Wide Externalism and the Roles of Biology and Culture in Human Emotional Development

Introduction

Within both philosophy and psychology emotions are considered to comprise, involve or essentially to be, thoughts or propositional attitudes, for example, judgements, cognitive appraisals, desires (Lazarus, 1991; Solomon, 1977, 2003); modulation of mental processes, for example, attention, memory (Oatley, 2004); subjective feelings of pleasure/displeasure (Ledwig, 2006); subjective feelings of bodily changes, that is, facial expression, orientation, musculature, visceral organs (Ekman, 1994; Griffiths, 1997; James, 1884; Prinz, 2004); action tendencies (Frijda, 1986); interactive orientation (Zinck & Newen, 2008); and paradigm scenario acquisition (De Sousa, 1987). Although all these theorists agree that the above components typify emotional experience, they disagree, as the list implies, on which particular component is the emotion or is essential to emotional experience. Theories range from the purely somatic-feeling (e.g., James, 1884) to purely cognitivist (e.g., Nash, 1989) with a range of less radical views intervening, for example, somatic (Prinz, 2004) and hybrids of various composites (e.g., Zinck & Newen, 2008). Relatedly, they also disagree on the extent to which emotions are genetically hard-wired and/or culturally/socially constructed. Theorists who opine the former tend to do so because there is at least some evidence of universal emotion elicitors and expressions and because there is increasing evidence of specific neural substrata for at least some emotions, for example, fear (Bush, Schafe, & LeDoux, 2009). Social constructionists, in contrast, deny the existence of universal emotions and, therefore, genetic hard-wiring. Emotions, to hard-nosed social constructionists (e.g., Harré, 1986), are socio-culturally determined.

I address the latter of these two issues in this article. I argue that recent theoretical developments in the metaphysics of mind offer new and constructive perspectives on the geneticist-biologist versus social constructionist debate. Indeed, incorporating insights derived from developmental psychology and biology, I demonstrate that they reconcile these two apparently conflicting positions by showing how they continuously and intimately inter-relate in human emotional ontogenesis and experience. Emotions are expressive-communicative devices which evolved to regulate and control mammalian, including human mammalian, social life, first through interpersonal mechanisms and subsequently through intrapersonal mechanisms. I argue that the environmental inheritance ¹ of human neonates, when functioning normally, provides the necessary and sufficient conditions for the development of a repertoire of full human emotions from a limited range of genetically inherited, culturally universal emotional precursors. This environment crucially includes a complexly interdependent, linguistically-mediated social relationship without which human neonates and children will fail to emote like human beings at all. They will fail to emote like human beings because they will fail to nudge into development/maturation/connection the necessary neural substrata of the emotions, including the affect programmes. They will also fail to become intrapersonal regulators of their own emotions. The close, linguistically-mediated social relationship of neonate and caregiver(s) provides the necessary sensory-perceptual stimulation to excite genetic activity and the resultant development/maturation of emotion-related neural substrata, including the amygdala.

Accordingly, this article, which is explicitly interdisciplinary, focuses on analyses of emotional ontogenesis, experience and metaphysical realisation. I argue that human emotional ontogenesis and adult emotional experience are both metaphysically realised (determined) in realiser systems which extend beyond the physical boundaries of the neonate/infant and individual adult into the sociocultural environments into which they are deeply embedded. I also argue that although the components in both realiser systems are similar, the crucial realiser component (termed the core realiser) in emotional ontogenesis and adult emotional experience differ. The close, linguistically-mediated, social relationship is, as indicated above, the core realiser of emotional ontogenesis. As the emotion-related neural substrata are developed and/or mature, however, the role of the amygdala becomes progressively more important (and emotion regulation progressively more intrapersonal) until, eventually, it becomes the core realiser of full adult (intrapersonally regulated) emotionality.

In order to contextualise these analyses, however, I begin with some general preliminaries with respect to emotion.

