Compassion as a Skill: A Comparison of Contemplative and Evolution-Based Approaches

A key part of skillful compassion is having clarity as to its functions and definition. On these issues, there is still some debate, with slightly different views on how to conceptualise and measure compassion (Condon & Makransky, 2020; Crocker et al., 2017; Gilbert, 1989, 2017, 2020a; Goetz et al., 2010; Mascaro et al., 2020; Neff, 2003; Strauss et al., 2016). Most current definitions have been derived from Buddhist concepts (Condon & Makransky, 2022; Ricard, 2015; Strauss et al., 2016). For example, Buddhist scholar Geshe Thupten Jinpa (2015; translator to the Dalai Lama) and colleagues, who helped develop the Stanford Compassion Cultivation Training, define compassion as:

a multidimensional process comprised of four key components: (1) an awareness of suffering (cognitive/empathic awareness), (2) sympathetic concern related to being emotionally moved by suffering (affective component), (3) a wish to see the relief of that suffering (intention), and (4) a responsiveness or readiness to help relieve that suffering (motivational). (Jazaieri et al., 2013)

Strauss et al. (2016) offered a systematic and integrative review of a number of different, mostly Buddhist, approaches and came out with their overlapping but also different set of competencies and skills. They concluded:

A range of definitions from Buddhist and Western psychological perspectives were considered and five components of compassion were identified: recognition of suffering; understanding its universality; feeling sympathy, empathy, or concern for those who are suffering (which we describe as emotional resonance); tolerating the distress associated with the witnessing of suffering; and motivation to act or acting to alleviate the suffering. Each of these components has been articulated by several published definitions of compassion, although no single existing definition explicitly includes all five of them. (p. 25; italics added)

In listing these qualities as ways of being compassionate, they also point to the skills that need to be developed. In a more descriptive but still Buddhist approach, mindfulness and compassion scholars Feldman and Kuyken (2011) highlight the multifaceted textures of compassion: They suggest that:

Compassion is the acknowledgment that not all pain can be ‘fixed’ or ‘solved’ but all suffering is made more approachable in a landscape of compassion.

Compassion is a multi-textured response to pain, sorrow and anguish. It includes kindness, empathy, generosity and acceptance. The strands of courage, tolerance, equanimity are equally woven into the cloth of compassion. Above all, compassion is the capacity to be open to the reality of suffering and to aspire to its healing. (p.143; italics added)

They go on to add:

Compassion is an orientation of mind that recognises pain and the universality of pain in human experience and the capacity to meet that pain with kindness, empathy, equanimity and patience. (p. 145)

As we will note shortly, if compassion is to take on the harmful side then two qualities are essential: courage and wisdom. One of the key distinctions requiring clarity therefore is what compassion actually is. For example, is it a motive, a feeling state or an attitude, and can we distinguish it from ways of being compassionate? Simmer-Brown (2022) relates how some Buddhist scholars think the West sees compassion as an emotion, and some definitions include an emotional or affective state component (Goetz et al., 2010). However, Simmer-Brown (2022) notes that the Tibetan language does not have a term that maps closely to concepts of emotion or affective state. Here we note that most approaches regard compassion as a “motive and an intention” and that emotions, thoughts, attitudes and behaviours are ways of being compassionate and guide compassion. In other words, just as depression is a mood disorder, with many different possible manifestations, so we can see compassion as a prosocial motive that has many different ways of being expressed. This enables us to distinguish the basis of compassion from ways of being compassionate.

We can start the explorations with motives and needs at the top of Fig. 1. Motives are the essential guiding processes of the mind that coordinate other functions. Baumeister (2016) says:

The primacy of motivation emphasizes that cognition, emotion, agency, and other psychological processes exist to serve motivation. (p. 1)

Psychological theorists should therefore devote considerable respect and attention to motivational processes. Broad theories that combine multiple processes should generally give priority to motivation as fundamental, because the full causal chain that leads to behavior and beyond will generally start with motivation. (p. 3)

Motives (including ones for compassion) evolved from the challenges of survival and reproduction to orientate and stimulate organisms to seek out what they need to do to sustain life and reproduce (Buss, 2019; Davey, 2019b). There are three major motives and life tasks:

These three “basic” motives give rise to many sub-motives, such as wanting to pursue a career, have a family and children, become famous or become a hermit and seek enlightenment, as well as small motives such as wanting to go to a party. There are different social motives called social mentalities, such as being caring, cooperating, competing and gaining control over resources, even at the expense of others (Gilbert, 1989, 2020b, c). Social mentalities differ from other basic motives because they require skills in the ability to understand social communication for role enactment. For example, an individual has to be able to identify any one signal from another as, say, a sexual signal or a threat signal, or a signal of caring or friendship and then respond appropriately. Social mentalities therefore require feature detectors for certain types of social signal, with processing systems to interpret those specifics signals that carry information about intent and emotional state. They also require the ability to send reciprocal dynamic information thus creating a particular social role relationship (Gilbert, 2017; Gilbert, 2022a, c). Through reciprocal, dynamic interactions, individuals create roles for caring (compassion), help seeking, cooperating, sexuality, competing and so forth.

