and defensive pessimism. We may therefore ask why the negatives of defensive pessimism are typically considered “true negatives” by positive psychologists who ignore the negatives of strategic optimism. One answer is that positive psychologists who bother with Norem’s findings tend to compare the functioning of those who use defensive pessimism, a context-dependent coping strategy, to the functioning of dispositional optimists, instead of to those who use the (equally) context-dependent coping strategy of strategic optimism. When that latter, more appropriate comparison is made, performance differences typically evaporate (Norem, 2008, p. 124). And when the functioning of anxious persons who do not use defensive pessimism is compared to those who do, defensive pessimism enhances functioning significantly (Norem & Chang, 2002).
Norem (2014) reminds us that a key question about adaptive functioning is what negative thoughts and affects do, what kind of functioning they motivate, not simply how they feel. We should therefore consider functioning itself to be an important outcome category, distinct from affect. Though positive psychologists certainly now emphasize functionality (over happiness), the idea of doing so independently of affect does not fit their a priori mission. And so they cannot tumble to Norem’s (2014, p. 259) own acknowledgment of the “hedonic failure” of defensive pessimism while also asking us to consider how it improves adaptive functioning nonetheless. They apparently cannot conceive of how hedonics may not conduce to eudaimonics, which is just what Norem’s research demonstrates—over and over. And if positive psychologists counter that defensive pessimism does not produce optimal functioning, then the burden is on them to demonstrate that with their positive interventions they can improve the functioning of those high in performance-debilitating anxiety beyond (or even equal to) the improvements Norem has demonstrated with defensive pessimism. But in that case, they would have to take Norem’s findings seriously, instead of ignoring or dismissing them.
It is hardly surprising that Norem (2014) now wishes she had not used the term defensive pessimism, because “defensive” is too negative: If I could go back in time, I would change the term defensive pessimism to reflective pessimism, as opposed to nonreflective optimism. . . . By choosing defensive to label the pessimistic strategy we were studying, but strategic to label the optimistic counterpart, we inadvertently implied that there was something better (i.e., more “strategic”) about strategic optimism compared with defensive pessimism before we had gathered any data. (pp. 265-266)
And so we see that terminology matters. Norem is surely most prominent among those who study the “positive power of negativity.” Yet the failure of positive psychologists to integrate Norem’s consistent findings into their own research programs is perhaps the single most glaring example of their inability to conceive even of the possibility that negative characteristics can contribute substantially to positive functioning. That is perhaps because positive functioning for them logically entails enhancing, via “positive” interventions, the characteristics that they deem to be in “positive” territory a priori, and anything less remains in “negative” territory, even if that territory becomes considerably less “negative.” 20
To conclude, if “negative” interventions and states can be demonstrated to enhance positive outcomes in some circumstances, and if “positive” interventions and states can be demonstrated to hinder positive outcomes in some circumstances, then the a priori, context-independent, movement-defining dichotomy collapses. And with it goes the entire positive psychology enterprise, which is, after all, founded on that a priori dichotomy. The enterprise could of course survive productively if positive psychologists were to limit their place for positivity well short of what they now claim for its miraculous effects and find a substantially more constructive place for negativity. But could they then rightly call their movement positive psychology?
Why Can’t Positive Psychologists Accept the Positive Value of Negativity?
Not only do positive psychologists ignore or dismiss Norem’s findings about the “positive power of negativity,” they also fail to acknowledge the decades of scientific findings accumulated by contributors to Parrott (2014), who make a strong context-dependent case for the positive consequences of negativity and the negative consequences of positivity. I have theorized that positive psychologists’ a posteriori findings, not least Fredrickson’s 3:1 positivity ratio, seemingly justify or validate empirically, and so reinforce, their a priori designations of positive and negative characteristics, designations based on their hedonic criterion of pleasantness. And once so “validated,” they perhaps think it unnecessary to reconsider the alleged benefits of their positive characteristics and the alleged detriments of their negative characteristics, especially in relation to circumstantial particularity. If this is so, then the completeness and accuracy of their science are called into question. On a more ironic note, this also calls into question their open-mindedness, which is one of their positive characteristics. Indeed, open-mindedness is one of Seligman’s signature strengths; it is a route to the virtue of wisdom and learning (Peterson & Seligman, 2004). Seligman (2002) said that the “opposite of this strength is thinking in ways that favor and confirm what you already believe. This is a significant part of the healthy trait of not confusing your own wants and needs with the facts of the world” (p. 142). Another way to state the problem is that positive psychologists may have logically conflated their a posteriori findings and their a priori designations. If these epistemic conditions are not distinguished sufficiently, then the case is closed, leaving positive psychologists presuming themselves free to ignore the possibility that hedonics might not always conduce to eudaimonics, that negative feelings can contribute causally to flourishing (and positive feelings to languishing)—well beyond the limited nod to negativity given by positive psychologists now and then (Fredrickson & Losada, 2005, p. 685; Ng & Diener, 2009; Oishi, Diener, & Lucas, 2007; see Note 7).
In short, positive psychologists’ context-free causal relation between hedonics (feeling good) and eudaimonics (functioning well) problematically reinforces their a priori designations, perhaps leading them to see their causal relation between positivity and flourishing as constitutive of positivity, which then blinds them even to the possibility of nontrivial negative effects of positivity and nontrivial positive effects of negativity. As McNulty and Fincham (2012) aptly put it, “Presuming and thus predicting that a particular psychological characteristic will benefit people could lead researchers to accumulate knowledge consistent with that presumption—even if it is inaccurate” (p. 107). But here I merely speculate. And yet something still gnaws at some positive psychologists nonetheless. Evidence for this we find in their renewed call for integration of the positive and the negative.
The Call for Integration—and “Integration”
Soon after the movement’s birth, some members called for a positive place for negativity in positive psychology, for integration of the positive and negative, in the service of optimal functioning. In Authentic Happiness, Seligman (2002) said, “Positive psychology aims for the optimal balance between positive and negative thinking. . . . Positive psychology is a complement to negative psychology, not a substitute” (pp. 288-289). And in A Psychology of Human Strengths, coeditors Aspinwall and Staudinger (2003) elaborated Seligman’s aim for balance: A psychology of human strengths should not be the study of how negative experience may be avoided or ignored, but rather how positive and negative experience may be interrelated. . . . Some philosophical perspectives suggest that the positive and negative are by definition dependent on each other; that is, human existence seems to be constituted by basic dialectics. (p. 15)
In their chapter, Larsen, Hemenover, Norris, and Cacioppo (2003) wrote, This discomfiting mode of coactivation [of positive and negative emotions] may allow individuals to make sense of stressors, to gain mastery over future stressors, and to transcend traumatic experiences. . . . Although positive psychology has made it clear that an exclusive focus on negative emotions is insufficient, the present perspective implies that an exclusive focus on positive emotions may also ultimately prove insufficient. (pp. 213, 222)
And in another chapter, Ryff and Singer (2003) stated, Chastised for its preoccupation with human failings, the field of psychology has been admonished to attend to human strengths. . . . We [need to] move beyond false dichotomies that separate positive and negative features of the human condition [by appreciating] inevitable dialectics between positive and negative aspects of living. (pp. 271-272)
These early calls, then, raised logical, theoretical, and empirical concerns. Here we saw an open-minded, less dogmatic quest for a complete, holistic psychological science of well-being. This gave me hope at the time, although in light of the dominant, polarized, and polarizing rhetoric of the movement’s leaders, my hope was cautious at best (Held, 2004, p. 16). My caution turned out to be warranted: Positive psychologists have not only failed to heed these calls, they have run with haste in the opposite direction.
In the past several years a positive role for negativity has been demoted substantially: For many, negativity must be attended to in order to diminish it with “negative interventions,” which should somehow be integrated with “positive interventions.” The call for an “integration” defined in just this vague and attenuated way (relative to the aforementioned calls for a thoroughgoing integration) is seen most clearly in A. M. Wood and Tarrier’s (2010) proposed “positive clinical psychology” (also see A. M. Wood & Johnson, 2016). Recall that for them and others in the movement, positive interventions directly increase positive characteristics, and negative interventions directly decrease negative characteristics. There is no positive role for negativity; it must be diminished as much as possible in a “balanced” practice. As A. M. Wood and Tarrier (2010) put it, “We suggest the development of a Positive Clinical Psychology in which the understanding and treatment of clinical levels of distress is based on a balanced and equally weighted focus on the positive and negative aspects of life” (p. 820).
This suggestion opens a Pandora’s box of questions, beginning with what the “positive and negative aspects of life” are and how the positive and negative aspects of life may be properly “balanced” in the service of optimal functioning. After all, A. M. Wood and Tarrier’s (2010) “integrated” practice aims only to eliminate negative characteristics (Held, 2016). As Norem (2001, 2008, 2014), Parrott (2014), and McNulty and Fincham (2012) have consistently found, what is positive by enhancing functionality for some can be negative by diminishing functionality for others. Yet again, the assumption about which characteristics can be designated positive and which negative in virtue of contributing causally to flourishing and languishing, respectively, does not hold a priori or independent of circumstantial particularity.
