By Norah Hannel, Connecticut College
This paper was presented at Prometheus’ 2014 Mid-Atlantic Philosophy Conference.
By offering two counterarguments to Jesse Prinz’s explanation of valence, I will ultimately defend the view that valence depends on an emotion’s pleasantness and unpleasantness. In his book Gut Reactions: A Perceptual Theory of Emotion (2004), Prinz gives an overview of five possible valence theories that he refutes to then propose his own alternative. Before delineating them, however, I will proffer a clearer understanding of what valence consists in. To embark with a definition in mind, “the difference between positive and negative emotions is called a difference in ‘valence,’”[i] across which each emotion can be divided. In other words, the positivity we associate with some emotions would be deemed “positive valence” and the negativity we associate with others, “negative valence.” Joy and hope fall to the former, and sadness and anger to the latter side of valence; Victor S. Johnston described this scalar polarization as falling on a “hedonic tone”[ii] continuum. Valence is “essential to emotionality,” [iii] for without it, we would have no basis for reacting positively or negatively to a stimulus. Regardless of their independent characteristics, joy, hope, and reassurance all have positive valence, while anger, sadness, and fear all have negative valence.
Some emotions are entirely positive, others entirely negative, others yet are mixed. Joy, we might say, is entirely positive and anger, entirely negative. Prinz mentions nostalgia and feeling touched as examples of mixed feelings. By “feeling touched,” he means the state of being surprised to joyful tears at the kindness of others, often after having endured some sort of tribulation.[iv] I also imagine homesickness when traveling to an exciting but unfamiliar place, relief after a drawn-out hardship, and nervousness on a first date to all be mixed emotions, since each of the three contains a positive as well as a negative element. Now that we have a clearer sense of valence as it takes form in our daily lives, let’s turn to the six theories thereof, as conveyed by Prinz.
The most minimalistic valence theory is that emotions are positive when they are perceived as pleasant and negative when perceived as unpleasant (Frijda, 1993). David Hume endorses this theory of emotions when he speaks of indirect passions as bifurcated into pleasant and painful impressions, which then indicate, together with the object of our emotions, what the passion is: love, pride, humility, or hatred.[v] Though this approach to valence intuitively makes sense to me, Prinz claims that it does not account for unconscious emotions since pleasantness and unpleasantness are necessarily conscious feelings. He gives the example of states that share properties with fear, including their negative valence, but are unconscious. After traversing the remaining theories, I will ultimately espouse the pleasantness-unpleasantness view as phenomenologically paramount.
The second theory of valence, called the “dimensional appraisal theory,”[vi] links positive emotions to the congruence of our goals and negative emotions to the incongruence thereof. The closer we come to fulfilling our goals, the more positively valent is the accompanying emotion; the further away we stray from fulfilling our goals, the more negatively valent it becomes. The problem here is no longer one of consciousness, since unconscious sentiments can align with or deviate from our goals. A possible example of such an unconscious deviation might be the case of “Anna O.” Sigmund Freud writes about Dr. Breuer’s patient, Bertha Peppenheim, alias Anna O., who suffered from a severe cough, unilateral paralysis, damaged hearing, vision, and speech abilities, contractures, and “hysteria.”[vii] While experiencing these symptoms, she was caring for her ill father, who soon died of tuberculosis. The grief surrounding her father’s death was masked by physiological pathologies. According to the dimensional appraisal theory, the death of a close relative possibly went against her goal of maintaining a family structure in her life so was incongruent to that goal, causing a negatively valent emotion.
However, while the dimensional appraisal theory does appear to account for conscious as well as unconscious feelings, as hypothesized in the case of Anna O., a new problem arises. This theory regrettably neglects to recognize that not all of our emotions have to do directly with our goals. If we assumed that each moment in which we are abruptly surprised or frightened were anchored in personal goals, we would endorse an excessively cognitive account of the emotions. For example, someone jumping out from behind a corner and scaring us is not contingent upon the goal of us not being scared. The dimensional appraisal theory of valence is therefore also unfavorable.