All motives are supported by if A then do B (stimulus response) algorithms, where A is dependent on evolved feature detectors that trigger B, action responses (Davey, 2019b; Deckers, 2014). Whether it is protecting from threats or seeking food or sexual encounters, understanding the algorithms of a motive is essential to exploring its biopsychosocial manifestations. Harm avoidance motives need feature detectors for threat that trigger physiological systems (such as the amygdala) to deal with threat such as fighting, running away or hiding. In contrast, feeding motives need feature detectors to identify edible foods and how to find them, and prepare the body for consumption and digestion. Sexual motives need feature detectors for sexual opportunities that stimulate bodily processes for courting and mating behaviour. Motives could not evolve without feature detectors, and they would be relatively useless without being linked to output and response systems. The evolution of feature detectors means that the brain can monitor for the presence or absence of certain stimuli and signals. It follows therefore that care and compassion evolved with feature detectors for detecting signals of distress and need, which then trigger physiological, cognitive and behavioural responses to address them. The stimulus detection of suffering, in contrast to the activation of responses to be helpful, is underpinned by different physiological systems and search routines (Di Bello et al., 2021). Individuals who are sensitive to distress (to self or others) but do not know what to do, or feel constrained in acting, can experience high personal distress.

We may wish to get fit and know how to get fit, but still not do what we need to do to get fit. Poulin (2017) noted that we can be quite capable of detecting suffering and know what to do, and even appear to be motivated to do it, but then do not actually perform the actions necessary. Motivation to take action, the wisdom of what action to take and actually taking action are therefore different processes and skills (Di Bello et al., 2020). In addition, people can take actions to address suffering and needs in others but not necessarily from a compassionate motivation. Instead, it is to be liked or to harness a good reputation (Böckler et al., 2016; Catarino et al., 2014). In fact, most of our behaviours have multiple and different motivations behind them rather than one, and many are unconscious to us anyway (Bargh, 2017). Hence, it is possible to behave compassionately both from a genuine wish to be compassionate and also to be liked or rewarded.

The evolution of caring and compassion motives have different roots with different functions and therefore compassion cannot be regarded as a single skill or process. We can roughly identify three interacting evolved pathways into compassion.

Over hundreds of millions of years, various caring motives have been evolving around these basic algorithms and processes of [A] evolved feature detectors (enabling attention sensitivity) for stimuli and signals of suffering/need that then [B] trigger appropriate actions to alleviate and prevent suffering and address needs. Following these steps enables us to see how we arrive at a definition of compassion as a basic algorithm which is:

[A] sensitivity to suffering in self and others with [B] a commitment to try to alleviate and prevent it. (Gilbert, 2014, p. 19)

This definition is informed by evolution and Buddhist thinking, and with slight variations is fairly typical of most definitions (Mascaro et al., 2020). However, it is recognised that just because individuals have the potential to be sensitive and take action does not necessarily mean they will (Poulin, 2017). In fact, even if people have compassion skills they may, for various reasons, be fearful or resistant to using them (Gilbert et al., 2011; Gilbert & Mascaro, 2017; Kirby et al., 2019). Thus, as noted above the competencies, skills and processes to be sensitive, moved by and understand suffering need to be distinguished from the competencies, skills and processes that guide and enable taking action.

Emotions are different to motives and are in the service of motives (Baumeister, 2016). In his book on the evolution of emotions, Good Reasons for Bad Feelings, Nesse (2019) says:

Emotions are specialized states that adjust physiology, cognition, subjective experience, facial expressions, and behaviour in ways that increase the ability to meet the adaptive challenges of situations that have reoccurred over the evolutionary history of a species. (p. 53)

Emotions come and go, and their basic function is to change physiology and the orientation of attention, thinking and behaviour to meet the challenge of pursuing motives. Hence, we can identify particular types of emotion associated with the three basic life tasks and life motives. Emotions that are associated with harm avoidance and threat include those such as anxiety, anger and disgust. Emotions associated with seeking and achieving resources tend to be what we call positive emotions such as joy, excitement and pleasure. Emotions associated with rest and digest, and with sources of satisfaction, are associated with a sense of peacefulness, safeness and contentment.

There have been many models which point out that how we understand our emotions significantly impacts our behaviour towards ourselves and others. For example, Irons (2019) highlights stages associated with emotion such as awareness of changes in the body, awareness of triggers, ability to label and recognise emotion including that we can have more than one emotion at the same time, and ability to understand and make choices about how to behave in any particular emotional state as opposed to being controlled by it. Some individuals will avoid certain emotions both on purpose and also non-consciously, which can make compassion difficult if they are engaging with suffering that stirs up painful emotions. Key skills of compassion will therefore be attention sensitivity allied with emotional tolerance and empathic insight in the face of suffering which support wise action.

