Halliday & Matthiessen (1999: 547):
We have not here explicitly foregrounded the concept of indeterminacy in language. But there is a reason for this. To foreground indeterminacy is to treat it as something special, as a marked feature that stands out from, and hence distorts and destabilises, the phenomenon under scrutiny. Here however we take indeterminacy for granted, as a normal and necessary feature of an evolved and functioning semiotic system. Rather than being something that needs to be especially remarked on, it is something that should be built in to our ways of representing and interpreting language: part of the background, rather than the foreground, to our account of the construal of experience.