Friday, August 31, 2012

A Matter of Semantics - Words, Concepts and Things


By Gary Berg-Cross


Modern society bombards us with words and no more than during a political season when candidates compete for our attention with punchy slogans and catchy phrases. It seems too close to the word manipulation of George Orwell’s 1984. 1984 isn’t yet the good old days, but people seem pretty confused by the flood of political language as we descend into a world of thin, distorted or made up meanings. Paid wordsmiths now hawk a language of fractional facts, non-sense, almost sense and truthiness that is captured in the Lewis Carroll quote:

“When I use a word,’ Humpty Dumpty said in rather a scornful tone, ‘it means just what I choose it to mean — neither more nor less.’
’The question is,’ said Alice, ‘whether you can make words mean so many different things.’
’The question is,’ said Humpty Dumpty, ‘which is to be master — that’s all.”


In ordinary conversations, when people debate a point and the words they are using for discussion, they often backhand this disagreement with the phrase “It’s a matter of semantics.” Well perhaps, but that seems to trivialize meaning a bit. It’s more than a choice of words. Semantics is about the meaning of things, which seems more important than a making it something like “you say Po-tat-o and I say Po-ta-to.” 



In comparison to marketers and political writers most layman seem to understand little about semantics, but it is a deeply refined field over the last 120 years or so. Semantics is often contrasted with syntax in language. Syntax gets at the structure of language expressions, while semantics has a focus on the relations between a class of signifiers, such as words, phrases, signs, and symbols, and what they stand for or denote. The meaning of meaning is a lot to think about so one needs a bit of model to help.



A core of what is written above and what we should know about semantics, readily applied  at least simple, single word semantics, is represented by what is called a triangle of reference shown below. BTW. it is also known, as noted in Wikipedia, as the triangle of meaning, and the semantic triangle. A version of the triangle, shown below, was published in The Meaning of Meaning (1923) by Ogden and Richards, but a more sophisticated version was penned earlier by the great American, pragmatic philosopher Charles Peirce, but also by French linguist Sassaure.

Ok, so what does this model assert? It’s a model of how linguistic symbols are related to the objects (or whatever) they represent. These are the 3 corners of the triangle and each relates to the other. I represent concept parts in sq. brackets [] and word parts in quotes ‘’ “ to make clear which vertex part is being discussed. 

Let’s start with the referent, which in my example is some thing we recognize as a [tree]. This is to say that we have a concept [] or category for the referent part of reality. A concept should be adequate and fit reality. That's part of a meaning, but only part. And we have a word for that concept “tree.” That's another part. If we say “tree” for that concept , we are correct. If we say “bush”, well close, but not correct. "Bush" shows we've used the wrong category.  That concept is not adequate for the isolated part of reality that makes up the referent - tree in reality.
A clearer example is when, say a child, has an initial concept for the category of thing that we see in the world –[bird]. If we categorize them by a flying attribute a child can quickly realize, or be told, that a flying thing I/we call a “plane” would then be classified as  [bird]. Not true, or at least we can do better and not be so confused as to not mean what we say. So the child should align the concept and referent part and not use the word "bird" for it.

And what about an ostrich? Not all birds fly. 
So I might make several types of mistakes in relation to a concept for a part of reality and its name. The correct, scientific alignment is that birds have feathers. That's a much better, consistent model for bird that aligns with we see when we test the world systematically with observations. Here we have the very important point that there should be a relation between our words and things.  The connection is through our thoughts but they are fallible. 

And even here the semantic game is difficult, since what we call "dinosaurs" seem to have had feathers. We have different names for things along an evolutionary path as our knowledge or reality advances.

So “it’s a matter of semantics" turns out to cover a pretty complicated, cognitive and scientific situation. Meaning is not in the referent of something like a tree. It’s isn’t just in the concept we have or in the words we use. When we are expressing ourselves with language the meaning is in the relation of 3 parts and is constructed.

There’s the rub. Or at least a starting part of a much deeper, but interesting story of semantics.

You can also see another challenge in the diagram in that the part of reality that I've tried to isolate and call "tree" is embedded in a larger context.  I could call it "tree in a yard" or "tree near a cemetery." What  I call it depends on the concepts evoked and this my meaning.  Or digging deeper into what Cognitive Science has revealed, the mental models we have that organize our understanding of the world. Another person hearing the phrase "tree in a yard" may have a similar concept and comprehend it in a similar way, but it is in this complicate conceptual processing that meaning comes about. Thinking fast with associative memory as discussed in my earlier blogs plays a big part in what meaning we come to.

When we are communicating we may share a common vocabulary, but differ on how we hook them up to our concepts. What do we mean by a term like “god”, “pro-life”, "legitimate rape" or “American”? Hooking "American" and "Exceptionalism" together is not just a syntactic affair and not something that has one, consistent meaning as discussed in earlier blogs

These are all pretty complex affairs and often used with different referents.  Using different words can evoke many different concepts in a cognitive agent.  The referents aren’t clear and the concepts are varied even in normal life let along in the wolf whistle world of current politics . Misunderstandings fly by and sometimes intentionally so as meanings get bent to a Humpty Dumpty purpose.  

Oh, and by the way, that is a third part of language - the pragmatics of what we say which is driven by what we are trying to communicate or not communicate.  We may just be trying to confuse someone's concepts and take over the usual meaning of words.  Over time words get weighed down by various associations. "Indian" used to mean something a bit evil, but the concept has evolved as we have gotten away from the Indian Wars.

This side of Socrates many don’t seem to make an effort to connect the parts and communicate. Next time you hear that it is “just a matter of semantics” you might have some concepts and new words to communicate back to the speaker.

                                          Image Credits
Tree: http://fineartamerica.com/featured/tree-in-grave-yard-danny-jones.html
God as a concept: http://www.patheos.com/blogs/christianpiatt/2012/01/atheism-a-null-hypothesis-on-god/

1 comment:

Edd.Doerr said...

Excellent column, Gary. My next Free Inquiry column has a similar comment on "Labels and Bottles".