Layers of Silence, Arenas of Voice: The Ecology of Visible and Invisible Work
Star & Strauss (1999) · Good and bad invisibility
Star and Strauss argue that no work is inherently either visible or invisible. Work is always seen through a selection of indicators; the indicators change with context, and that context becomes a negotiation about the relationship between visible and invisible work. What will count as work, they write, depends on the definition of the situation, and they take from Toni Morrison's Beloved the line that definitions belong to the definers, not the defined.
The paper is a theoretical essay for computer supported cooperative work (CSCW). They examine processes of invisibility: creating a nonperson, where the work is seen but the worker is not; disembedding background work, where the workers are visible but their work is relegated to a background of expectation; going backstage, where preparation is hidden behind performance; and abstracting indicators away from the work setting.
For the first they lean on Judith Rollins, an African-American sociologist who passed as a maid: employers usually oversaw the work, sometimes to the point of micromanagement, while the workers were socially invisible to them; in extreme cases the micromanagement became surveillance, once a tape recorder in a respondent's bedroom, and a passage they quote from Rollins describes being related to 'as if I wouldn't hear'. For the second they draw on Bowker, Timmermans and Star's work on the Iowa Nursing Intervention Classification, more than 300 interventions, one primary motive for which is to make nurses' embedded work visible to the medical record. More visibility may mean more surveillance and paperwork, they note, citing Wagner, and nurses struggle to be visible while holding areas of ambiguity and discretion. Of indicators they write, citing Markussen, that when the relationship is 'solely traded in abstract indicators', 'both silence and suffering result'.
They are not recommending more visibility in any simple sense. They find 'good invisibility' and 'bad invisibility', often traced to discretion, autonomy and power over one's resources; 'illuminating one corner may throw another into darkness', and every gain in granularity of description may bring increased risk of surveillance. Where serious inequities and divergent frames of reference exist, they add, artefacts meant to decrease the complexity of articulation work will displace it rather than reduce it, and if a system does not account for the matrix of visible and invisible work, 'those at the bottom will suffer'.
It offers no data of its own: the cases come from published studies, including Star's own with Bowker and Timmermans on nursing, some chosen as extremes to underscore the range of indicators of 'real work'. The paper does not name psychological safety; the bridge is ours. We read its arena of voice as the struggle to have one's work recognised, and its warning about surveillance as a caution that watching a team's work more closely is not the same as hearing from it.
Save this paper to a bibliography, keep notes on it, and export citations. Create a free account →
