Visual crowding is the disruptive effect of “clutter” on our ability to recognise objects, despite their being identifiable in isolation. This effect is particularly strong in our peripheral vision, making crowding the fundamental limit on over 95% of the visual field. Understanding crowding will thus reveal much about how we recognise objects, as well as how to present information to those who rely on peripheral vision (e.g. macular degeneration) and those with elevated crowding (e.g. amblyopia).The dominant view of crowding is as an excessive integration of object features that simplifies the crowded region into efficiently processed texture. A key assumption here is that the position of objects is lost – otherwise, the true features would be seen in their true locations. This is, however, inconsistent with the high positional specificity of crowding: its magnitude relies strongly on the position of objects, both in relation to one another (more crowding for adjacent objects) and across the visual field (more crowding peripherally). In addition, while crowding affects object identities, it rarely fills the spaces in between. Simple feature integration would not retain this specificity. We call this discrepancy between gross object position and the position of their features “the paradox of position” in crowding.Our psychophysical experiments and computational modeling will examine this hitherto unexplored aspect of crowding in two key ways. The first will delineate the circumstances where position encoding is reliable and where it breaks down. The second will examine the rules that govern “what goes where” by manipulating aspects of the stimuli likely to influence this process (e.g. spatial frequency and border ownership). The outcome will be an insight into both the mechanisms of crowding, a key process in vision, and the way that object identities are tied to positions, a key issue of the ‘binding problem’ in object recognition more generally.