Emotions: Some General Preliminaries

Given the burgeoning scientific evidence (see above), and despite the views of the most hard-nosed social constructionists, my view is that emotions can usefully be construed as falling into two basic groups. These are affect programmes and the higher sentiments (Griffiths, 1997). Affect programmes evolve from reflexive affects, what I refer to below as precursor emotions, a group of eruptive and transient responses which are closely tied to precipitating environmental conditions, for example, fright, gustatory disgust, pain and the various homeostatic distresses (e.g., hunger, thirst) and endogenous pleasures. Affect programmes, that is, those typically referred to by the English words fear, anger, sadness, joy, affection and interest can markedly outlast precipitating environmental (and, in humans, internal precipitating) conditions. They are governed by a set of neural circuits in intermediate areas of the brain (Panksepp, 1998, 2000) and conceptualised as sensory-motor command circuits. They orchestrate the complex and coordinated behavioural, physiological, cognitive and affective responses characteristic of fear, anger, etc. In terms of complexity, they involve several expressive elements which include changes in facial musculature, changes in musculo-skeletal functioning, for example, flinching, orienting, expressive vocal changes, changes in autonomic nervous system (ANS) activity, changes in subjective feelings and a range of cognitive phenomena, for example, attention direction. In terms of coordination, various elements co-occur in recognisable patterns or sequences and, in terms of automation, they unfold without need for conscious attention (Ekman, 1994). In addition, their underlying neural circuits appear to be present in homologous fashion in the intact adults of all mammalian species, human and non-human (Griffiths, 1997).

In contrast, the higher sentiments are the emotional-affective processes that emerge from the recent evolutionary expansion of the forebrain (Panksepp, 1998, 2000). These processes are responsible for human beings’ subtle, social emotions, that is, shame, guilt, contempt, envy, humour, empathy, sympathy and jealousy. Unlike the affect programmes, higher sentiments do not display stereotypical patterns of physiological effects. They are more integrated with cognitive activity than affect programme responses, leading to planned, long-term actions (Griffiths, 1997). The higher sentiments rely on symbolic representations which enable higher cognitive functioning, for example, inferencing, reasoning and planning. Symbolic representations enable language, a social life governed by convention and rich cultural traditions. The higher sentiments presuppose language; they can be experienced only by creatures with the ability to semantically conceptualise, for example, shame presupposes the conceptual grasp of self, responsibility for transgression, etc. In addition, the cultural traditions which structure human social life account for differences in their fine-grained natures and the relative plasticity of their expression. Which higher sentiments an agent experiences, therefore, are consensually considered a function of her language and culture. The ontogenesis of the emotions, however, demonstrates clearly that language and culture are necessarily involved in the development and expression of human affect programmes.

The Socio-Cultural Context of Development: Preadaptations and Interactions

“Organisms inherit environments along with their genes … parents create very specific conditions for infants and the adult is the product, not of its genes, but of the interaction of its genes in this structured developmental context” (Griffiths 1997, p. 61). This is particularly true in relation to mental capacities. There is ample evidence accumulated over the last 40 years that both human and non-human primates will fail to develop normal adult mental capacities and certain normal adult behaviours if denied the social interaction to which they are preadapted (Bowlby, 1969; Harlow & Harlow, 1962). The development of any trait or characteristic is a consequence of coactions between complex molecular interactions within and across cells and the nature of the physical, biological and social environments through which individuals pass during development (Lickliter, 2008).

It is now recognised that one of the clearest links between gene activation in individual neurones and experience is that of immediate early genes (IEGs) and sensory stimulation (Johnston, 2008; Johnston & Edwards, 2002). Auditory, visual, tactile and olfactory stimulation almost always results in the activation, within minutes, of one or more IEGs. These genes produce transcription factors to control the production of proteins by other genes and these proteins construct and reshape neural circuits. The genetic activity is not restricted to early development; it is, rather, decreasing but ever-present during the life of human and non-human animals. This explains why animals, including human animals, who are raised in total or relative social isolation fail to develop normal cognitive and emotional capacities; without the requisite auditory, visual, tactile and olfactory stimulation, there will be no activation of IEGs and no construction and reshaping of neural circuits.

Some preadaptations with which human neonates are equipped are general adaptations; these include sensory-motor preferences and mimicry, and these general preadaptations are exploited during emotional development. Others are emotion-specific, for example, emotion precursors and both general and emotion-specific preadaptations combined with emotion-specific caregiver preadaptations in social interaction to realise full emotional development. The point is that emotions evolved to control and regulate the behaviour of members of social species (Soussignan & Schaal, 2005), and it is a necessary condition of emotional ontogenesis that social animals are exposed to the social lives their emotions evolved and developed to regulate.