A third alternative focuses on “approach [and withdrawal] tendencies.”[viii] Positive emotions prompt us to move towards the object of our emotion, whereas negative emotions prompt us to back away. Prinz’s challenge to this theory is that approach and withdrawal are not always linked to positive and negative emotions, respectively. Some emotions inspire inverse behaviors. Prinz gives the example of anger, which, on this model, should prompt withdrawal since it is a negative emotion. Instead, it often impels us to yell or attack—approach behaviors. Conversely, we may feel joy when listening to a certain song, yet do not move closer to the speaker from which the music plays; we tend, instead, to sit back and bask in its pleasant effect. Since emotions do not uniformly translate into predictable actions, they should not be broken down into approach and withdrawal tendencies.
The fourth theory draws from Silvan Tomkins (1962), who suggests that positive valence leads to an outward focus of attention, whereas negative valence leads to an inward focus of attention. As I understand it, joy, for instance, causes us to engage with the people and world around us, inviting us to laugh, frolic, and cheer. We feel pleasantly connected to and in tune with the spatiotemporal context we find ourselves in; our attention is outwardly focused and our emotions therefore, positively valent. Sadness, on the other hand, drives the direction of our thoughts inward, isolated from the greater context we are embedded in. The behavior, in turn, becomes more individualistic, pushing us to brood, fret, and nostalgize, for example. We sever ourselves from the world around us in favor of blue-tinted introspection. However, since there are positive emotions, such as pride, that turn our focus inward, and negative emotions, such as contempt, that turn it outward, directionality of attention, is not necessarily correlated with valence of emotions.
Having exposed dilemmas in all four preceding valence theories, Prinz turns to a fifth, “animal learning theory,”[ix] for which he calls upon behaviorist Jeffery Gray (1982, 1987, 1991). The premise of animal learning theory is that “properly selected stimuli can be used to regulate behavioral responses.”[x] The stimuli that increase the probability of an active response are called positive reinforcers and those that decrease such probability are called negative reinforces. Put simply, positive reinforcers give a reward or take away a punishment; negative reinforcers give a punishment or take away a reward. Gray explains that the brain contains a “behavioral approach system” (BAS) and a “behavioral inhibition system” (BIS). He has two leading hypotheses, the second of which is particularly relevant to our discussion of valence. [xi] High BAS is related to positive valence, he claims, and high BIS to negative valence.
In presenting his own theory of valence, however, Prinz criticizes this hypothesis by saying that it echoes the approach and withdrawal tendencies model. BAS, which precipitates behavior that is anticipated to lead to a positive emotion, resembles approach, and BIS, which terminates behavior that is anticipated to lead to a negative emotion, resembles withdrawal. Anger, despite its negative valence, might increase the probability of a response by prompting the angry person to punch a wall or accost a perceived adversary, for example. Prinz maintains that behavior and emotion should be neither sufficiently nor necessarily linked in a phenomenologically astute model of valence, as we discovered in looking at the approach and withdrawal tendencies theory above. Emotions can be flexible and unpredictable, so the behavior that results “depends on what we identify as the source of past emotions…[and] on what we identify as the best coping strategies”[xii] for the future.
Prinz, however, does not entirely dismiss the BAS-BIS system that Gray introduces. He thinks it prepares us to respond physiologically to a rewarding or punishing stimulus, thereby contributing to the embodied appraisal that constitutes our emotions. He suggests that emotions might, during the process of initiation, have different levels of BAS and BIS activation. These systems, then, could prove helpful in distinguishing between the unique profiles of emotions.