Competencies are a third class of evolved functions. These are processes that facilitate action and emotion regulation. For example, birds need wings to fly and a brain that will navigate their flight. The motives for and reasons they are flying could be various, such as escaping a predator, looking for food or returning to the nest, but the brain systems and competencies they are using to fly will be exactly the same, regardless of reason. One of the important distinctions between humans and other animals is the degree of complexity of our cognitive competencies (Gilbert, 2022a). These include competencies for:

Together, these competencies offer the basis for knowingly and intentionally, that is paying attention and knowingly developing intentions that enable a choice of action. Unfortunately, they also come with built-in (non-conscious) biases and at times deep irrationalities because the brain is not evolved to be rational as such and uses many (jump to conclusions) shortcuts (Gilbert 1998; Nesse, 2019). These built-in biases are partly why we can be so harmful to ourselves and others and why we need to learn to become mindful of them. In addition, individuals can be skilled in one competency but not necessarily another. For example, individuals may be extremely intelligent and win Nobel Prizes, but may be poor at empathy and vice versa. Or they may practise a form of mindfulness but are still not that empathic or sensitive to others. Indeed, developing the skills for the different competencies supporting compassion may require specific training such as empathy training, mindfulness training or the use of imagery and metacognition (Favre et al., 2021; Singer & Engert, 2019). Crucial, too, is the motivation behind the use of competencies because any of these competencies can be used to be helpful or harmful. Empathy, for example, can be used to be deceptive or to ingratiate oneself with others in order to exploit them. One of the reasons we have designed such horrific tortures is because we can use empathy to maximise pain and terror.

Behaviours depend on competencies to engage in actions, and through actions we learn and change. In other words, skills grow from the doing. In Fig. 1, UCR stands for unconditioned response. These are the responses that are built into us and are relatively automatic. For example, we suddenly see a snake or somebody jumps out at us and we automatically go into anxiety. CR stands for conditioned response. These are the behavioural responses we learn through experience. To some extent compassion requires us to be sensitive and mindful to our UCRs and CRs especially if they link to impulsive or unhelpful behaviours.

Many behaviours are not simple responses to stimuli but emerge out of motives and goals shaped by various complex cogntive competencies. For example, we want to drive or play the piano but these require dedicated practice behaviours integrated with a knowledge of cars and roads, and for the piano, music. To become skillful we need both. No matter how much one studies and intellectually knows about driving or playing the piano, unless one actually drives or plays, one will not get better at it. Intellectual wisdom cannot be translated into experiential wisdom. The same applies for compassion, self-transcendent states and insight. People may be highly motivated to engage in insight or compassion practices or actions but do not follow through with actual actions (Poulin, 2017). Translating motives and “know-how” into behaviour change is one of the challenges for many change-based interventions. Knowing, intending and wanting to and actually doing are different. At times it is doing things with others that give us that push into action. This is what makes the support of the “sangha” or community so important for facilitating behaviour change. Skillful compassion arises then from the integration of multiple processes.

The competencies for insight and knowing intentionality are crucial for compassion because compassion is more than just the caring “instinct”. Lions are clearly motivated to hunt and kill but not knowingly. They cannot decide if they need to train to run faster, lose weight or choose to become a vegetarian because hunting causes suffering. Humans can help or harm, with aware intention, with knowing, self-aware insight into the nature, causes and experiences of suffering. Partly because we have the three types of competencies outlined above, we can work out what is likely to be causing suffering and distress, and what is likely to be helpful or not. For example, with empathy, “I know that you are suffering today because it is the anniversary of the death of your beloved partner”. Another empathic recognition for action might be “taking you out to meet friends for support or just being with you today”. It is when we knowingly and intentionally orientate ourselves to be helpful to self and others that we enact compassion. Without this, compassion may become synonymous with more automatic caring. This is depicted in Fig. 2.

Hence, compassion can be viewed as the motive that stimulates and guides us to knowingly notice suffering in self and other sentient beings, knowingly and empathically connect to make sense and understand the nature and causes of the suffering, and knowingly use empathy and our basic knowledge to search for and implement possible solutions (Gilbert & Choden, 2013). These competencies may also enable us to become aware of, and then override, our inhibitors to compassion such as not being compassionate to strangers (Kirby et al., 2022; Loewenstein & Small, 2007)

This bears closely on the concept of dukkha, which is important in Jainism, Hinduism and Buddhism. There are different types of dukkha arising from the fact that we, as self-aware conscious beings, have conscious experiences of suffering that are (1) inherent in pain, disease and death; (2) inherent in change and the arising of loss of the wanted (i.e. grief) and the emergence of the unwanted; and (3) due to being aware of the suffering inherent in existence itself, including in other sentient beings. Indeed, these forms of dukkha can only arise because of our consciousness awareness. Clearly, artificial intelligence cannot experience dukkha. It follows, then, that the way we use our new brain competencies impacts how we experience and cope with fear, pain, loss and misfortune.

