According to the authors, a statistically coherent view on confounding is presented. Confounding by a covariate C in the presence of data on C is distinguished from confounding in the absence of data on C. A covariate C is defined to be a nonconfounder in the absence of data on C if the parameter of interest can be asymptotically unbiasedly estimated absent data on C. When data on C are available, C is defined to be a nonconfounder for a parameter of interest if inference on that parameter does not depend on data through C.
A related, and by the present authors often quoted paper is O. S. Miettinen and E. F. Cook, Confounding: essence and detection. Am. J. Epidemiol. 114, 593-603 (1981).