As to the significance of the BAS and BIS system in valence, Prinz claims that valence markers are synonymous with reinforcers. Again, positive reinforcers in the BAS model present a reward or take away a punishment, whereas negative reinforcers in the BIS model present a punishment or take away a reward. A positively valent state has an inner positive reinforcer (IPR) and a negatively valent state has an inner negative reinforcer (INR).[xiii] These inner reinforcers tell us whether we want more of something in the case of positive emotions, or whether we want less of something in the case of negative emotions. The reinforcers allow for more flexibility in associating behavioral responses to positive and negative stimuli, as they honor our individuality in experiencing the stimulus. We then decide, based on the personal inner reinforcers, whether we want to indeed approach or withdraw from the stimulus.
Inner reinforcers notify our bodies of how we should respond, in the form of an embodied appraisal, to the stimulus. This holds true not just at the precise moment in which we face the stimulus but also further down the road. IPRs and INRs, associated with positive and negative emotional responses respectively, become stored in our memory so as to prepare us for similar circumstances in the future. “Emotions serve as memory markers, which factor into decision making,”[xiv] meaning that the IPRs and INRs we experience in the present can impact our future behavior.
Prinz does anticipate a possible counterclaim to his hypothesis of inner reinforcers. If IPRs cause us to increase behavior that leads to positive emotions and INRs cause us to decrease behavior that leads to negative emotions, then we might not be able to account for cases in which behavior and valence have an inverse relation. This counterclaim seems to presume that valence is normatively intersubjective. In other words, the vast majority of a community should agree upon which emotions are positively valent and which are negatively valent. The people who defy these social norms are ostracized because they are invoking an inapt application of valence. Valence, according to this view, seems to depend on how ethically apt an emotion is in its given circumstance. IPRs should correspond to positive emotions and INRs should correspond to negative emotions as judged by society at large. A masochist who associates pleasure with self-destruction or a Stoic who associates displeasure with joviality go against the societal grain regarding classifications of valence. By counting on valence to be intersubjective, this counterclaim of inner reinforcers threatens the realism of human variability in the interplay between behavior and valence, as may be the case for the masochist and the Stoic.
While this counterclaim, as a prudent move in his argument, does not consume a large portion of his investigation, Prinz offers three succinct responses. First, he calls upon Freud, who points out that “compulsion to repeat patterns can outweigh hedonistic pursuits.”[xv] The increase or decrease of a certain behavior must not correlate with the increase or decrease of the degree to which it is hedonistic. Second, through learning and experience, typically positive or negative emotions can reverse their expected responses. For example, fear of heights can be seen as positive in creating the pleasant rush of being on a rollercoaster, converting the negative valence of fear into a positive one of enthusiasm. And third, “opponent processing”[xvi] holds that we have psychological mechanisms in place that want to promote an emotional equilibrium and prevent an access of either positive or negative emotions. At one point or another, we are all likely to be overwhelmed by tears that eventually turn into laughter in between sobs because we have come to a point of relief. This inclination towards harmony of emotions might explain “[behavior that seems] to be at odds with the reinforcement properties of our emotions.”[xvii]
After having resolved the anticipated counterclaim, Prinz further elucidates his own theory of valence, for which inner reinforcers inform embodied appraisals. In explaining emotions, he perspicaciously weds the appraisal approach with the James-Langer approach of emotions, seeing merit in both the cognitivist and the physiological views. When we experience emotions, we are thinking about life but are using our whole bodies to think. In other words, our bodies participate in cognitively appraising a situation. So, embodied appraisals are the “internal states that register patterned bodily changes.”[xviii]
What about valence? While “appraisals represent things that matter to us…they do not represent the fact that they matter. That’s where valence markers come in.”[xix] An emotion’s embodied appraisal reflects the positivity or negativity of an event—for example, laughing at a birthday party or crying at a funeral—but the emotion’s “valence marker”[xx] indicates what about the event makes us perceive it as positive or negative in the first place, for example the presence (in the case of the birthday) or absence (in the case of the funeral) of loved ones in our lives. Embodied appraisals and valence markers are two separate but complementary elements of an emotion.