Linked to our evolved cognitive competencies is the way we form beliefs that bring different motives, cognitive competencies and sources of information together. Our beliefs are not just about things and people in the world, and our expectancies, but also about ourselves and the kinds of minds we have and the people we are and want to become. Beliefs arise from, and also direct, our cognitive competencies, frame how we see and think about the world, the minds within it, and choose what values to pursue. Beliefs can guide us to what we want to know, what we turn away from and hence what is knowable. If we believe that Buddhism is (say) a dysfunctional religion then it is unlikely we will want to learn about it and discover its wisdoms and mind trainings. People can develop beliefs that support prosocial but also very harmful behaviours (Eidelson & Eidelson, 2003). Beliefs can fuel fear and avoidance of compassion; that compassion is a weakness or is undeserved (Gilbert et al., 2011; Kirby et al., 2019). Indeed, beliefs can guide us to compassion or intense cruelty, as well as a sense of hope or despair (Beck, 2002). Buddhism does not describe concepts like beliefs as such but certainly has many; for example, the four noble truths represent basic beliefs while the eightfold path represents beliefs about the nature of ethical behaviour (Huxter, 2016). Buddhism generates many beliefs about how to liberate ourselves from the illusions of a separate self and the benefits of doing so, but recognises that belief and faith are not enough and these have to be experienced for wisdom to really change the mind.

Buddhist psychology has key beliefs about the causes of suffering as being linked to types of attachment that generate ignorance, greed and grasping. The obvious antidote is to guide people to develop non-attachment and non-grasping on the basis that all things are impermanent (Austin, 2009). In other words, the process of non-attachment is an essential compassion practise and skill because of how it undermines the three causes of suffering. Western psychotherapy approaches this issue of over attachment or clinging-grasping in slightly different ways. For example, Ellis (1979; see Dryden et al., 2010), the founder of rational emotion behaviour therapy, highlighted that people get into trouble when they turn a preference into a must a have to and awfulise. He called it “mustabating the self” and highlighted that humans make unreasonable “have to” demands on themselves, life and the world that are impossible to meet for reasons linked to not accepting limitation, and imagining catastrophic outcomes. Mustabating and demanding can be linked to beliefs such as “unless I succeed I’m nothing”; “I must not upset people or make mistakes”; “I have to be loved or admired otherwise I will be alone”; and “it is awful and unbearable to be alone or rejected and therefore I must at all costs stop it from happening”. In fact, individuals can become highly competitive and non-compassionate because they have a fear of inferiority, being left behind and marginalised, or are distrusting of others to help them (Gilbert et al., 2007; Basran et al., 2019). Crucially, there is clearly a connection between motivation to avoid harms and these beliefs (Gilbert, 1992). This indicates that the more threatened we feel, the more such beliefs can dominate our emotions and behaviours. Therefore, to address attachment and striving, we have to address underlying fears which are mostly to do with fear of rejection and marginalisation.

This approach does not stress the importance of non-attachment, which can be easily misunderstood in the West as non-engagement, but rather the importance of having rational-helpful desires and preferences (not demands) supported by rational-helpful beliefs, with the skills to cope with the ups and downs if the preferences do not work out. Like a potential surfer who wants to ride the wave, first they must learn how to swim and be OK with falling off. When there is fear of falling off and drowning then they will grasp the surfboard and fail to learn how to ride or not try at all.

There have been a number of fascinating comparisons between Buddhist approaches and rational emotive behaviour therapy (Holt & Austad, 2013; Kwee & Ellis, 1998) and other cognitive approaches (Tirch et al., 2016). In their different ways, they both address the way our recently evolved cognitive competencies, such as the way we reason, the way we empathise or don’t, and our ability to be mindful and discerning of what is helpful versus unhelpful, play key roles in our mental states. Western therapies might explore unconscious fears or unprocessed trauma or look at the beliefs that underpin “grasping and greed” or certain attachments, because they are not always what they appear on the surface. In Western psychotherapy, they may be revealed through processes of therapist-guided discovery, Socratic inquiry, reflection and behavioural experiments.

Rather than working through unconscious processes or trauma, Buddhism places emphasis on mind training such as with meditation and at times specific behavioural practises and experiments. A classic story is of a woman experiencing profound grief over the death of her child. Asking the Buddha how she could cope, he advised her to find mustard seeds from a house that had not been touched by grief. Searching high and low for such a house, she encountered many stories of grief and in that way understood the nature of common humanity and the reality of suffering. Monks must rely on the charity of others to provide their food and in this way understand the importance of interdependence and gratitude. Interestingly, insofar as one of the primary fears of humans is social disconnection and aloneness which can give rise to “grasping”, if people have an experience of self-transcendence associated with interconnectedness and love, then that could settle that fear system in a way quite different to Western psychotherapy.

Being motivated and having cognitive competencies and beliefs that guide and inspire compassion require other characteristics for compassion to be utilised. Central ones are courage and wisdom, because compassion intention is not enough to ensure appropriate compassionate actions (Gilbert, 2009; Jinpa, 2015; Ricard, 2015). If a person jumps into a river to save someone but cannot swim, or enters a burning house without proper safety equipment, they are going to cause more problems than solve. Courage without wisdom can be reckless. Equally, if a person knows how to swim or firefight but is too anxious to act on that knowledge, then their knowledge and wisdom are ineffective. So both are necessary. In a fascinating interpretation of the way various historical Buddha images began to emerge, Vessantara (1993) indicates that they became associated with these different qualities and skills of mind:

as Buddhists over the centuries meditated on the archetypal Buddha, more and more forms appeared. They sprang up profusely like wildflowers in different shapes and colours […] All these forms were expressions of the life and vitality of the Dharma […]