The environments into which human neonates are born vary enormously across the globe; it is entirely reasonable, therefore, that they should have to learn which objects and events in them should elicit emotional responses. It is unremarkable that, although emotion-eliciting stimuli are learned, their learning is preadapted by the provision of precursor emotions which evolved to deal with experiential saliencies in ancestral environments. Such saliencies clearly included stimuli which were dangerous, noxious, challenging, novel and pleasurable. These match the precursor emotions with which human neonates are equipped, that is, fright (danger), disgust (noxious stimuli), distress (challenges), interest (novelty) and endogenous pleasure (pleasant sensations). These precursor emotions serve to benefit the neonate directly in terms of its relationship to the environment, for example, disgust results in opening of the mouth and retching, which removes noxious substances from the digestive tract. In human emotional ontogenesis, however, they also benefit the neonate indirectly by serving as signs of physiological/psychological status to caregivers, for example, noticing signs of disgust during feeding and, as a result, interrupting feeding. This response presupposes the making of the correct interpretation of the sign and the provision of the appropriate intervention in light of this interpretation. It is enabled by the complementary preadaptation of caregivers which takes the form of intuitive parenting skills.

Human neonates are the most neurologically immature of mammalian neonates (Precht, 1993) and this is particularly true in relation to their motor abilities (Thelen, 1984). In light of this, and to ensure survival and well-being, certain competencies evolved via a process of natural selection in both neonates and caregivers. In neonates, these take the form of a limited range of precursor emotions, precursor strategies for regulating emotions and sensorimotor competencies and, in their caregivers, they take the form of “intuitive parenting” skills (Papousek & Papousek, 1987).

Precursor emotions, that is, the referents of the English words distress, disgust, fright, interest and endogenous pleasure (Izard, 1978), which are consistently observed in neonates from birth, are triggered solely by absolute stimulus thresholds; they are not directed towards particular objects. They are expressive reactions to sensed positive or negative physiological change and presuppose a further innate endowment, that is, interoceptive and proprioceptive receptors in viscera, muscle and skin. The function of distress, disgust and fright is to signal needs-related deficits (e.g., hunger) or impairments to physical integrity to caregivers (Sroufe, 1996). The function of interest and endogenous pleasure, however, is to assist in building representations of the external and internal environment. In terms of precursor strategies for regulating emotion, neonates are equipped with only two and these operate only within a limited arousal range. Neonates engage in sucking to calm themselves and look away from overwhelming stimuli (Blass & Ciaramitaro, 1994) to control arousal. When arousal exceeds its stimulus range the infant shows signs of (i.e., expresses) distress. Distress is typically elicited by a threshold-exceeding deficit state and is expressed initially through motor unrest. This is followed by unfocused crying which increases in volume if distress continues or increases. Neither the motor unrest nor the crying indicates which threshold-exceeding deficit state is causing the distress (Lester, 1984). Caregivers have to discover for themselves what ails their infant and what to do to remediate it.

What these distress expressions do, depending on their relative sensitivity, is elicit in caregivers a measureable degree of psychophysiological arousal, that is, measurable physiological stress symptoms (Boukydis & Burgess, 1982), feelings of intense pressure to give assistance until the crying ceases (Gustavson & Harrison, 1990) and to seek the cause of distress, remove it and engage in intuitive calming actions (Papousek, 1990). Sensitivity means that caregivers perceive their infants’ precursor emotions, interpret them correctly and react to them appropriately; it is, therefore, the basis of successful intuitive parenting. Sensitive caregivers perceive unspecific and subtle cues from their infants and respond promptly, regardless of their own current motives. Sensitive parenting results, at 3, 6, and 9 months in reduced negative (precursor) emotional reactions, reduced frequency of crying and increased communicative abilities (Ainsworth & Bell, 1974).

It is important to note, however, that sensitive intuitive parenting is not simply reactive. It is also proactive in terms of selecting appropriate environments to optimise infant arousal and in actively seeking contact with the infant from birth. Caregivers talk, smile, etc. and position themselves to make mutual gaze possible, which engages precisely those resources with which infants are innately equipped.

