p.1
Part I
Evolution and the nature of compassion
p.3
Chapter 1
Compassion
Definitions and controversies
Paul Gilbert
Compassion is clearly linked to the evolution of caring behaviour (Gilbert, 1989/2016, 2009, 2015a, 2015b; Keltner, Kogan, Piff & Saturn, 2014; Mayseless, 2016), altruism (Preston, 2013; Goetz, Keltner & Simon-Thomas, 2010; Ricard, 2015) and prosocial behaviour in general (Bierhoff, 2005; Brown & Brown, 2015; Penner, Dovidio, Piliavin & Schroeder, 2005). The last 20 years have seen increasing research into the nature and benefits of all these. For compassion, these include a range of effects on psychological processes (Jazaieri et al., 2013; Keltner et al., 2014; Singer & Bolz, 2012), social relationships (Cozolino, 2006; Crocker & Canevello, 2012; Penner et al., 2005) and physiological processes (Klimecki, Leiberg, Ricard & Singer, 2014; Kogan et al., 2014; Simon-Thomas et al., 2011; Weng et al., 2013) that extend to genetic expression (Slavich & Cole, 2013; Slavic & Conway, Chapter 9, this volume). Hoge et al. (2013) found that women with experience of loving-kindness meditation had longer relative telomere length than controls (see also Fredrickson et al., 2013). Compassionate motives benefit social relationships and well-being, whereas ego self-focused motives do not (Crocker & Canevello, 2012). Compassion has become the focus for psychotherapeutic interventions with increasing evidence for its effectiveness (Gilbert, 2000, 2010, 2015c; Hofmann, Grossman & Hinton, 2011; Kirby & Gilbert, Chapter 15, this volume; Leaviss & Uttley, 2015; Neff & Germer, 2013). In addition, all the chapters of this book address these themes too. However, despite this progress there remains controversy and discussion around the actual nature of compassion, its definition and constituents. This chapter explores some of these controversies.
Issues of definition: what’s in a word?
We can only understand phenomena by agreeing a set of properties by which phenomena will be known. So, for example, we can distinguish between a chair, an elephant, a tiger and a cat by agreeing a set of qualities and properties that belong to each and those that are specific. Hence, they share qualities as four-legged things, but not of being living organisms. We can then distinguish between an elephant, a tiger and a cat and then finally between a tiger and a cat, and even different breeds of cat. Each subdivision has more and more overlapping features, but also more fine-grained distinctions. This is important in areas like compassion too which are easily fused with concepts such as altruism, benevolence, heartedness, prosociality, kindness and love. Sometimes it’s the fine-grained distinctions that are important. Second, as the late Prof Kendell (1975) used to drum into us clinical trainees in Edinburgh in the 1970s, there is no such ‘thing’ as depression, anxiety or paranoia as rarefied states. Rather there are a set of phenomena (signs and symptoms) that we agree may cluster/fit together and we will give labels to, so that we can agree what we are discussing when we use the label. So too, the definition of compassion will depend upon the shared properties we give to this concept and its differentiation from similar concepts. Third, definition problems can plague psychological research because of the insufficient attention it gives to functions. For example, shame and guilt, envy and jealousy, are frequently used interchangeably when they are in fact very different, with very different evolutionary histories, competencies, functions and focus (Gilbert, 1989, 1998, 2005, 2009). So, as will be discussed, although there are many general ideas around what compassion is, currently there is no clear agreement about what the specific attributes of compassion are. Let’s take a look at the range.
p.4
Defining compassion
One origin of the word compassion is from the Latin compati, meaning ‘to suffer with’. However, words change their meanings with use over time and across cultures and this is certainly the case for the word compassion. For example, to ‘suffer with’ need not imply a compassionate motivation to do anything about it. Today ‘to suffer with or on behalf of’ might be more closely related to what we now call sympathy (Eisenberg, VanSchyndel & Hofer, 2015) or empathy (Batson, 2009). Aristotle argued that we only had compassion if we thought the suffering of another was nontrivial; to use one of his examples, ‘we would not have compassion for the upset of an emperor who’d lost a shipment of Larks tongues he’d ordered for his party!’ (Nassbaum, 2003). Aristotle also thought that a sense of deserve influenced compassion, and third that we would have to have some sense of that suffering; that we could imagine ourselves in their situation (Nassbaum, 2003). So the seriousness of suffering, sense of deserve and empathy texture compassion for Aristotle. Buddhist scholars would not see these qualities as central.
