The purpose of this study was to examine assignment writing as a phenomenon of academic writing. This was done through exploring the experiential accounts of members of a university writing community. Their accounts described the community's perceptions and experiences of literacy practices needed to write assignments, of how students developed these practices, and, of what constituted success in the writing. A multi-method, embedded, case-study approach was used. Quantitative data were derived from first-year, second-year, and fourth-year respondents' perceptions and experiences related to assignment writing. A cross-sectional comparison of groups showed consistent year-level effects. Fourth-year students were more confident as writers than first-year and second-year students, and had less difficulty with declarative and procedural aspects of writing assignments. These findings were replicated in a repeated-measures study using a sub-group of first-year and fourth-year students. However, when students contextualised their responses by nominating a subject and referring to their completion of its written assignment, first-year students reported less difficulty with the declarative aspects while fourth-year students were more positive in the procedural aspects. Year-level effects were found for what they reported as helpful in acquiring declarative and procedural knowledge of writing. First-year students reported a wider range of sources as helpful than fourth-year students did, with two exceptions. More of the latter had found information gained in consultations helpful in understanding an assignment question. Additionally more had found friends helpful. Second-year students generally were more positive than first-year and fourth-year students about the usefulness of information in helping them understand an assignment question and in writing it in an academic genre. Knowing how to write predicted success more strongly and consistently than any other factor. Qualitative data informed findings from the quantitative analyses by providing experiential accounts about students' perceptions of themselves as assignment writers, their experiences when writing assignments, and how these experiences developed literacy practices that contributed to success. Additionally, qualitative data were collected from lecturers who convened first-year subjects and those who convened fourth-year subjects. The qualitative data indicated students' strong reference to experiences of writing and of seeking help. Both had shaped their self-perceptions as writers and these had changed over time. First-year students believed that knowing what lecturers wanted in writing assignments was an important factor in success. They described their efforts to access this information and to give lecturers what they thought was wanted. Fourth-year students recognised the same factor, but were more self-reliant in approaching an assignment task. The change to greater internal control appeared to be an outcome of encountering inconsistent and confusing information from external sources over their four years of writing assignments. For their part, lecturers of first-year students said that successful students knew what to write and how to write it. However, lecturers of fourth-year students believed knowing what to write should be subsumed by knowing how to write, and concentrated on the procedural aspect. They believed a coherent assignment resulted when students conceptualised subject matter in ways that enabled them to write academically. Findings in this study extend recent reconceptualisations of literacy as 'literacies' and socio-cultural, socio-cognitive theories about literacy as social practice. They demonstrate limitations of an apprenticeship model for acculturation and suggest a more agentic role for novice members in accounting for learning outcomes as students develop as assignment writers. The experiential accounts reported by members of the academic writing community described their shared and idiosyncratic perceptions of literacy practices and relations of these practices with success in assignment writing. Their descriptions enhance our understanding of the complexity and consequences of these experiences. They also account for why calls for the community to be more visible and explicit in sharing communal expectations of what is privileged and valued in academic assignment writing generally may not be a solution. Based on findings here, this is not a solution. Expectations need to be co-constructed within the community, among students, and lecturers within the context of the writing task. An outcome of understandings reported here is the development of a model from which factors, conditions and critical events that situate learning within a rhetorical conundrum may be described and predicted. This model offers a framework for members of a writing community to explicate individual experiences and expectations in ways that help everyone make sense of those critical events that contribute to a rhetorical conundrum and shape encultured knowledge.