17 November 2004


Language, again

Just came across this piece by John Humphrys in the Sunday Times. I've read him on the subject before, and it's very irritating. All the points he makes about language are absolutely correct, and so well put that I'm cheering him on.

Then come his examples. They're poor examples. This time, he takes a document from some investment bank, describing a particular team's role:

The structuring team sits within the Equity Derivatives Group. Its main roles are: i. Product innovation: define and write new payoffs with sales, traders and quants (pro-active and reactive), participate to the study of the risk management of the new payoffs. The objective is to increase sharply the amount of pro-active business . .
It is in fairly obscure jargon; the nature of the business is that different companies, or even different divisions within the same company, develop their own, different terms for the same thing. As a result, while the passage makes sense to me, working in the same field, I am not quite sure I am not misinterpreting it. "Pro-active" is a danger word, often used as a meaningless buzzword, but here it has a quite obvious and well-defined meaning.

It's certainly not well written. Participate to?. This is just the sort of thing he has been talking about, but here the writer gets away with it, in that it doesn't obscure his meaning. Next time he uses the wrong preposition he might not be so lucky. There are criticisms of style that could be made too: the last (incompletely quoted) sentence is ugly. But as the whole passage is likely to be obscure to people who are not familiar with the details of the derivatives business, it is impossible for Humphrys to separate the essential obscurity from the avoidable errors. He just calls it "meaningless drivel".

Private Eye's "Pseuds' Corner" makes similar false steps. In amongst the execrable mounds of advertising, management and political mumbo-jumbo, there are odd extracts from technical journals covering particular specialties. To me, many of them might as well be Swahili, and some of them are as clear as the speaking clock.

Humphrys is conflating several issues. One is the general sloppiness of language, which the bulk of his article describes excellently. It is a major problem, as it leads to ambiguities, and, in some cases deliberately, to statements which appear to mean something, but on closer inspection state no actual identifiable facts, but convey only a vague impression or emotion. Another is that jargon, which is necessary, but in some cases seems to be intentionally less accessible than it could be, in order to exclude outsiders from a clique. A third is that when those within a particular jargon-using culture need to communicate to outsiders, they frequently are unable to do so effectively because they have forgotten the non-jargon terms they need. A fourth is that most of us, unlike Mr Humphrys, are not professional communicators. We have other work to do, and writing down a description of our work, when requested, is often an unwelcome chore to be dispensed with with as little time and attention as possible. A fifth is the desire of some to show off by indulging in some fashionable cliche or phrase which they don't really understand, in preference to the perfectly simple and ordinary way of saying what they want. (That is probably to blame for the "sits within" of the example).

I am not sure even that the claim that jargon is deliberately intended to obscure and exclude is true. It often seems that way because specialists completely exclude the normal words for their subjects from their vocabulary. You will never hear a computer specialist talk about a "computer", or a telephone engineer talk about an "exchange". That, however, has a reasonable explanation. The common-usage words are just not precise enough for technical use: does a "computer" include the monitor? Is an "exchange" all the switches on a site, or is each one an "exchange"? It is clearer for the specialist to avoid these words.

Humphrys claims that his example could not be understood even by those it was intended for. Frankly, I don't believe him. It is more likely that, in an example of another modern bad habit, it has been sent to far too long an email list, including a bunch of people who have no reason to know what the role of the equity derivatives "structure team" is.

When selecting targets to attack for bad English, force must be concentrated on those who ought to be making themselves clear to a wide audience: journalists, politicians, teachers and people selling products.

Here is a classic, from Oracle, highlighted by Bruce Schneier, when the product they advertised widely as "unbreakable" turned out to be full of security holes:
Oracle's security chief, Mary Ann Davidson, claims that the campaign "speaks to" fourteen independent security evaluations that Oracle's database server passed.

What does "speaks to" mean? In English, it means that the security evaluations were the target of the campaign, which is obviously rubbish.

What is actually means in the context is "is a lie inspired by", which is not what Ms Davidson wants to be caught saying in so many words.

Instead, a meaningless phrase has been used in attempt to hide the gap between the facts at the base of the campaign (some researchers had looked for security holes and not found any), and the claims made by the campaign (there were definitely no security holes). To anyone at all familiar with software, that's a yawning great gap, but the spokesperson attempts to bridge it by saying that the claims "spoke to" the facts, which doesn't mean anything at all and leaves the gap unbridged.

There is another modern curse and obstacle to communication: the spell-checker, but that's off the point and I'll rant about that on another occassion.

No comments: