Lies, damned lies and chart statistics
American author Mark Twain once reported British prime minister Benjamin Disraeli as saying: "There are three kinds of lies: lies, damned lies, and statistics."
It's a maxim, however apocryphal, that has power, and which drives a lot of the thinking of commentators in politics, economics and culture to this day.
It has its place in music too: how regularly do we hear statistics bandied about when it comes to the 'best-selling' or the 'fastest ever sales' and so on and so forth.
But where's the truth? And does the truth even matter?
The Bee Gees
Today the sad news came through of Robin Gibb's death. The Bee Gees singer and songwriter has rightly been eulogised for his artistic accomplishments, but in passing a figure of 200m record sales has been mentioned. Sometimes generally as 'records' but sometimes more specifically as 'albums'.
For those of us immersed in the business of music, it matters how facts and figures are reported. Call it geeky if you will, but with the continuing assault by Adele upon the UK's best seller list, it's in the news almost every week. Cower before the might of Adele as she vanquishes Dire Straits, Pink Floyd and even Michael Jackson.
The message to news consumers, to people who might even occasionally buy records, is that sales figures are inherently newsworthy. If that's the supposition, then a certain degree of accuracy should be striven for, as with all other subjects of news journalism.
The trouble is that music is inherently unreliable, muddled and muddied in the upkeep of its own history. Until the 1970s at the earliest, charts were often manipulated - the 'payola' scandal in America had record companies paying radio stations to play their works. Skulduggery and underhand tactics characterised the industry across the world. It still does in some territories.
But that's not to say that educated guesses can't be made.
In America, the world's largest music market, an album goes platinum at sales of one million, while in the UK it's 300,000. France, Turkey and Germany are 200,000 and Japan 250,000. These are the biggest music markets in the world, by far. The rest of Europe, Asia and Australasia add another significant portion of course, but individual countries have far lower thresholds than these examples.
It is plausible that the 'rest of the world' could double the sales of these main markets.
Sales figure data is easy to come by for these main markets, owing to three decades of 'point of sale' electronic scanning. So...
The Bee Gees' best-selling record is the Saturday Night Fever soundtrack. It's 15 times platinum in the US, with more than 15m sales. Its sales in the other main markets add about five million. At the upper-end this album might have sold 40m.
Their only other US number one album was 1979's Spirits Having Flown. Its total sales from the main markets is 5.5m. So let's estimate total worldwide sales, at absolute maximum of 11m. USA Today puts it at 20m, so just for the sake of argument let's take that.
This gives us 60m so far. There's another 140m to account for. Let's estimate that the other studio albums sold another 10m (they really weren't very successful). Can singles sales take care of 130m? That's far from likely.
It seems likely that reported sales figures are cobbled together from online encyclopaedias, which only require one source for stated facts. It's therefore easy to imagine figures amalgamating, expanding and getting a life their own, without any mathematical analysis being applied to them.
Why is any of this important? Well, it's part of a wider picture. The more inaccurate facts people have, the less they understand about the industry. Down the bottom of the ladder, artists have to deal with their audiences fundamentally misunderstanding their capacity to earn, to live, to make a career.
The vague notion people have that any kind of chart appearance or magazine coverage means riches and stardom is so far wide of the mark that it's almost humorous, but it's a serious issue when the reality is so far wide of the mark.
Accuracy helps to give a more complete picture of an industry in which a tiny percentage of acts make any money at all. Inflated estimates of the top acts' sales figures knock that drive for accuracy before it's even begun. Why should anything be accurate if this isn't?
Feel free to comment! If you want to have your say, on this or any other BBC blog, you will need to sign in to your BBC iD account. If you don't have a BBC iD account, you can register here - it'll allow you to contribute to a range of BBC sites and services using a single login.