To begin with, two new Buddhas appeared. These expressed the two great aspects of Enlightenment – Wisdom and Compassion. As people dwelt on these qualities as embodied in, Skakysmuni, so they took on a life of their own and became two new archetypal Buddhas. (p. 62)

Compassionate courage and wisdom became sources of meditation in their own right, with different bodhisattva images representing courage and wisdom. One meditated on them, imagining them flowing into oneself, and becoming them to try and stimulate these qualities within oneself (Gilbert & Choden., 2013; Leighton, 2003; Vessantara, 1993). In the secular approach to compassion, people practise different skills of courage and wisdom relating to the different types and contexts of suffering. Firefighters who risk their lives or clinicians who risk COVID-19 infection need physical courage but also specific skills and wisdoms for intervention. Standing up against injustice and risking ridicule is a different type of courage needing different skills. Working in a hospice with end-of-life care again requires different kinds of courage and skill. And being a good parent requires special skills, too. Clearly, one might be good in one context but not another. A courageous firefighter might not make the most empathic friend or parent.

Courage and wisdom are also at the heart of people’s journeys through psychotherapy because they often have to engage with states of mind they find frightening and painful (Rachman, 1990). It is not uncommon for therapists to work with veterans who have shown extraordinary physical bravery, risking death and physical injury, but who struggle when it comes to their own emotional pain. As one veteran said, “I have the courage to die but not to cry”. Clearly, then there are different forms of courage: one linked to physical risk and one linked to the ability to tolerate emotional pain. Both Western psychotherapy and the contemplative approaches agree that mind training can be about fostering courage so that we do not use our mental training to get rid of emotional pain as a form of avoidance, but rather to be able to accept, tolerate and work through the pain.

Zimbardo (2015) highlighted what he calls heroic compassion, where the “archetypal hero” knowingly engages risks to achieve some goal. Hence, we could define courage as the knowing engagement in risk of harm or loss to self in order to pursue an intention. Motives are crucial in respect of whether courage and wisdom are used to be harmful or helpful. A skilled burglar may show courage in taking a risk to climb through high windows, or a drug trafficker in risking detection. Forms of compassionate courage may involve altruistic sacrifice when we give up some of our resources and privileges to help those less fortunate than ourselves, for example, donating a kidney to save someone’s life (Kirby et al., 2022). Other forms, including assertiveness, risk taking, forgiveness, kindness and tenderness, are all ways of being compassionate rather than compassion itself. Indeed kindness has been shown to be different to compassion in relationship to the emotions that are generated (when engaging in compassion emotions are more threat focused) and the courage and sacrifices involved (Gilbert  et al., 2019).

Seeing compassion as an algorithm, with two different components of engagement and response, supported by the competencies for (1) reasoning and conceptualising, (2) mentalizing, and (3) observing and mindfulness that facilitate knowing intentionality, enables the distinguishing of different processes underpinning compassion (Gilbert, 2009, 2014, 2022b). For example, we can further differentiate specific competencies that can be targets for therapeutic intervention. CFT has suggested six competencies or processes underpinning our ability to engage with suffering and distress. These include (1) being motivated to address suffering and thereby (2) attune our feature detectors to be sensitive to distress; (3) being moved by distress as in our personal sympathy reactions; (4) being able to tolerate that distress from turning towards suffering; (5) being empathically able to gain insight into the nature and causes of distress; and (6) non-judgemental, condemning or harshly critical. 

When it comes to the action components, we have to (1) switch our attention away from the nature suffering into what is likely to be helpful; (2) run imaginary scenarios in our minds that orientate us to helpfulness; (3) use our reasoning and empathy skills for working out what is likely to be helpful; (4) set out to act on our wisdoms, (5) when needed, to ground ourselves in our body and that supports helpful action; and (6) tolerate the feelings and emotions that arise when we engage in action. For example, sometimes helping people in some situations can involve threat to the self and anxiety.

The twelve competencies are only guides that support compassion cultivation. They are not meant to be exhaustive lists. Rather they are identifiable trainable skills. In other words, we can develop our own and our client’s skills for paying attention, helpful reasoning, empathy, emotion tolerance, mindfulness and so forth. These competencies form the bedrock for the core compassion qualities of the wisdom of how to understand and act, the courage of how to understand and act, and the commitment to be sensitive to suffering in self and others and take action (Gilbert, 2022a). Importantly, all these competencies can also be expressed as “flow” in the sense that there is the sensitivity, empathy and helpfulness we give to others, that we can feel coming from others, and we can direct to ourselves. Condon and Makransky (2020) highlight the importance of the moral and ethical principles of compassion, particularly the focus on compassion for others for compassion to be sustainable. In regard to training and therapy, some competencies may require more efforts than others but training in these competencies and skills in the context of a secure and safe therapeutic relationship is at the heart of compassion focused therapy (Gilbert, 2010; Gilbert & Simos, 2022) (Fig. 3).

