Attention! It's a Revolution.

When I began planning this blog I did not expect to be reporting on a revolution, at least not right off the bat. But a revolution is underway and it turns out that this blog reports on its crossroads, the place where attention meets society. The origins of speech are critical to that junction.

This blog declared itself for the revolution almost immediately in a series of posts headed “What Evolved?” (Parts 1, 2, and 3). They argued that it was a mistake to think of language as an evolution of communication as expressed in theories of how to control machines (information theory). Speech evolved as an elaboration of an already existing capacity for joint attention and provides a tool for the social sharing of attention.

Other scholars and researchers had declared for this revolution earlier. Michael Tomasello of the Max Planck Institute for Evolutionary Anthropology wrote in a volume titled Language Evolution (2003),

My proposal is that symbolic communication is the process by which one individual attempts to manipulate the attention of, or to share the attention with, another individual. (pp. 94-95)

Still earlier, over a quarter of a century ago, M.A.K. Halliday noted the critical role of joint attention in children's acquisition of language. It's not just that they need to pay attention; they need joint attention before they can start speaking about things. What this blog calls the attention triplet (speaker, listener, topic) is required from the beginning of each person's speech.

Now, an Italian scholar, Giorgio Marchetti, at the University of Urbino has provided a full throated manifesto for the revolution in an article titled, “A Presentation of Attentional Semantics,” published in the current issue of Cognitive Processing (abstract here). It argues that words are tools to pilot attention and proposes a semantics (attentional semantics) that aims to find the attentional operations produced by the meaning of words.

Marchetti's paper is about semantics, not origins, but it provides a foundation for the claim that speech is an extension of a rudimentary power among among social mammals, a faculty that that became something new when words were introduced. His semantics takes seriously the relation between meaning and attention.

The problem with meaning is the way semantic theories tend to be incomplete. Common sense typically uses the word as interchangable with intentions. "Sir, what do you mean by that?" can commonly be understood as asking about intentions. This approach can be useful in thinking about why politicians, sales people, and master sergeants say the things they do. It is less helpful for analyzening concrete statements of the the-cat-sat-on-the-mat variety.

Logical semantics does well with cat/mat type statements. For that school, meaning refers to the objective relationships between things. Symbols are "tokens" pointing to something in the world. Einstein compared symbols to coat-check stubs. A stub is not a coat, but it points to one somewhere.

That approach is well regarded these days, but gets into trouble with sentences like, "Bob's passion for Alice has reached its twilight hour." Metaphors like twilight hour can (at least sometime) be translated into something objective and concrete. But why should translation be necessary, and where does the knowledge to effect the translation come from?

Some analysts, therefore, have proposed that the meaning lies in the relationship between the words themselves. That approach can tell you the difference between "He gave the dogs to Pete" and "He threw Pete to the dogs," but it has a hard time with the way the same words can have different meanings. A famous example comes from Owen Wister’s novel The Virginian. Several people slap the hero on the back, laughing and saying, “You son of a bitch.” Then somebody says it without laughing and the Virginian draws his gun, “When you call me that, SMILE.” (Gary Cooper made it, “If you wanna call me that, smile.”)

All of these confusions arise from thinking of words as representatives of something else. Where is that something else? In the head, in the world, or between the words themselves? If the answer is "any of the above," how do we know in any particular circumstance where to look? Marchetti's attentional semantics says words are go-betweens rather than representatives. They can direct attention to things in the world like the cat and the mat. Since attention can be directed toward subjective things as well as objective ones, directing attention to a memory of twilight hours poses no speciall challenge. All attention occurs within a context so that reactions to son-of-a-bitch, or most any word, will naturally vary with the context.

Marchetti does not discuss an attentional syntax that presumably would explain how attention makes senses of its shifting focus. His article focuses on the nature and source of an attentional semantics and on the taxonomy of words when they are considered as tools rather than symbols, but this blog is not really about the technical work of semantics. The important thing as far as Babel’s Dawn is concerned is the news that a wholesale reconsideration of the very nature of words and speech is underway.

• The function of speech is joint attention.
• The elements of speech are words, the tools for directing attention.
• Speech arose as part of general evolution from joint attention on the ape level to joint attention on the human level.

This week the blog will discuss some of the implications of Marchetti's paper for where speech came from and what need it serves.

Comments

[TLTB]

This has little to do with your post, but reading your blog last week it occurred to me that many who resist or argue against the idea of innateness being a source for syntactic universals don't really know what syntacticians claim those universals are. Any debate takes on a bit of a futile flavor.

So I decided to post something at my own blog on the topic (even though my blog isn't about linguistics). I enumerate four principles that syntacticians would widely agree are universal and have some basis in innateness. I really think any argument against innateness has to provide non-innate stories for at least these four specific phenomena.

---------------------
BLOGGER: This is a great service. The post can be found at http://lameandblind.blogspot.com/

[hugo]

In case you are not aware of that, Dan Sperber and some of his collaborators defend a theory of language evolution that puts the stress on pragmatics, and that convincingly argues that theory of mind must have evolved before language.
See particularly these two papers
http://www.dan.sperber.com/evo-lang.htm
http://www.dan.sperber.com/metarep.htm
and my post on the topic
http://www.cognition.ens.fr/~alphapsy/blog/?2006/09/19/30-language-evolution-where-is-my-mind
By the way, Dan Sperber is also the creator, with Deirdre Wilson of relevance theory, the most important single theory in pragmatics, and the theory has obvious links with attentional phenomena, so you might also be interested.

[Giorgio Marchetti]

Hugo

Personally, I do not know Dan Sperber's work. Thank you for the indication.

Giorgio Marchetti