A new method for determining whether nonverbal organisms such as human infants have consciousness has recently emerged. According to this so-called marker methodology, we do not use theories of consciousness to assess whether creatures are conscious by meeting the conditions set out by those theories. Rather, we identify a range of markers that putatively correlate with consciousness to make educated guesses about whether and when consciousness is present. Despite drawing much academic and mainstream attention, this methodology is questionable; here, we raise a new problem for it. In short, we argue that, despite appearances, the method is sui generis and unlike other investigations within the sciences. When other empirical studies into phenomena use markers, those markers are determined either by theory or by our commonsense conception of the target phenomenon. The markers posited by the marker method in consciousness studies, however, are neither. Consequently, we argue that this marker method ought to be abandoned—and we sketch how we might instead use commonsense markers to explore infant and other forms of consciousness.