Interest and focused attention are triggered by novelty of external stimulation, and novelty is tied to certain temporal, sensory and spatial contingencies. Neonatal visual perception is limited to 20–25 cm and exhibits a preference for face-shaped forms (Simion, Cassia, & Turati, 2001), and neonatal auditory perception exhibits a preference for the frequency, range and pattern of human speech (Papousek, 1994). The “speaking” and slowly moving face of the person holding the baby in her/his arms (Langsdorf, Izard, Rayias, & Hembree, 1983), moving slowly enough to track (Brazelton, 1983), therefore, is one such contingency. Neonates do not simply respond to such contingencies, however; they also actively search for them (Stern, 1992).When interest is stimulated, the baby ceases unfocused motor activity, orients towards and visually fixates the stimulus object (Malatesta & Wilson, 1988), which signals receptivity for information/interaction to the caregiver. The caregiver assesses the arousal state of the baby and her readiness to interact, and adjusts her facial, vocal and gestural behaviours accordingly.

I stated earlier that human neonates are born equipped with certain sensory-motor competencies and cited sucking and turning of head as two of them. A third and critically important competency is that of motor mimicry (Meltzoff & Moore, 1988, 1989). Neonates imitate the facial expressions of caregivers, including sticking out of tongues and opening of mouths and even facial expressions for emotions, for example, smiling, ² frowning and pursing lips, within 2 days of birth. What this seems to imply is that neonates can translate visually perceived motor patterns into the appropriate proprioceptive ones. The mimicry competence of neonates corresponds to a complementary competence in caregivers. They adapt their communication patterns intuitively to the infant’s perceptual limitations to reduce complexity; they mirror babies’ expressive reactions in their own expression and use “exaggeratedly succinct” (Holodynski & Friedlmeier, 2006, p. 100) expression signs, for example, exaggerated smile, in order to elicit an imitated smile, with its accompanying feeling, in the infant. This enables the infant to experience contingencies between expression and feeling (Gergeley & Watson, 1999). These exaggeratedly succinct expression signs are accompanied typically with “baby talk”, for example, “Oh, Amy love Teddy …” “Amy cuddle Teddy …”

The Socio-Cultural Context of Development: Emotional Outcomes

As a result of this type of repeated interaction, in the first 2 years of life the infant (a) constructs differentiated emotion systems mediated by expressive reactions such as joy, anger, sadness; and (b) acquires a repertoire of appropriate coping mechanisms. It is through sensitive and prompt nurturing that carers create contingencies between emotion elicitors, their infant’s precursor emotional responses to these, their own interpretation of the infant’s needs and their own actions to deal with them. These contingencies constitute mind–world constancies which are exploited by the neonate to fine-tune her assistance-soliciting devices in emotional ontogenesis, and by the caregiver to fine-tune her assistance-providing devices. It is through this process that the differing components of the emotion become integrated into differentiated emotion systems.

If the caregiver reacts sensitively, promptly and consistently to the infant’s expression signs, the infant will construct temporal contingencies (mind–world constancies) between cause, own/other appraisal, own expression and action by others. Certain expression signs in this sequence of contingencies then prove more efficacious in eliciting caregiver assistance, thus increasing the probability that they will be used again in a similar situation. What this implies is that during this process the infant’s assistance-soliciting devices are progressively fine-tuned.

The reciprocal influencing of appraisal patterns and expressive reactions is generated by the special context in which infant emotions develop: Fogel (1993) terms this interdependent relationship coregulation to emphasise the interdependence. Infants build up their emotion-specific appraisal patterns not just in physical space but also in semantic space, in which their emotional experiences are mediated by the interpretations of caregivers. Infant expressive behaviours play an important role in the emergence of these interpretations and not just as instrumental adaptation reactions to the physical environment, but also as indexical and iconic signs that appeal to the caregiver. Caregivers infer their infant’s emotions and intentions through the coincidence of contextual features, knowledge of their infant’s developing character, her current expressive and body reactions, and coordinate their responses with this interpretation of their infant’s expression. In turn, they infer the rightness or wrongness of their interventions based on their interpretation of feedback from the infant’s expressive and body reactions.

