But first, let’s look at a simple analogy – and remember, this is purely illustrative. It demonstrates just how a simple ‘fact’ and the value of that ‘fact’ can be eroded or undermined by a set of assumptions or methodological decisions to the point where it ceases to be a fact at all.
In a speaker review, one of the standard pieces of information provided is how much space the product needs, or how close it can be placed to room boundaries. Traditionally, reviewers will give a distance to the rear wall, stated in centimetres, inches or even feet. It’s a discrete value and, on the surface, what could be simpler? Well, for starters, its value depends on the accuracy and discrimination of the units of measurement. Inches are two and half times the size of centimetres. Therefore, unless the writer using inches gives a measurement that involves fractions, that information is almost certainly less precise than the metric alternative. A measurement given in feet? As a guide to placement – only meaningful in the loosest possible sense.
Now let’s look at the tape itself. Is it rigid? Is it accurately graduated? Does it have a fixed length and zero point? How accurately is it placed between speaker and wall? How accurately is the reading taken? An awful lot of expanding steel rules have a right-angle plate on the end that stops the tape disappearing inside the housing but is also designed to slide, compensating for whether the tape is hooked over a surface or butted up against one. Just how accurately do you think that functions? Then there’s the question of what you are actually measuring? The distance from the rear of the speaker to the wall will vary, depending on where on the speaker you measure from (inner or outer edge, or somewhere in between), whether the speaker is toed in and whether the tape is parallel to the side-wall and floor. If the speaker slopes or is raked, then the distance top and bottom will differ too. If the speaker is on spikes, what are you actually measuring to? The rear of the cabinet (which is raised off the ground and thus angles the tape) or do you leave the tape on the ground and essentially guess the exact distance to the cabinet itself, an approach prone to parallax affects?
“Without facts, the world is just your opinion.” Bernard Beckett
These are all error mechanisms that apply directly to an apparently simple measurement – and ultimately affect its accuracy. Now, you might say that it’s only an approximation anyway, but what if it wasn’t? Given the degree of error that impacts such a simple exercise, what error mechanisms potentially apply to something as complex as a review? And we haven’t even got to the unknown variables yet. The rear wall distance is absolutely going to be a function of not just room size but the room’s acoustic behaviour. And if you want to get really esoteric, how about considering the implications of an elastic ruler? I know that steel rules can be considered mechanically stable, but what if your means of ‘measurement’ isn’t stable? In a theoretical model that’s a very real consideration. What if your ‘tool’ varies with environmental or operational conditions? What if it depends on human agency (and let’s be honest, what review doesn’t)? After all, give six people a steel rule and ask them to measure the distance between six different sets of objects and you are going to get 36 different answers…
