Insights from Language: What Are Thoughts Made Of?
Cognitive linguistics explores how language expresses thought. What does language tell us about cognition, and what does cognition tell us about language?
I have come to the International Cognitive Linguistics Conference to find out. This weeklong conference of 480 speakers from 50 countries is taking place July 10-14, 2017 in Tartu, Estonia. Its program and book of abstracts provide a hands-on overview of this active field.
The conference is also the perfect chance for me to discover whether my own ideas are novel. I am trying to explain, from first principles, how we experience life in terms of three conceptual languages: argumentation (how issues matter), verbalization (how terms mean) and narration (how events happen). It is a challenge for me to figure out what implications my model of abstract thinking has for natural language.
What is the fundamental unit of language? How is a thought expressed? Different disciplines diverge on this issue. Formal logic strings together well-formed formulas, such as mathematical statements. In this spirit, analytic philosophy has focused on philosophical propositions and whether they are true or false. Traditionally, linguists diagram sentences, breaking them down into their syntactic parts. Alternatively, cognitive psychologists study concepts and associations, much as neuroscientists study neurons and connections.
Cognitive linguists fuse these approaches by considering an assembly of connected elements, a frame or model, an instantiation of a constructional schema. "May I have a glass of water, please?" refers to a shared schema - a container ("a glass of water"), along with our relationships to it and to each other. We understand that the glass is a container for the water, and that if you hand me the glass, then I will receive the water inside of it; hence the "of" in "a glass of water". The cognitive model suggests that participants in a conversation leverage a vast body of knowledge, interlinked in a variety of mental schema, and with each new sentence they explicitly and selectively refer to particular aspects of a schema, introduce new schema or go back to earlier schema.
The basic idea of cognitive linguistics is that language use depends on the same set of mental faculties as the rest of cognition. Semantics, morphology, syntax, pragmatics, idioms, metaphor, etc. can all be assembled with the same set of cognitive operations. This stands in contrast with generative grammar, as developed by Noam Chomsky and others since the late 1950s, which supposes that the human brain has specialized modules which apply rules to generate grammatically correct statements.
A few months ago, I started watching linguistics videos, such as 50 Years of Linguistics at MIT, but also George Lakoff on Embodied Cognition and Language and How Brains Think: The Embodiment Hypothesis. I was looking for insights as to what are the basic elements of semantics and syntax that I might try to metaphysically derive from first principles. I stumbled upon informative video lectures by Martin Hilpert on Cognitive linguistics and Construction grammar. I realized that the biannual worldwide conference would be held in Estonia, eight hours by bus from my home in Lithuania. Ronald Langacker, the founder of cognitive grammar, was to speak there! I started reading his textbook, Cognitive Grammar: A Basic Introduction.
Early on in his book, section 2.2.2., he gives an example that got me thinking. Traditionally, logicians, mathematicians, philosophers and dictionary editors have taken definitions to be the foundations of meaning. Linguists and psychologists, in studying sentences from everyday life, have understandably moved away from such a lifeless outlook, which supposes an axiomatic, hermetic, closed system. Instead, they considered meanings of words to be given by flexible bundles of attributes. Cognitive linguists step beyond words by grounding meaning in schemas that can be referenced by parts of words, entire words, phrases, sentences and groups of sentences. Langacker characterizes a "drinking glass" as accessing a matrix of domains. It indicates a space; a shape - roughly cylindrical; an orientation in space - along an axis, with the closed end at the bottom; a function as a container for liquids; a function as a utensil for drinking; a material - typically, glass; a size; and other domains related to its cost, washing, storing, dropping and breaking, its position on a table at mealtime, matching sets, method of manufacture, etc.
I grew skeptical that we assemble meaning in this way, even though I find his various schema quite believable. I wondered how this all served the purpose of language. A common view is that language serves to coordinate information exchange. But a simple thought experiment shows that people typically don't share knowledge in any absolute sense. Two people may seem to agree on what a "glass" is. But if we gradually alter its shape, for example, if we gradually make it wider, then at some point the two people will disagree. Thus people only agree pragmatically, tentatively, for the sake of something else, for example: We want to quench our thirst! Most any "glass" will do.
What is the purpose of language, given the lack of necessity of complete agreement, and the challenges faced by adults or children who have little knowledge of a language? I concluded that language serves to coordinate activity, especially by referring to standardized, recurring activity.
Architect Christopher Alexander's theory of pattern languages is very helpful here. I think of him as the Plato or Kant of our times. He is arguably the most influential thinker
What is real...
Lithuanian women - Giedrė Junčytė - explained that while people were asking what language is, little progress was made. But once that question got put aside, a lot of knowledge was accumulated about language.
Ronald Langacker, the founder of cognitive grammar, gave the plenary talk, "Functions and Assemblies".
Talk he gave.
Absolute truth... and nonabsolute truth...