Book Reviews

Jennifer Greenwood, Becoming Human: The Ontogenesis, Metaphysics, and Expression of Human Emotionality, MIT Press, 2015

Becoming Human by Jennifer Greenwood is one of the most thought-provoking books on emotion and its expression I have read. At its core, it attempts to provide an account of the development of full human emotionality and in so doing argues the emotions are “transcranial.” Emotions are radically realized outside our nervous systems and beyond our skin. As children, we are functionally integrated affectively with our mothers; so much so that in a sense our emotions are not ours alone. Regardless of whether one agrees with her radical claims, it is a must-read for those interested in emotion and expression. In order appreciate the significance of this book, let me sketch its contents and raise a few criticisms.

Many, but certainly not all, psychologists and philosophers assume that there are basic emotions (BEs) and higher-cognitive emotions (HCEs). The former include fear, anger, disgust, happiness, surprise, and sadness; and the later include guilt, shame, and pride amongst others. BEs are thought of as natural kinds involving facial expression, homologous traits shared with non-human primates, specific brain structures, and stereotyped behaviors. HCEs differ in that they often do not have unique physiological profiles, facial expressions, dedicated brain regions, and culturally vary quite a bit. Greenwood argues that there are affective precursors that develop into BEs and HCEs. However, the distinction between BEs and HCEs lulls us into naïve views about nature and nurture, biology and culture. We have not taken their development from childhood as seriously as we should. Both develop through time.

Greenwood has us consider human infants. They are completely dependent on their caregiver who is usually their mother. They can cry and exhibit motor unrest to convey how things are going for them. Their emotional precursors are ostensive and expressive. They are ostensive insofar as they direct the attention of the mother to sources of displeasure and they are expressive insofar as they signal that displeasure. They are referentially opaque but through time become less so. Thus, these emotional precursors are natural signs which occur when proprioceptive and interoceptive thresholds have been crossed and help is needed. Crying and motor unrest are assistance-soliciting devices. Mothers are equipped with “intuitive parenting” skills by which they can identify what things are helpful and harmful to their child. They instinctually respond to the child’s stress and try to remove the sources of it. This also involves mimicry in speech and mirrored facial expressions of their child. These are assistance-producing devices. Through repeated interactions of these assistance-soliciting and assistance-providing devices, the child’s emotionality advances through functional integration with the mother. Not only do these interactions provide feelings of pleasure but also new neural machinery develops as the result of these dynamic interactions. There are complex causal interactions between child and mother, which scaffold the child’s emotions and their expression. It is also through the same mechanisms that language appears as well. “Motherese” helps the child to understand the boundaries between clauses and how to command, request, declare, and to question. Thus, our first conversations are with our mothers. Greenwood offers an explanation of how infants begin with affect expressions and species-typical behaviors to express themselves, move to communicating with gestures and inflected vocalization, and finally develop language. And, referential clarity increases with the development of the emotions and language. The child’s species-typical behavior patterns coupled with the mother’s intuitive parenting skills provides the necessary and sufficient conditions for generating a semantic lexicon. These natural and conventional signs have their declarative and imperative content secured by a kind of teleosemantics inspired by the work of Ruth Millikan.

These are bold hypotheses that Greenwood has offered. She additionally argues that the hypothesis above gives strong support to Transcranialism, or the Extended Mind hypothesis, about the emotions. It says that mental states and processes extend outside the nervous system. Intracranialists deny this claiming that the scaffolding provided by the world outside the body is simply that of “support”. That is, they claim Transcranialists conflate causation and constitution. Greenwood utilizes the work of Robert A. Wilson (2004) on the metaphysics of realization to challenge this. Roughly, a property R realizes G if it is plays the G-role or has the G-function. Philosophers of mind have typically assumed that the physical properties sufficient for mental roles or functions were exhausted by the intrinsic physical properties of individuals. However, this is exactly the assumption that Wilson and Greenwood challenge. According to Wilson, there are core and non-core realizers. Core realizers are those parts of a system which play a “crucial role” in producing or sustaining a property of interest. Non-core realizers are necessary but are not crucial for the producing or sustaining the property of interest. For example, a core realizer for fear is the amygdala whereas the thalamus is a non-core realizer. Transcranialism about emotions claims that the realization of emotions is radically wide; a total realization of an emotion can have core and non-core realizers located outside the individual experiencing the emotion.

As we have seen, Greenwood claims emotions are essentially ostensive-expressive devices that evolved by natural selection for interpersonal tasks and only later for intrapersonal ones. She also argues that Wilson’s model of radically wide realization provides the metaphysical machinery for defending transcranialism about human emotions. Going beyond earlier externalisms, she contents the internal and external resources are deeply functionally integrated and play complimentary roles in our emotions. The mind extends our emotions into the world, and the world extends into our emotions. It is “synchronous modulation” of the child and mother during the development of emotions and language which widely realizes those emotions.

Greenwood’s claims and arguments are extremely bold. Let me conclude by raising some points of criticism. Greenwood has effectively argued that human infant’s emotions develop through a dynamic coupling with their mothers. They are functionally integrated. I find the evidence she presents fairly convincing. If child and mother are functionally integrated with respect to the child’s emotions, then they are widely realized. If they are widely realized, then those emotions are transcranial. Therefore, those emotions are transcranial. First, like Victor of Aveyron, some feral children develop without the benefit of a mother or even a direct caregiver. They are raised by wolves as it were. Though they may be developmentally challenged in various ways, one might argue that they experience a full range of human emotions. This suggests that the dynamic coupling with a caregiver, and specifically a mother, is not necessary for the development of human emotions. Without good empirical evidence, I am suspicious of this claim. Anecdotal reports of feral children strike me as inconclusive at best. Second, let’s suppose that this dynamic coupling is required. Still, one might object even accepting that it is through this functional integration the emotions are widely realized, we needn’t accept that they are transcranial. Striking a match causes a fire at the end of the stick but only so if there is oxygen present. We treat the striking as the cause and the oxygen as a background condition. Thus, even if we grant that a core realizer of a child’s fear is the amygdala and non-core realizer are the mother’s assistance-producing devices, it seems that we can still appeal to the cause/background condition distinction. Through this, we can resist the more radical transcranialist’s claims. Of course, much more work is needed to make good on this claim. For example, the cause/background condition claim must be more than pragmatic. However, if we can, then I think we can argue that functional integration is causal and not constitutive. Third, let’s suppose that Greenwood has made her case that the emotion experienced by human children are transcranial. She grants human emotions move from interpersonal functions to intrapersonal ones. I doubt that emotions are generally interpersonal or intrapersonal en masse. Still, it is open whether at least some human emotions after childhood are exaptations to new intrapersonal challenges. If this is correct, then even if human emotions during childhood evolved to be transcranial they become intracranial later in life. The emotions of adults are screened-off from those of human children in a way that doesn’t threaten the project of psychologists and philosophers I mentioned at the beginning.

Regardless of my worries, Greenwood’s is an impressive book from which I learned quite a bit. I strongly encourage those interested in emotions and expression to read it. Your mother would have wanted you to.

Jay Odenbaugh
Associate Professor
Department of Philosophy
Lewis & Clark College
Portland, OR, USA


Wilson, Robert A. 2004. Boundaries of the Mind: The Individual in the Fragile Sciences Cognition. Cambridge University Press.