ABSTRACT
CLI has often been treated as a you-know-it-when-you-see-it phenomenon ( Jarvis, 2000a). Most researchers probably do know it when they see it, but do they always see it? What happens when CLI is so subtle or so obscured by other factors that it cannot be detected simply by looking at the data, no matter how carefully? Conversely, what happens when the researcher sees something that looks like CLI but really is not? And, in cases where CLI combines with the effects of other factors, how can CLI be teased apart from those other factors? That is, how can the specific effects of CLI be isolated, identified, and measured? Clearly, the all-too-common practice of assuming the liberty to label as transfer any or only the language use data that the researcher subjectively deems as such, is inadequate. What is needed is a more principled approach to identifying and measuring CLI.