It is worth emphasising that this interdependence of neonate and caregiver in human emotional ontogenesis has been increasingly acknowledged in developmental psychology since the early 1990s, for example, in Fogel’s (1993) coregulation model and, if Vygotsky’s work is to be included, since the 1930s. Conceptions of emotional development have progressed from unidirectional models (where parental behaviours were viewed as antecedents and neonatal behaviours as outcomes) to bidirectional models, to transactional/circular models (Maccoby, 1992) and, most recently, dynamicist models (Fogel, King, & Schanker, 2008; Lewis & Granic, 2000). The early “constrictions” (Lewis & Granic, 2000, p. 1) have been progressively challenged as relatively impoverished in explanatory power, especially since the development of sophisticated microanalytic videotaping and computer technologies. Such technologies capture the synchronised, continuous parent–child–parent relationship in which it is impossible to pinpoint the influence of one partner on the other at any given moment. This is demonstrated in mutual gaze interactions: The infant’s gaze elicits the caregiver’s gaze which perpetuates the infant’s gaze . . . In short, the two gazes are both effects and causes simultaneously in mutual gazing and this is because they each affect the other’s central nervous system simultaneously (Shore, 1997).

An important result of this child–caregiver interaction is the ability of toddlers to coordinate the satisfaction of their own motives with the cultural norms of their social environment. This, too, is a function of caregiver behaviour. Parents cease to respond unconditionally to their children’s demands and begin to urge them to regulate their actions and their emotions themselves. They also start to explicitly evaluate their children’s behaviour in light of cultural norms and values. This results in a growing ability in children to evaluate their own reactions in light of such norms, to attempt to control norm-violating emotions and to divert reactions into the appropriate channels. This is the process through which shame, pride and embarrassment are developed. This development is scaffolded by the increasing use of language in both caregivers and children in emotional regulation. Parents increasingly require children to request assistance verbally rather that just whine and hold out their hands, and children increasingly use verbal instructions/requests to others and to themselves (Holodynski and Friedlmeier, 2006). It is through this process that the interpersonal regulation typical of juvenile emotionality is gradually replaced by the intrapersonal regulation typical of adult emotions.

There are two important points to note at this juncture. First, the ability to emote properly at all, as a human being, is a function of social relationship. When everything is functioning normally, that is, ideally as it was designed to (but seldom does) in terms of neonate and caregiver input and output, the relationship provides the necessary and sufficient conditions to initiate the development of full human emotionality from a set of culturally universal and monomorphic emotional precursors. This emotionality includes the identification of culturally relevant emotions, their culturally relevant elicitors, what they are called/labelled in the emoter’s particular culture and what the culturally appropriate response is.

Second, and more specifically, the assistance-soliciting mechanisms in the infant and the assistance-providing mechanisms in the caregiver(s) are coconstructed or learned through linguistically mediated social interaction. Referentially opaque precursor expressions in infants, for example, unfocused crying and motor unrest, trigger non-specific general purpose assistance-providing devices in caregivers (to search, identify, correct) needs/deficits. These repeated search processes serve to fine-tune the assistance-providing devices to the infant’s idiosyncratic needs, causes and satisfaction conditions; these are fed back to the infant in linguistically-mediated social interaction (“Tommy sad . . . Tommy lost toy . . .” “Tommy look for toy . . .”) to promote the maturation of the neural substrata of: (a) specific assistance-soliciting devices, for example, sadness in the infant, and (b) her own assistance-providing devices. What this implies is that emotion-related assistance-soliciting and assistance- providing devices in human beings are largely learned through social interaction. These assistance-soliciting devices will, of course, include a limited range of in-built emotion-elicitors, for example, fright response to precipice stimuli, but it will also include a whole raft of culturally relevant and culturally appropriate elicitors and responses, and culturally relevant linguistic labels for these. In Japan, for instance, amae is a pleasant feeling of dependence on individuals or institutions (Morsbach & Tyler, 1986) which is entirely appropriate to Japanese culture where dependence is viewed positively.