In an effort to explore and measure competencies for engagement and competencies for action, Gilbert et al. (2017) developed the compassion engagement and action scales (CEAS). We developed three complementary forms for the flows of compassion that measure: self-compassion, compassion to others, and compassion from others. These were translated into a number of different languages and have now been used in a number of studies (Arı et al., 2022; Asano et al., 2020; Duarte et al., 2021; Matos et al., 2021, 2022). In terms of the effects of CFT on harmful behaviour, there is growing evidence that it is helpful for those who have harmed others and are being treated in forensic contexts (Ribeiro da Silva et al., 2019, 2021).

In summary, an evolutionary and biopsychosocial approach to compassion, and its benefits, views compassion as a basic motive (Mascaro et al., 2020). Motives are different processes with different functions from emotions, competencies and behavioural dispositions. Motives orchestrate them in different ways according to context (Gilbert, 2020a, b; Gilbert & Simos, 2022). We can track the origins of care-focused motives and their evolved psychophysiological processes over hundreds of millions of years (Cassidy & Shaver, 2016; Mayseless, 2016; Seppälä et al., 2017) and how caring became “compassion” when we evolved the competencies for different types of “knowing” that made us able to (1) reason and understand the relationships between cause and effect and can have insight into how things work in the world (Byrne, 2016; Stewart-Williams, 2018); (2) consciously and knowingly empathically resonate with suffering in sentient beings (Luyten et al., 2020); and (3) develop a form of mind awareness from evolving self-awareness and “conscious of being conscious”. These different skills and competencies of compassion have become a focus for mind training in this approach. The contemplative traditions that train the brain and the mind overlap with these factors but also have a different focus and source of insight for cultivating compassion.

Although, compared with those of Western psychology, the different schools of Buddhism have different ways of mapping our inner psychology, the issue of mind training is central to both. A crucial focus of Buddhism’s mind training is mindfulness (Brown et al., 2015; Germer & Siegel, 2012; Gilbert & Choden, 2013) with a specialist journal Mindfulness dedicated to its study. Mindfulness has two crucial aims. One is to become an observer of one’s mind so that one becomes familiar with the arising of different, thoughts, emotions and motives, keeping track of these in relationship to intention particularly compassion intention. Then, rather than unhelpfully ruminating on them or acting them out impulsively, to be able to choose how to act helpfully not harmfully. The second skill of mindfulness is to enable the mind to settle, quieten the default mode, create the conditions for more subtle awareness of the nature of consciousness and eventually move into self-transcendent and non-duality states (Kitson et al., 2020; Ornstein & Ornstein, 2021; Siegel, 2023; Yaden et al., 2017). Mindfulness also supports compassion in that it facilitates maintaining one’s behaviour on the eightfold path (Huxter, 2016). As noted above, empathy is very important for skillful acting with compassion. The eightfold path has within it the notion of developing empathic awareness because it requires us to be sensitive to the impact we can have on other people so as not to cause them to suffer. Hence, we can have empathy for what has happened to others, but also empathy for our impact on others. These are regulated through different processes.

Whereas Western psychotherapy seeks to help people understand and utilise basic biopsychosocial processes to shape compassion and helpful mental states, the contemplative traditions have a very different focus. This relates to enabling people to experience new and enlightened states of consciousness that dissolves illusions of a separate self and generate profound changes in the experience of consciousness itself. Indeed, one key route to the inspiration to pursue compassion came from some (but by no means all) mystical self-transcendent experiences (Ornstein & Ornstein, 2021; Yaden et al., 2017). As humans evolved self-awareness and “consciousness of being conscious” minds, they could knowingly explore questions about the nature and meaning of their short existence in ways that no other animal could (Dunbar, 2022). From the earliest known societies, humans used substances and rituals to try to create experiences that transcended their everyday ones and that gave insight into alternative realities (Dunbar, 2022; Muraresku, 2020). Luke (2019) and Muraresku (2020) outline the fascinating history of the use of psychedelics to deliberately create altered states. Luke (2019) reviewed some of the typical paranormal experiences people have and found that other practices employed to produce such states include complex breathing patterns, fasting, intense exercise and spinning. Common to such altered states is losing a sense of a separate, individual self to “being one with all”, such as one with nature, with the universe, or with a representation of God. In addition, these self-transcendent states were often associated with feelings of overwhelming love, awe and joy that left individuals with a genuine increase in compassion for all beings, partly through that sense of complete connectedness (Ornstein & Ornstein, 2021; Siegel, 2023; Yaden et al., 2017). More recently, psychedelic-induced experiences have been used to help people with depression, anxiety, trauma and the fear of death, with promising results (Carhart-Harris et al., 2018; Nayak & Johnson, 2021; Schimmel et al., 2022; Stellar et al., 2017; Tagliazucchi et al., 2022). Again, it appears to be the sense of loving, safe interconnectedness, “a merging of the self” and self-transcendence, which are crucial (Stellar et al., 2017; Schimmel et al., 2022).

