How do children learn how to speak?

During the last couple of weeks of our unit on foreign language learning, we will explore a bit the theoretical foundations of many of the principles we have put into practice in the previous weeks, which led us to make a proposal with some activities to learn English in preschool, which I presented here a few days ago.

So far, we have only begun to read one chapter of Michael Tomasello's book titled Constructing a Language: A Usage-Based Theory of Language Acquisition in groups. Our group will work on the chapter dealing with abstract syntactic constructions, which we will need to understand in order to explain its contents to the rest of our class. Besides doing that, we will have to write a summary on two articles about related subjects among a selection offered by our teacher. I have chosen the following two:
   Andrea Tyler (2010). Usage-Based Approaches to Language and Their Applications to Second Language Learning. Annual Review of Applied Linguistics, 30, pp 270-291. doi:10.1017/S0267190510000140.
   Nick C. Ellis (2002). REFLECTIONS ON FREQUENCY EFFECTS IN LANGUAGE PROCESSING. Studies in Second Language Acquisition, 24, pp 297-339. doi:10.1017/S0272263102002140.

I was thinking about the book by Tomasello when I woke up this morning. I love weekend mornings, because I can lie in bed, quietly thinking, while the day starts. Objects begin to recover their shapes first, then their colours, while I let my mind wander.

I only managed to read a few pages of the chapter yesterday. As one of my group mates pointed out, Tomasello is an honest writer, and he sets very clearly what he intends to defend and his view on linguistics at the beginning of the chapter. The problem is that this reading is far from Ausubel's meaningful learning for us, in the sense that it takes a giant leap from our previous knowledge on the matter, leaving us with a basis too small to grant full understanding of what we read. Nevertheless, we will do our best; explain what we believe we have understood, and make an educated guess on the rest.

The title of Tomasello's book states clearly that his theory is set within the usage-based approach to language acquisition. Now, I hope to learn a bit more on this approach in one of my additional readings, but what I know so far is that this approach defends that language use is the main driving force for the acquisition of language, and that communication is at the core of the whole process.

Before continuing with Tomasello's introduction on his chapter, I must say that the title of his book seems a bit contradictory to me. Is language constructed, or acquired? By acquisition I understand getting something that is outside of yourself; taking it. Constructing would refer to building it inside out; just the opposite. Developing it would refer to some previously existing guideline or plan. So, I see that one needs to carefully choose the verb on a topic like this. From the little I know, I would say that I think that as individuals we recreate language; we recreate the universe that our community offers to us. And through the negotiation on the individual recreations of the whole community, we create language, understood as a common cultural element. I am sure there are proper scientific terms for my "recreation" and "creation", but unfortunately I don't know them. How else can we explain why languages change and evolve, if there is no room for individual and collective variation? The closest I have come to reading anything on language ecology is actually related to the phylogeny of a tale, for an assignment on our unit about plastic arts, but as a biologist the idea of languages evolving similarly to living organisms looks attractive. It would be nice to learn a bit about that, but it doesn't seem like it will be in our group's chapter, although another one towards the end of the book looks promising.

Returning back to the introduction on our chapter, I must admit I don't quite understand what abstract syntactic constructions are, as the examples of abstract experiential scenes Tomasello mentions seem quite concrete or objective to me: people acting on objects, objects changing state or location, people giving people things, things being acted upon. The most abstract among those he mentions are people experiencing psychological states and objects or people being in a state.

Regardless of whether Tomasello's abstract is abstract enough or not, it is quite clear what he is talking about. He also sets some basic concepts in the introduction of his chapter. On the one hand, he proposes some cognitive processes that will enable children to create these abstract constructions: analogy and distributional analysis. Thus, apparently children compare utterances (units of speech, so to say) and build new ones based on the structural similarities they detect or foresee. In other words, children "guess" language through analogies. When children apply distributional analysis to language they analyse the components of language and group together elements that behave in the same way, although Tomasello talks about functionally based distributional analysis, where the classification of language elements would result in grouping together elements that play similar communicative roles. So, children would also put together elements of language that are used to name objects, for instance, in a different group from those which are used to name actions.

