Tuesday, August 21, 2007

Matters of Semantics

One of my pet peeves is the saying "it's just a matter of semantics" which is used to say that something is a completely trivial matter. Fact is, semantics is anything but trivial.

The study of language can be roughly divided into three parts: syntax looks at the formal structure of the language, things like word order, where the punctuation goes, how to put the blocks together kind of stuff. Semantics examines notions of meaning, reference, and truth. Pragmatics looks at utterances in context, how do we clearly get speaker's meanings which are nowhere contained in the spoken sentence -- if someone turns to you apropos of nothing and says, "The train is leaving the station," you do not look for railroad tracks, but know to check your fly. There are very intricate, technical, and difficult questions in all of these areas, but semantics is probably the trickiest of the lot.

And not only difficult, but important. What we mean by our words is not a matter to be taken lightly. If we define our words problematically, the results can be tragic. Two stories this week to highlight exactly this point:

The first is this article from the BBC which highlights a debate over the definition of clinical depression.

As I have learned from reading Aspazia's research over the years, the way that mental illnesses in the DSM (the psychiatrist's Chilton guide) are defined is not the clean neuro-anatomical sort of process you might think at the outset. For the most part, a group of psychiatrists chosen by a larger group of psychiatrists sit in a room and negotiate (a) whether the behavior pattern is outside the norm enough, debilitating or dangerous enough to the person or others with whom he may come in contact with, or in some other way sufficiently undesirable to warrant classification as a mental illness, (b) if so, which symptoms, thought patterns, emotions, or behaviors that tend to be associated with those diagnosed with the problem are the ones that ought to be the defining characteristics and which are mere accidental regularities. These are renegotiated every few years and change with advances in understanding of the causal mechanisms, advances in treatment, cultural norms, and "informational seminars" given by drug companies.

In this debate, one psychiatrist,

Professor Gordon Parker claims the threshold for clinical depression is too low and risks treating normal emotional states as illness.

Writing in the British Medical Journal, he calls depression a "catch-all" diagnosis driven by clever marketing.

But another psychiatrist writing in the journal contradicts his views, praising the increased diagnosis of depression.
The defining characteristics of clinical depression as we now have them set out are so broad, that people with normal blues are being categorized as suffering from a mental illness.

On the other hand,
Professor Ian Hickie writes that an increased diagnosis and treatment of depression has led to a reduction in suicides and removal of the old stigma surrounding mental illness.

Under the current diagnosis guidelines, around one in five adults is thought to suffer depression during their lifetime. This costs the UK economy billions in lost productivity and treatment.
According to this line, the essential properties of depression are set out to make sure that all depressive cases are captured and this has significant social value.

Who is right? Don't know, but this question of semantics is certain more than trivial word play. Medicating those who do not need it and preventing suicide certain do not strike me as unimportant.

The second comes from a discussion over at the Maiden's blog. We use the term "killing" to distinguish certain acts from "murder." In both cases, someone takes life, but the former term sheds the moral component that is a part of the latter. A similar move has been tried by those who have introduced the term "enhanced interrogation techniques," into our political and cultural lexicon. The idea is to differentiate it from torture by removing any ethical connotation. This, then, allows us to torture, but because we have recategorized it, it no longer should give us second thoughts.

We have had two semantic assaults upon our social vocabulary in the last couple of generations. Political correctness sought to replace terms pregnant with the biases of the past. The labels we give to people come loaded with images and stereotypes that enforced an unfair distribution of power. Naively, it was thought to be "only a matter of semantics" that could be fixed by using new terms by all "colloquial English speaking-Americans."

That turned out not to be the case, but the plot was picked up by the right and political correctness became patriotic correctness and the same game was played. "Conservative" became "pro-American." "Civilian deaths" became "collateral damage." The damndest thing, though, was that where the first pc failed, the second one has been much more successful. The results, of course, are now out there in the world for all to mourn.

If we really want to change the world, we not only have to change politics and legislative and executive power holders, we need to reclaim our language. The path to a better world is "just a matter of semantics."

UPDATE:
And then there's this little semantic gem from LGM:

"The Anti Defamation League has fired its New England regional direction for insisting that the group recognize as genocide the circa-1915 slaughter of perhaps a million Armenians by the Ottoman Turks... A resolution pending in Congress would make it official U.S. policy to recognize that the Armenians were genocide victims. But the ADL, along with other leading Jewish-American groups, apparently considers friendly relations between Israel and Turkey--whose government takes genocide claims as a massive provocation--more important than the underlying historical question."


In other words, the Anti-Defamation League is trying to prevent recognition of the Armenian Genocide, an event that Adolf Hitler *explictly* used to justify the Holocaust.

Wow.
Wow, indeed.