Another mystical experience that is attracting increasing scientific attention is near-death experiences (NDE) and the potential for “consciousness” to exist separately from biology, typically referred to as an out-of-body experience (OBE). While some NDEs can be frightening (Cassol et al., 2019) Greyson (2010) and Woollacott and Shumway-Cook (2020) note that most are associated with a sense of self-transcendence, cosmic unity and overwhelming joy. NDE accounts also typically describe a form of unconditional love that is not like human love (see below). Crucial to this paper, NDEs not only reduce the fear of death but also increase a sense of compassion and connectedness (Cavarra et al., 2022; Greyson, 2010, 2021). One of the central controversies is the degree to which these experiences are any more than complex neurophysiological changes and (as science fiction writers show) our ability to create amazing fantasies (Dunbar, 2022), or do they offer windows on more complex understanding of the nature of consciousness (Greyson, 2021; Woollacott & Shumway-Cook, 2020)?

An increased sense of unconditional compassion is also consistent with a pre-planned form of NDE invoked through meditation, known as the meditation-induced NDE (MI-NDE), that has been studied in advanced Buddhist meditators (Van Gordon et al., 2018). As part of the key features of the MI-NDE, such as an OBE, altered perception of time and space, volitionally manifesting a non-corporeal form, and helping other beings was identified as both a rationale for and an outcome of the practice. All these experiences typically became referred to as self-transcendent (Coxhead, 1985; D’Aquili & Newberg, 1993; Ornstein & Ornstein, 2021;Wilber, 2007; Yaden et al., 2017), although this term is now used in different ways (Kitson et al., 2020). Neuroscientists are beginning to identify particular brain areas that appear associated with particular experiences, such as self-transcendence and timelessness (Ornstein & Ornstein, 2021).

Related to self-transcendent experiences, as generated in the ways discussed above, are the dimensions of non-dual consciousness awareness that emerge out of mindfulness and other mind trainings. One of the main aims of Buddhist mindfulness and compassionate mind training is to facilitate people’s experience of transcendental states, and experience “the non-duality of consciousness” as the ground of all being (Austin, 2009; Spira, 2008). Non-duality means that everything is a property of one “process”, such as energy (Siegel, 2016, 2023). In this approach, consciousness is not created by the brain or electrical chemical processes but is a property or dimension of all things (Austin, 2009; Ornstein & Ornstein, 2021). One of the functions of mindful “present moment awareness” is to facilitate receptivity to non-duality experience (Austin, 2009; Spira, 2008). Whether or not consciousness is only an emergent property of the brain or not, meditative techniques are known to use a range of practices that have neurophysiological effects. These are seen as able to facilitate the experience of the non-duality of consciousness and of the interconnectedness of all things (Austin, 2009; Ornstein & Ornstein, 2021; Siegel, 2023). This experience is different from Western psychotherapy because it dissolves the distinctions between self and other, object and subject, time and space.

Non-duality of consciousness is at the heart of Buddhist approaches to insight and compassion (Anālayo, 2015; Van Gordon et al., 2017). However, when we think about insight we have to ask: insight into what? Van Gordon et al. (2021) suggest that there have been three stages by which Western science has become interested in Buddhism insight and compassion. The first stage was interest in the relationship of mindfulness and attention to mind states. This was of special interest to those helping people with stress, painful states and self-regulation (Germer & Siegel, 2012; Kabat-Zinn et al., 1985). The second was concerned with ethical and moral principles and the eightfold path. The third stage is becoming interested in the dynamic of consciousness itself, that is, the nature of emptiness (Sanskrit: śūnyatā), the non-duality of consciousness as it underpins, and is the source for, the arising and falling of all things (Austin, 2009; Siegel, 2016, 2023; Van Gordon et al., 2017).

The editors invited consideration of potential dangers from efforts at Buddhist and secular integration, such as the common danger due to the translation of words conveying concepts across languages and cultures. For instance, in some languages, the word “compassion” does not actually exist, or can be seen as pity, which is a completely different notion. Philosophers Arthur Schopenhauer and Friedrich Nietzsche had heated debates about the value of compassion to underpin morality because Nietzsche defined compassion as pity, whereas Schopenhauer took the Buddhist view (Cartwright, 1988).

The use of the word “love” is another problem area. The experience of enlightened states, particularly with their sense of complete interconnectedness, self-transcendence, joyfulness and “infinite oneness”, is sometimes described as one of overwhelming “love” (Yaden et al., 2017). However, this is very different to our evolved feelings of love linked to liking, desire and dopamine, of “I love you and you love me”, or “I love pizzas”. In addition, the Buddhist concept of love is “a wish” or “desire for”, that is “the wish for all beings to be free of suffering and the causes of suffering”. In English, this is more closely related to the words “altruism” and “benevolence” rather than the word “love”. A doctor will try to operate as best s/he can whether s/he likes the patient or not. The medical motivation is not based on “love” but on the desire to alleviate and prevent suffering. That motive can stand separate to the motive to love and to want to be loved or feel affection and warmth. The definition of love is therefore crucial. Indeed, Anālayo (2015) and others have noted that, although now installed in Western approaches, the term loving-kindness can be problematic for the same reason. He argues it is better to focus on the concept of benevolent wish, where benevolence is the wish for one (and others) to be free of suffering, of the causes of suffering, and to experience the causes of happiness (Dalai Lama, 1995). As a modern take, compassion therefore emerges as insightful empathic benevolence, which seeks to help us with the suffering we are all caught up in. This suffering is partly the result of being a DNA-built, without our consent, short-lived, wanting and fearing, vulnerable-to-sickness, conscious and self-aware, socially textured being. We exist as separate beings in the boundaries of our skin. However from the contemplative point of view — we suffer this too because we are ignorant of our true interconnected, non-biological, non-dual consciousness nature. It is unclear whether some secular approaches to compassion that have now proliferated on the internet and promoting “love” understand the different meaning and experience of love from self-transcendence. Instead, some may be engaging what the Dalai Lama (2001) calls sentimentality. The problem with this is that individuals then start to try to create feelings of love which can get in the way of benevolent self-transcendence and psychotherapy (Gilbert & Simos, 2022). Research also shows that kindness tends to drop away for people we do not like, whereas this is less the case for compassion and addressing suffering (Kirby et al., 2022). Hence, benevolence and altruism are better terms to focus what underpins compassion (Anālayo, 2015; Ricard, 2015).

