Randolph Quirk Well, the Survey is in part a collection of data. Now this makes it sound very dull – to contemplate filing cabinets filled with slips registering the repertoire of the educated, mature native speaker. But that is, of course, only part of what constitutes our data in the Survey. The Survey has always from the outset had two deliberate and quite specific aims: first, to collect samples of continuous usage of that mature, educated native speaker, and, secondly, to devise psycholinguistic techniques that would elicit from a sample of such adults constructions and usages which would only by a sort of happy accident come to be enshrined in the texts, whether spoken or written, that they actually recorded.