Prinz’s account is compelling but too quick, I believe, to dismiss the pleasantness/unpleasantness model, or, as I will refer to it from here on out, the “phenomenological theory.” The first of my responses to Prinz’s counterarguments against this theory looks at the status of unconscious emotions in the context of valence, and the second questions the feasibility of inner reinforcers. For Prinz, the phenomenological theory is insufficient because it does not account for unconscious emotions. I take Prinz to imagine the fallible structure of the phenomenological view to proceed as follows:
(1) The terms “pleasant” and “unpleasant” necessarily describe conscious emotions
(2) An emotion that is unconscious cannot be considered pleasant or unpleasant
(3) An unconscious emotion, thereby, cannot have valence
(4) Without valence, an emotion cannot be an emotion
(5) Conclusion: An unconscious emotion is not an emotion
His concern for this conceptualization of emotions is that it excludes the possibility of unconsciousness. I believe, however, that we should not be troubled by emotions that lack consciousness since they arguably do not even have valence until they become felt.
If the emotion were felt, it would be felt as pleasant/unpleasant. The unborn state can have hypothetical valence when it is unconscious, but it needs to be actualized by consciousness to have full-fledged valence. An unconscious fear, for example, can have negative valence in a hypothetical way. Valence is itself a factor of how the emotion is subjectively felt, which I take to mean how pleasant or unpleasant it is. Therefore, dismissing the phenomenological theory because it shelves unconscious emotions, which appear to fall short of valence anyway, does not offer an adequate counterclaim.
For the sake of clarification, pleasantness and unpleasantness might also be described as pleasure and pain. The latter dichotomy contains a further distinction to which Martha Nussbaum directs our attention: there are both nonintentional and intentional pains. Nonintentional pains, such as bodily aches—can in fact be unconscious but are not essential to our emotional judgments. Intentional pains, on the other hand, are integral in acknowledging that, for instance, “an important part of one’s life is gone”[xxi] when a loved one passes away. As Nussbaum nicely puts it, “Not just an arm or a leg, but a sense of life, gets the shock of grief.”[xxii] Thus, the valence of emotions, not to be confounded with physiological sensations, is always conceived within our consciousness.
Prinz’s reliance on unconscious analogues to emotional states is tenuous. He says, “Even those who think all emotions must be conscious agree that there can be unconscious analogues of emotional states. There can be unconscious states that share many properties with fear, for example, but lack the distinctive fear feeling (Ledoux, 1996).”[xxiii] He continues, these “unconscious analogues of fear are negative states. Therefore, negative valence markers cannot be defined by unpleasantness.”[xxiv] However, “analogues of emotional states” are not emotional states so should be kept from composing the crux of his argument against a theory of, precisely, emotional states.
Nostalgia, going back to Prinz’s example of a mixed feeling, is not an analogue to an emotion but a de facto emotion, only the latter of which, I assert, requires valence. “Unconscious emotions” are only analogous to emotions until they become actualized and from that point on carry valence. Therefore, I think the urgency Prinz places on accounting for unconscious emotions is irrelevant. The conclusion in the above-stated list of what I conjecture to be Prinz’s concern for the phenomenological theory are true: unconscious emotions do not have valence and they are not emotions. But these arguments do not threaten any theory of valence, which is necessarily conscious. Therefore, Prinz’s counterclaim to the phenomenological theory does not effectively undermine it.
My second rebuttal focuses on Prinz’s replacement of the phenomenological theory with a system of inner reinforcers. To recapitulate, Prinz describes metaphorical voices in us saying “more of this!” for positively valent emotions and “less of this!” for negatively valent emotions in reacting to a stimulus.[xxv] Note that these positive and negative reinforcers are not equivalent to pleasantness and unpleasantness. In the case of sadness, for example, an inner negative reinforcer would ask us to expose ourselves to less of the sad-making stimulus, indicating that it is a negative emotion. However, how would his theory pan out in the case of a mixed feeling, such as nostalgia? Since the constitution of nostalgia combines positive and negative valence, both IPRs and INRs would have to be firing at the same time without canceling each other out. If one side’s commanding “more” were to counteract the other side’s commanding “less,” there would ultimately be no valence, which is empirically untrue.