As noted, the concept of non-attachment is also tricky because it does not actually mean detachment or disengagement. It is a much more subtle relationship between self and world. In Western psychology, attachment relates to the way we link our sense of ourselves to what we own or what we achieve in the world, and the way we seek to form a self-identity which we will then defend, reasons why Leary (2004) called this the curse of the self and Hood (2012) the self-illusion. Also as noted above, attachment refers to the way we fixate on these things and turn them into “absolute musts and have to’s”. The dialectical conflict between identifying with a sense of self and not comes from developmental studies of how a sense of self emerges through childhood. Without a sense of self, of what hurts and what brings pleasure, what brings acceptance and what brings shame and rejection, what values to aspire to and live by, biological life would become almost impossible. There is therefore a balance to be struck. Thus, non-attachment does not mean that one is not fully engaged in relationships; that one cannot fully love and grieve with the loss of love. It seems easy for westerners to use the concept of nonattachment as another vehicle to emotional avoidance — seeking a simple protection against the pain of loss or failure. Indeed, one would argue that it is the courage, wisdom and emotional maturity to be able to tolerate (potential) grief that allows one to love fully.

Another area where non-attachment can be confusing is in the nature of self-transcendence. In a sense, if one has experienced states of non-duality, where one phenomenon embodies all phenomena, then the concept of attachment to any particular thing seems irrelevant. Indeed, it seems to be the experience of being part of, and one with, all things that corresponds to self-transcendence and the loosening of ego-based attachments that loosens self-attachment identity (Siegel, 2023; Yaden et al., 2017). In other words, once individuals experience that their existence is implied in all things (and vice versa), the striving of a self-identity becomes less important and one seeks the joys and freedom of non-duality.

Non-duality transcends the notion of being “inter-connected”, which seems to imply “duality” on account of one thing being connected to another thing(s). Hence, Siegel (2023) refers to being “intraconnected”. Beyond being a functional label, it is impossible to assign an absolute separation between one phenomenon and the next. For example, it is incredibly difficult to locate the exact separation line where a valley connects to a mountain, where a wave begins, and arguably it is more accurate to say that the two phenomena exist inseparably and flow into one another. Thus, notions such as interconnectedness can constitute useful didactic means for beginning to think about non-duality and self-transcendence, but can also incur limitations.

As we have noted, the concept of emptiness is fundamental in Buddhist thinking but again easy to misunderstand (Anālayo, 2015). In English, emptiness can also mean “nothingless” or “without anything”. The Buddhist meaning however is empty of individual existence. Everything is pattern generation. For example, brains are patterns of electrochemical activity that is constantly changing. The atoms of our brains may have been part of a mountain or dinosaur. Every 7 years our bodies have completely regenerated the atoms they are using to make us. So today the fascination is to try to explore how energy gives rise to matter gives rise to a form of “self”-awareness which is also at the heart of self-transcendence. Self-transcendence is not annihilation but “expansion into” or a “dissolving of boundaries” (Austin, 2009; Ornstein & Ornstein, 2021; Siegel, 2016, 2023). The question remains therefore if self-transcendent and non-dual states, and the urges to compassion that arise from them, are simply the products of neurocircuits or are of a brain that has an expanded awareness of the domains of consciousness (energy fields) beyond biological, yet may also depend on the biological. For example, in order to experience colour, we need eyes and a brain, but the source of the experience lies in light frequencies falling on the eyes. Furthermore, human vision is sensitive to only a very narrow bandwidth of energy, hence a narrow range of experienced colours; eyes cannot create colour experience out of nothing. While neuroscientists debate the circuits that seem important for consciousness, physicists are debating the implications of quantum mechanics and consciousness. How these sciences, including those from the contemplative traditions, unfold and reveal new insights will have major implications for how we see ourselves as a species and want to relate to each other to create a more compassionate world.

As noted throughout this paper, western psychotherapy has as its focus helping people organise their minds and brains to address mental health problems and promote well-being. Creating the conditions for non-dual experience and insight is not (usually) its focus. However, experienced meditators and psychotherapists have explored these variations in therapeutic goals and how one can inform the other (Gill et al., 2015). What is agreed is a need for better precision in terms of what different practises are designed to do, in what context and for whom. This is especially important for compassion research and when we are trying to develop practises not only to help individuals develop compassion but also for helping groups and communities.