On the other hand, Tomasello explains two other processes that children use to create these abstract constructions: generalisation and constraint. Generalisation would be an outward driving force, which would lead children to apply a rule identified in several elements of the same group to others. For example, I remember that when I was in my Practicum I, one of the 4 year-olds in the class was new to Basque, and he had begun to construct Basque verbs by adding the particle -tu to Spanish words for verbs (rompi-tu, corta-tu, etc.). He was generalising; actually, he was overgeneralising. Constraint would be an inward driving force that would prevent children from overgeneralising or, more precisely, would allow them to generalise only in the direction their linguistic community deems appropriate.

Quite clearly, constraint requires more fine tuning than generalisation does, so more efforts have been employed trying to explain it. Generativists rely mainly on innate universal grammar to explain it, so an intuitive knowledge about language would guide children in the complex dynamics between generalisation and constraint. Tomasello isn't among those who believe such things, but rather believes in the usage-based approach, where three processes have been proposed to explain how this happens: entrenchment, preemption and formation of mainly semantic verb classes. Tomasello doesn't explain in depth these three processes, so in order to understand them a bit I have found another article. Preemtion means blocking in this context, so the "right" or conventional form used by the adult would eventually block the unconventional form created by the child. Entrenchment refers to a repeated use of a verb in different sentences, which would lead the child to infer that a use different from what has been heard will be ungrammatical, so it works well for elements that are quite frequent in the input children receive. The third process occurs when children group together verbs in different categories (verbs about "ways to move", such as bounce, roll or slid, for instance), and predict how a verb new to them will behave based on the behaviour of another one of the same group that they already know about.

We must note that the concepts Tomasello's book deals with are currently being debated among the scientific community, and Tomasello's views might not be as widely accepted as one might think while reading the book. For instance, the author of the article I just mentioned is much more critical about the effectiveness of the three processed described above regarding constraint. I highly doubt the average preschool teacher, among which I intend to be one day, will be able to form an opinion on such matters but, nevertheless, this reading is a very good exercise for many other things.

So far, I am afraid I have only covered the first page of the chapter, and I wasn't even meant to write about it! Well, in summary, Tomasello believes that children develop syntax (the principles that rule how small elements of language are combined to create larger elements, such as sentences) through the use of language, and paying special attention to the communicative intention and function of those elements, that is, what they are used for, and why they are used.

I would like to end with a reflection on the cognitive processes of children, compared to those of adults. I haven't been able to read what Tomasello writes about it, so I really don't know what he has to say over this matter. Still, I have a feeling that, in general, we tend to think about children's productions as more rudimentary than those coming from adults and, by extension, of children's thinking as more rudimentary than adult thinking too. I just revised this morning what we wrote about this issue in our experimental sciences didactic sequence, and added a reference to a really interesting documentary series titled Baby Human, which we watched last year and I liked very much. One of the six episodes of the series was devoted to baby thinking, and it proves how sophisticated it is. As a matter of fact, it seems children are very smart before they start school, as Ken Robinson states in his talk on changing paradigms:

Apparently, they lose part of their divergent thinking as they get older in school. On the other hand, during the degree we have had the opportunity to learn that children have some anatomical or physiological adaptations which are lost after some months or years after they are born, such as reflexes, the ability to breathe and swallow at the same time without choking. They are even able to put their toe in their mouth, something that only few well trained adults can do! Maybe they also have additional cognitive abilities which adults lack. Maybe they ARE smarter than we are. Actually, I would argue that THEY MUST be smarter, because they need to learn more than adults. If children on the Baby Human series could tell us what they are thinking, I bet we would be quite surprised!

I can't help but think on "A room of one's own" by Virginia Woolf I was reading last night:

"Women have served all these centuries as looking-glasses possessing the magic and delicious power of reflecting the figure of man at twice its natural size. Without that power probably the earth would still be swamp and jungle."

Children are also comforting mirrors for us adults, but I wouldn't be surprised if in a few years from now the idea of children being smarter than adults were quite widespread.