We all presumably have some idea of a shadow. For the seeing of shadows is part of ordinary experience, and talking about shadows is part of ordinary language. In this paper we propose to investigate our common idea of a shadow. We will first present a preliminary account of shadows which is meant to be plausible. We trust you will agree in finding it so. Then in terms of a special imaginary case we will throw this preliminary version into doubt, leading to the suspicion that our common operative idea of shadows is confused or inconsistent. Afterwards, however, we will attempt to resolve this doubt by showing that closer attention to both the ordinary phenomenon and the ordinary language of shadows reveals our common idea of a shadow to be clear and consistent after all, but more subtle than at first appears. Finally, we will characterize in a general way the phenomenologico-linguistic method of analysis which seems to be successful in handling our particular problem and therefore likely to be fruitful in similar cases.