It is both exciting and intimidating to engage in this dialogue with Ed Jackson and Karla Henderson. Their expertise in the area of leisure constraints is widely acknowledged and their reviews evoke critical questions about our article. Since the main point of this exchange is to raise issues for reflection, we will respond only in general terms to some basic points in their commentaries. Jackson noted some ambivalence in our article, pointing out that we interpret our data as evidence of leisure constraints while also rejecting that interpretation as being too limited. This ambivalence is intentional and reflects our attempt within this paper to capture the processes by which we came to these understandings. Part I data analysis was initially written and submitted as a conference presentation that offered support for leisure constraints. During the four months between acceptance of that paper and the conference itself, our own thinking evolved into a radically different under standing of these data. In fact, when the proceedings were released we did not recognize our own abstract that we had written four months earlier! In this paper we tried to parallel the unfolding nature of that understanding. The ambivalence that Jackson noted is real, in part because the process took us from one understanding to another as we worked our way through the data. But also, by intentionally showing how our data support the leisure constraints framework we hoped to illustrate the vagaries of theoretical interpretation. Our intent was to raise questions rather than to offer answers. Both Henderson and Jackson criticize us for relying too heavily on an older model of leisure constraints and not effectively representing the more recent literature which conceptualizes constraints much more broadly. That may be a major fault in our paper, which was initially drafted in 1992 when the Crawford, Jackson and Godbey (1991) model was still current. However, we feel our discussion would be substantively the same even if it were premised upon the more recent literature. Contemporary discussions of leisure constraints still encourage us to understand people's leisure by viewing it as a process of encountering and negotiating threats and constraints. Whether we used an older model or the more recent literature, we would still reach the conclusion that this perspective seems limited for understanding the breadth of factors that shape people's leisure choices. We have a serious concern about how the original discussions of leisure constraints have evolved to encompass too much. Jackson himself refers to this as the Pacman problem, suggesting that the constraints model eats and incorporates everything that gets in its path. It is ironic, then, that Henderson and Jackson accuse us of using a narrow interpretation of leisure constraints that focuses only on participation. In our view, the broader conceptualizations that they prefer (relating constraints to factors that shape preference or experience) would simply magnify our concern. It was troubling enough when leisure in the presence of constraints was viewed as evidence of successful constraint negotiation; now it appears that our leisure preferences and experiences are, themselves, the consequence of constraints or constraint negotiation. The all-encompassing and irrefutable nature of this claim illustrates how the constraints framework has evolved into a pervasive system of beliefs that shape and filter the way that many researchers see every aspect of leisure. …