What these observations further imply is that the emotions that an individual develops, their elicitors and their responses depend on features that are a function of the socio-cultural environment in which the subject-emoter is embedded. What this in turn implies is that culture makes a serious constitutive contribution to the development of human affect programmes. More specifically, culture and language are as important to the maturation/development of human affect programmes as they are to the development of higher sentiments.

Individualism and Externalism: A Very (Very) Short, Potted History

One of the current orthodoxies in the philosophy of emotion is that they occur at the interface of our internal concerns and the outer environment and, as such, are essentially what they are in virtue of their relational/functional properties. This orthodoxy, however, conceals what continues to be a deep division in philosophy of mind, that is, the division between individualists and externalists in terms of both metaphysical determination and how psychological states should be taxonomised/individuated. Very briefly, individualism arose as a means of “disciplining” psychology (Wilson, 2004), that is, of rendering psychology scientifically respectable. Individualism began as a normative constraint on psychological methodology, the principles of which were thought to be derived from the most successful explanatory practices in science. Such practices taxonomised their investigation targets as supervening on the intrinsic physical properties of the individual object or organism which instantiated them.

Individualists and externalists agree that agents causally interact with: (a) their physical environments, to collect information relating to them using their different sensory apparati, and (b) their social environments, gathering information from their communicative interactions with others (Wilson, 2004). The nature of their minds, therefore, and in particular the intentionality (“aboutness”) of their mental states, is at least partly causally determined by the character of their physical and social environments. Individualists and externalists agree, therefore, that the environment causally contributes to the contents of mind, in terms of what is perceived, thought about and desired. Where they differ is that individualists insist that the differences in the world cause some corresponding change in physical brain states which gives rise to differences in content. Taxonomic externalists deny this. They insist that physically identical doppelgangers could have brain states with different content and, given that mental states with different content are ipso facto different types of states, their bearer’s intrinsic physical states do not fix/determine her mental states (Wilson 2004, p. 80).

The classic Putnam–Burge thought experiments of the 1970s ³ sought to challenge individualism by demonstrating the importance of external conditions on individuating mental content. Taxonomic (or semantic) externalism arose as a result of these challenges; it continued to construe content-laden mental states as being internal to the individual who bears them, but allowed that they were not metaphysically determined (i.e., did not supervene on) the intrinsic physical states of the individual agent.

The last decade has witnessed a seemingly progressive radicalisation of externalism (Robbins & Aydede, 2008), the most radical of which is currently vehicle or locational externalism. Vehicle externalism claims not only that meaning can be externally derived (as taxonomic/semantic externalists do), but that the vehicles (carriers) of mental content can extend beyond the skull and skin of the cognising individual into her physical, technological and socio-cultural environments.

Metaphysical Realisation

The realisation theory of mind began as the thesis that the brain should be understood as a computing machine and the mind as a programme or set of programmes. This thesis reflects the view that computers implement or realise the programmes or algorithms they run. Certain electrical states of the device realise computational states; the electrical activity of the device is not identical to any programme state of Microsoft^® Word, but it implements or realises such programme states (Polger, 2007, p. 236). In exactly the same way, it was claimed, mental properties are realised by physico-chemical brain states but are not identical to them. Despite the importance of this claim, however, realisation has received surprisingly little serious attention until recently (Polger, 2004, 2007; Wilson, 2001, 2004).

Realisation is a relation of metaphysical determination (Wilson, 2001, 2004); realisers determine the relevant realised properties and currently there are two views of realisation physicalism (Polger, 2007), namely, the Standard View and the Context-Sensitive View. The Standard View is espoused by individualists, and the Context-Sensitive View by the proponents of vehicle externalism. This latter view, it is claimed, does more justice to the concept of realisation operational in the biological, behavioural, and social sciences (Wilson, 2004).

Summary

The metaphysical realisation of full human emotionality is a function of the heterogeneous but complementary combination of partial realiser systems which are internal (genetic/biological) and external (sociocultural). Together, these partial realiser systems constitute the total realiser systems of distinctly human emotionality and provide the necessary and sufficient conditions for human emotional ontogenesis and experience. Distinctly human emotionality, therefore, is neither (mainly) genetically hard-wired nor (mainly) socially constructed; rather, it is crucially dependent on the complex coaction of genetic endowment and social environment.