Given positive psychology’s vanishing place for the positive power of negativity and its inattention to circumstantial particularity, it is noteworthy that Fredrickson finds a positive context-dependent role in flourishing for “appropriate negativity,” which is “time limited and soluble feedback connected to specific circumstances” (Fredrickson & Losada, 2005, p. 685). Although this statement about circumstances may be consistent with her 11.6:1 upper limit to the positivity ratio (above which languishing allegedly returns), it also presses the question of why she has not embraced Norem’s, Parrott’s, McNulty and Fincham’s, and many others’ circumstance-dependent findings in her “broaden-and-build theory.” Of similar interest is A. M. Wood and Tarrier’s (2010) atypical realization that “any designation of a characteristic as positive or negative is simplistic and inaccurate, as any trait or emotion can be ‘positive or negative’ depending on the situation and concomitant goals and motivations” (p. 827). Yet they, like Fredrickson, retain the a priori designation of positive and negative characteristics nonetheless; this despite their explicit nod to circumstantial particularity. Thus, Fredrickson severely limits a positive role for negativity, far beyond what the aforementioned findings indicate. And A. M. Wood and Tarrier give negativity no positive role at all; it must garner attention only for the sake of its diminishment if not elimination.
To be sure, (renewed) attempts at “integration” would not be needed if positive psychologists had not made the ontologically dubious move of carving psychological reality a priori into positive and negative components, which components then must be put back together again if they want a complete and comprehensive science of human flourishing. Yet they are stuck (seemingly happily) with their a priori positive and negative designations; they do not seem inclined to renounce or question them, even in light of the robust empirical evidence that cannot seem to penetrate their movement’s filtered vision. Alternatively, if positive psychologists have been aware of the findings that challenge their ideology and have deliberately chosen not to entertain them seriously, then they may not be quite as open-minded or as serious about advancing a rigorous, accurate, complete, and comprehensive science of flourishing as they may suppose.
In any event, it is time for positive psychologists to broaden and build their own science of flourishing, by revisiting their a priori designations with open minds and giving all due consideration to the many decades of research that have found positive effects of negative characteristics and negative effects of positive characteristics.
Conclusion
I have argued that positive psychology’s positive/negative dichotomy, in which positive and negative characteristics are designated a priori, independent of circumstantial particularity, is a form of false dichotomy. The terms do not refer to characteristics consistently when circumstantial particularity enters the picture empirically; this obtains for both the pleasantness/hedonic criterion of positivity and the causal contribution-to-flourishing/eudaimonic criterion of positivity. Moreover, the conflated, ambiguous, and/or inconsistent use of a priori designations and a posteriori findings, of constitutive and causal relations regarding positivity and negativity, may contribute to positive psychologists’ failure to notice and/or investigate the many robust findings about the positive consequences of negativity that have been studied for some three decades by mainstream psychologists. A progressive, flourishing science of any sort cannot be founded on a false dichotomy. A science so founded cannot flourish; it languishes eventually, even if that languishing is not apparent to those who participate in advancing that science.
The goal of positive psychology, to discover and enhance the determinants of flourishing, is laudable; it requires and deserves as complete, comprehensive, and accurate a science as is possible. To achieve their goal, positive psychologists can no longer afford to ignore or dismiss the many decades of research that demonstrates empirically the positive consequences of the characteristics they designate negative and the negative consequences of the characteristics they designate positive. Nor can they afford a conflated, ambiguous, and/or inconsistent use of constitutive and causal relations in their conceptualizations of positivity and negativity, which use infects their science.
To advance a complete and comprehensive psychological science in which flourishing gets top billing, it is just as important to study the adaptive enhancement of “negativity” as it is to study the adaptive enhancement of “positivity”—to let the scientific findings drive and determine what characteristics in what circumstances for what persons with what goals determine adaptive and maladaptive outcomes. If positive psychologists had done this, they would not be stuck with their false dichotomy and thus their a priori problem; indeed, they might not have become positive psychologists in the first place, since positivity psychology is founded on the a priori designation of positive and negative characteristics. This leads us to wonder whether their mission is first and foremost in the service of the complete and comprehensive science they have propounded or in the service of the hegemonic movement for which they have been criticized (Held, 2004; 2005; Katzko, 2002).
In light of the conceptual confusion positive psychology has bequeathed to our discipline, should we consider expunging the terms positive and negative from psychology’s lexicon (for other than mathematical purposes)? After all, the totality of relevant and robust empirical evidence—and not just positive psychology’s self-generated evidence—press those who attend to it to acknowledge that positives can be negatives and negatives can be positives. This is psychological doublespeak. To succeed in their mission, scientists must remain doggedly determined to gain knowledge of reality as accurately as possible. This mission depends foundationally on the conceptual clarity that has eluded positive psychologists, in their dogmatic determination to preserve their a priori ideology of positivity.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.