Instead of lacking valence, nostalgia seems to have both positive and negative valence, for which the phenomenological theory can account while the IPR-INR theory cannot. The latter approach would have trouble splitting its attention between concomitant positive and negative feelings. Furthermore, Prinz’s “more of this” and “less of this” schema begs the question, “…of what?” What do the IPRs and INRs in nostalgia want more and less of as they fire simultaneously? The phenomenological theory is more apt in describing nostalgia’s bittersweetness, a feeling of concurrent pleasantness and unpleasantness.
To continue looking at nostalgia as a paradigmatic mixed emotion, the bittersweet longing seems to be an overall positive reminiscence that carries a mild bite as we lament the loss of a feeling we relished in the past. On Prinz’s theory, we would have to say that the IPR is demanding “more of this” to a greater degree than the INR is demanding “less of this,” though we cannot be sure what the demand is referring to exactly. And since the demands, as polar opposites of each other, fire simultaneously, the concern that they might cancel each other out arises. The phenomenological theory, however, allows us to intuitively say that there is a greater intensity of pleasantness than there is unpleasantness, though the latter does still contribute to the overall emotion. Now we know how to resolve the “of what” as well as the “canceling each other out” dilemmas. Moreover, Prinz’s inner reinforcers can be traced to pleasantness and unpleasantness: positive reinforcers give us a pleasant feeling and negative reinforcers an unpleasant one. Therefore, his criticism of the phenomenological approach does not hold water.
By challenging Prinz’s critique that the pleasantness/unpleasantness approach neglects unconscious emotions, as well as casting doubt on the “more of this/less of this” structure of IPRs and INRs, I hope to have dissolved some of the obstacles that ostensibly stood in the way of accepting the phenomenological theory. I propose we think of positive valence as essentially pleasant and negative valence as unpleasant. In anticipating criticism of the phenomenological theory as reductionist, I believe the burden of proof to fall upon those who claim that every other theory of valence does not ultimately boil down to feelings of pleasantness and unpleasantness.
[i] Prinz, Jesse J. Gut Reactions: A Perceptual Theory of Emotion. New York: Oxford University Press, 2004. Page 160
[ii] http://www.theoryofmind.org/pieces/HTT.html (accessed on October 24, 2013)
[iii] ibid. Page 164
[iv] ibid. Page 165
[v] Hume, David, L. A. Selby-Bigge, and P. H. Nidditch. A Treatise of Human Nature. Oxford: Clarendon, 1978.
[vi] ibid. Page 168
[vii] http://www.richardwebster.net/freudandcharcot.html (accessed on November 11, 2013)
[viii] Prinz, Jesse J. Gut Reactions: A Perceptual Theory of Emotion. New York: Oxford University Press, 2004. Page 168
[ix] ibid. Page 169
[xi] For the purpose of this paper, I will leave aside the discussion of neurological mechanisms distinguishing BIS and BAS.
[xii] ibid. Page 172
[xiii] ibid. Page 173
[xiv] ibid. Page 174
[xv] ibid. Page 175
[xix] ibid. Page 178
[xxi] ibid. Page 64
[xxii] Nussbaum, Martha Craven. Upheavals of Thought: The Intelligence of Emotions. Cambridge: Cambridge University Press, 2001. Page 78
[xxiii] Prinz, Jesse J. Gut Reactions: A Perceptual Theory of Emotion. New York: Oxford University Press, 2004. Page 168
[xxv] Prinz, Jesse J. Gut Reactions: A Perceptual Theory of Emotion. New York: Oxford University Press, 2004. Page 174