In regard to dictionary definitions there are quite a variety. The Oxford Dictionary (2016) suggests that compassion is ‘Sympathetic pity and concern for the sufferings or misfortunes of others’. Teaching in France, Germany and Italy, colleagues inform me that ‘compassion’ is a difficult word to translate and can indeed be linked to pity. This is obviously a problem because pity is a very different process altogether and involves a sense of looking down on others (Nassbaum, 2003). This is a good example where misunderstanding about the use of a word can lead to heated debates and serious confusions. Schopenhauer (1788–1860), who was deeply influenced by Buddhist thought, and one of the first Western thinkers to introduce compassion into philosophy, argued that compassion was one of the highest and most important qualities of humanity. Neither self-interest, nor duty, social conformity nor fear of punishment could, in his view, be a source of genuine morality (Cartwright, 1988). However, Nietzsche wrote a number of significant critiques of Schopenhauer arguing that compassion is a poor source for moral development. The problem is, as Cartwright (1988) observed, these two philosophers were talking at cross purposes because of poor word definition:
p.5
(p. 1)
This confusion of concepts remains, especially in translations, and may help explain why, at times, compassion is regarded as unhelpful (Who wants to be pitied?), or even as a weakness. In contrast, the Cambridge University Dictionary (2016) suggests compassion is ‘a strong feeling of sympathy and sadness for the suffering or bad luck of others and a wish to help them’. Here, ‘pity’ is absent and is replaced with ‘sympathy’ and ‘feelings of sadness’, and the ‘wish to be helpful’ is also added. The concept of sympathy or ‘natural sympathy’, as David Hume called it, as underpinning human benevolence, is prevalent in many early philosophical texts. This definition also has the idea of a specific ‘feeling’ in compassion that stimulates motivated helping behaviour. In Goetz et al.’s (2010) major review of some of the evolutionary and historical origins of compassion they also place feeling centre stage:
(p. 351; italics added)
They also distinguish compassion from distress, sadness and love. As helpful and as detailed as this review is, there is a difference in defining compassion as a feeling state rather than a motivational state, and it is important not to confuse motives with emotions (Deckers, 2014; Gilbert, 2015a, Weiner, 1992). Indeed, Ekman (2016) surveyed researchers who study emotion. While 91% saw anger and 90% saw anxiety as basic emotions, compassion was rated one of the lowest with only 20% agreeing that compassion is an emotion. Without an evolved caring motivation system there would be no emotions arising to signals of suffering/distress. It is the motivation system that necessitates emotions in contexts, and those emotions may well then stimulate motivated actions (Gilbert, Chapter 3, this volume).
p.6
It is also unclear what the feeling state of compassion is. For example, what is the emotion when rushing into a burning house to save a child, consoling a recently bereaved or dying friend or fighting for human justice? It is likely that in the first case the feeling states/emotions are of urgency and anxiety, in the second sadness and sorrow, and in the third case a degree of anger at injustice. It’s not the emotions that unite them, but the motivation to pay attention (on suffering and needs) and the motivation and intention to do something about it. That intention can be present even when people are not thinking about it or are not activated. Note too that compassion can represent blends of emotions. For example, a doctor is struck by the suffering of the Ebola virus and dedicates his/her life, working long hours, to finding a cure. Emotions here may be a blend of anger that such viruses exist in the world, sadness for the suffering caused and the way people die from it, and anxiety that if it gets out into the world it could even reach his/her own family. In addition, whereas emotions wax and wane, a motivation can guide behaviour for the whole of one’s life.
The facial expressions of compassion can also differ according to context. The facial expressions we may have when consoling somebody in physical pain may be different to being with their grief or anger. The facial expression of a therapist meeting a patient for the first time and wishing to present themselves as compassionate, trustworthy and a safe listener will be different from compassionate expressions as different aspects of the story unfold. Whereas compassionate, kind and trustworthy faces are usually perceived as friendly, affiliative, happy and positive, such facial expressions may be less positively experienced when the viewer is in pain (Gerdes, Wieser, Alpers, Strack & Pauli, 2012; Godinho, Frot, Perchet, Magnin & Garcia-Larrea, 2008). When asked to pose compassion faces people automatically create soft expressions with gentle smiles of softness, friendliness and signals on being ‘safe’, kindness and gentleness (McEwan et al., 2014). Furthermore, it is now recognised that although compassion is seen as a positive act, it actually engages aversive emotions which arise when we are in touch with suffering (Condon & Barrett, 2013). So context is crucial.
The Free Dictionary definition (2016) suggests compassion is a ‘Deep awareness of the suffering of another accompanied by the wish to relieve it’. In this definition there is no pity, sympathy or ‘specific feeling’ aspect like sadness or sorrow, but a focus on ‘deep awareness’ with (again) a motivation (the wish) to do something about it. This definition gets closer to the traditional philosophical and contemplative traditions of compassion, as we will see.
The contemplative traditions and the multifaceted approach: Many approaches to compassion see it as multifaceted, although they do not always agree on what those facets are. Buddhist scholar Geshe Thupten Jinpa (translator to the Dalai Lama) and colleagues, who developed the Stanford compassion cultivation training, define compassion as:
p.7
(Jazaieri et al., 2013)
Mindfulness and compassion thinkers Christina Feldman and Willem Kuyken (2011) also highlight the multifaceted textures of compassion. They suggest that:
(p. 143)
They go on to add:
(p. 145)
Kuyken and his colleagues offer a more specific listing from their major, more recent literature review (Strauss et al., 2016).
(p. 19)
Dutton, Workman and Hardin (2014), who have done considerable work on compassion in organisations, relate compassion to four core aspects that also touch on cognitive, affective and behavioural processes: (1) noticing/attending to another’s suffering, (2) sensemaking or meaning making related to suffering; (3) feelings that resemble empathic concer...