Thursday, July 25, 2013

Empathy and language

The practice of pointing by infants raises some interesting questions about the psychological foundations upon which human communicational and linguistic capacities are built.

As explained in an article cited in the comments section of the previous post, young children routinely point to direct the attention of a nearby adult to something the infant finds interesting and apparently wishes the adult to see and appreciate also.

When an infant doesn't start pointing by the appropriate age (about 12 months), it's often a sign that they don't have an intuitive sense of other minds – and also of linguistic problems ahead. (I originally came across discussions of this phenomenon in material on identifying the early signs of autism.)

The article referred to above draws on papers by Michael Tomasello and his colleagues which explore the phenomenon of infant pointing and associated behaviors. Tomasello and his fellow researchers argue for "a deeply social view [of the process] in which infant pointing is best understood – on many levels and in many ways – as depending on uniquely human skills and motivations for cooperation and shared intentionality (e.g., joint intentions and attention with others). Children's early linguistic skills are built on this already existing platform of prelinguistic communication."

The researchers note that the kind of pointing they discuss is unique to humans and depends on certain key insights about the existence and nature of other minds as well as emotional factors – essentially a desire to share one's perceptions and to share in the perceptions of others.

A cursory reading of sources cited in the Slate article and related material suggests to me that Tomasello and his colleagues may well be overplaying their intuitions about sharing in their claims about the origins and development of human communication and language.

Of course, emotional factors cannot be ignored, but could not these elements be explained in terms of cognitive imperatives and the practical benefits of collaboration and reliable information transfer?

György Gergely and Gergely Csibra explicitly challenge Tomasello's views on the centrality of the emotions associated with shared intentionality and focus instead on the communication mechanisms necessary to ensure efficient cultural learning.

A crucial point relates to the efficacy of the highlighted emotions. Tomasello and his colleagues posit the desire to share emotional states as a key explanatory factor rather than merely as one element in a diverse suite of human abilities and behaviors.

But I am nowhere near having a sufficiently strong grasp of the material to take sides in this dispute.

It is clear that the same (or similar) perceptions and feelings which apparently motivate gestural communication – however we might characterize them – certainly do seem, in normal infants, also to motivate and facilitate the child's rapid and apparently easy acquisition of whatever language or languages they are routinely exposed to.

Significantly, though, the complexities of language can be learned (albeit often with some difficulty) even by those who lack a strong intuitive sense of other minds.

It's certainly plausible that the historical development both of prelinguistic modes of communication (like pointing) and language amongst our ancestors was dependent upon (amongst other things) certain empathetic perceptions and feelings. But, of course, the cognitive and affective factors involved are in practice always inextricably linked, sometimes in very complicated ways.

In his work on autism, Simon Baron-Cohen distinguishes between the cognitive and affective aspects of empathy. Cognitive empathy is all about what we perceive and understand about the mental states of others, whereas affective empathy concerns our emotional responses to this knowledge. Strength or appropriate responses in one area does not necessarily entail strength or appropriate responses in the other.

For example, the autistic person typically scores poorly on tests of cognitive empathy (e.g. reading particular emotions in pictures of faces cropped to reveal little more than the eyes), but often exhibits appropriate affective responses (e.g. to perceived suffering). By contrast, the psychopath typically has no problem at all with cognitive empathy (or language, for that matter), but displays deficiencies in terms of affective response.

Speculations about the way language evolved will necessarily draw on the findings of cognitive and developmental psychology as well as other areas. But, while it is reasonable to assume that affective responses played a role in the development of language, I have some doubts about the way Tomasello and his colleagues present the basic issues and about some of their key claims.

Also, as someone with a background in formal approaches to language and syntax, I am naturally wary of approaches which downplay the significance of this side of things. I was unimpressed, for example, by the comments by one of Tomasello's co-researchers, Malinda Carpenter, quoted in the Slate article.

The fact that pointing seems to call on a sophisticated understanding of what is going on in the heads of other people, she noted, "suggests that [infants] can do so much more with pointing prelinguistically than we ever thought before."

Until recently, people thought that this sort of knowledge only emerged with language. But when Carpenter, who was drawn to this work through an initial interest in language, started looking at prelinguistic gestures, her perspective changed.

"[E]verything’s already there!" she said. "I completely lost interest in language because you can see so much complexity already in infants' gestures."

It depends on what you mean by 'everything', I suppose, but I would have thought that language adds a little something to the mix.

Monday, July 8, 2013

A science of language?

A large part of the fascination which language holds for many is that it is one of the key markers of our humanity. Language is at the heart of human culture and human consciousness. Tense and aspect mark our sense of time, grammatical mood our sense of possibility, personal and possessive pronouns our very sense of identity and how we see ourselves as relating to other people and things.

Partly because language is an inextricable and defining part of us – and at once social and individual – it is impossible to clearly define a science of language in the way most other sciences can be defined.

To what extent should the study of language be subsumed into psychology and neuroscience? Language is behaviour, and the human language faculty can only be said to be understood to the extent that the neurological processes which drive it are known.

On the other hand, language is also a cultural object which can be studied in its own right, both structurally and historically.

It's hardly surprising, then, that, since its rise to prominence in the 19th and 20th centuries, linguistics has, as sciences go, been unusually riven by competing frameworks and approaches, and these divisions have, if anything, increased over time. (Though I sometimes wonder how different things might have been if the later-20th century's most prominent linguist had not been such a relentless intellectual warrier and contrarian!)

Ultimately, the divisions between the sciences are merely for practical and administrative purposes: the quality – and worthwhileness – of research is not generally determined by discipline-specific but rather by more general criteria.

But I don't want to get into an abstract discussion about the unity of science or related matters. I really just wanted to make the point that language represents not so much a subject area as a number of interrelated subject areas. And, because the phenomenon of language can be approached from very different directions, it is difficult, if not impossible, to pull all these perspectives – and the knowledge implicit in them – together.

Perhaps, then, the best we can do is to focus on specific questions which may happen to relate to language in one way or another and to renounce as unrealistic the desire for a comprehensive understanding of the phenomenon of language per se.


I'll finish by mentioning a couple of language-related topics which I have been thinking about lately.

Last month I referred to the ideas of Simon Fisher and Matt Ridley on culture-driven gene evolution. The work of Fisher and others has shown that the FOXP2 gene has a crucial role to play in human linguistic abilities. The gene occurs in other species in slightly different forms and it plays various roles. Interestingly, it has been shown to play a key role in vocal expression in both birds (canaries and finches) and chimpanzees as well as in humans. Neanderthals are now believed to have had exactly the same form of the FOXP2 gene as modern humans.

I can't help thinking that the question of the origin of language retains its fascination in part because it promises to reveal something important about who we are and where we came from.

This is, I think, largely an illusion based on the idea that the abrupt discontinuity we see between ourselves and our nearest relatives (chimpanzees) always was. But intermediate forms did exist (until relatively recently, in fact).

In practice, I think we tend to assume, consciously or unconsciously, that our species has an essence.

It hasn't. Nonetheless, the development of human language as we know it does mark a clear historical and cultural discontinuity.

On a more practical note, I have also been thinking about the reputed benefits of bilingualism. It has been claimed, for instance, that bilingualism can delay the onset of the symptoms of Alzheimer's disease by about five years. I have some reservations about the significance of these claims. More another time.