Situated Knowledges: The Science Question in Feminism and the Privilege of Partial Perspective

2022

The article presents a critical analysis of the theories and practices of scientific objectivity and outlines a promising feminist concept of objectivity. The author begins by criticizing two mainstream approaches to a feminist evaluation of scientific objectivity: a social constructionism combining the techniques of semiology and deconstruction, and a feminist empiricism. The former insists on the rhetorical nature of truth existing in the power field of a textualized world, while the latter legitimizes scientific objectivity adjusted for the results of gender analysis to yield a “successor science.” Rehabilitating the metaphor of vision resolves the dichotomy between the historical contingency of knowledge claims and subjects and the trust in scientific explanations of the “real” world. Any vision is embodied and therefore presupposes a location in the world. Consequently, the objectivity sought is inevitably embodied, and knowledge is situated and local, always associated with some place or position. Haraway subscribes to the constructivist critique of disembodied scientific objectivity and adds that it is an example of the “god trick” in which objectivity is a view “from above” or “from nowhere,” from outside the field of particular positions so that the scientist is distanced from the object of research. The point of view — of a human, any living being, or a machine — is an allegory for feminist objectivity. But in contrast to positioning approaches, the objectivity promised does not consist in identification with another position, but rather in a partial connection: to see together with the other, without claiming to be other. Situated knowledge from a partial perspective does not monopolize the truth by ontologizing subjugation but instead opens a way to learn how others see. Thus, only a partial perspective combined with others of the same kind can guarantee objectivity and accountability. However, the shared conversation on which rational knowledge is based should not be limited only to human beings. Any object of knowledge, Haraway argues, even when mediated by technologies for visualization, has agency, is active and is a generative node for meaning. The idea of material-semiotic actors whose boundaries are materialized in social interactions is introduced to conceptualize this kind of agency.

Citations in the corpus

• AI and the Everything in the Whole Wide World Benchmark (2021, background): “All datasets come with an embedded perspective — there is no neutral or universal dataset [Haraway, 1988, Stitzlein, 2004, Gebru, 2020, Denton et al., 2021, Scheuerman et al., 2021].”
• Data Governance in the Age of Large-Scale Data-Driven Language Technology (2022, background): “Meanwhile, the participants of our research project remain embodied in their own subjectivities [68], which necessarily represent but a small portion of these contexts.”
• On the genealogy of machine learning datasets: A critical history of ImageNet (2021, background): “In this sense, we understand ImageNet, and the myriad of image datasets that have followed in its stead, as a technical instantiation of Haraway 1988 ’ s “ god trick, ” the view that “ sees everything from nowhere ” (1988, 581).”
• Artsheets for Art Datasets (2021, unknown): “We build on critiques of universal views of data collection and analysis derived from technoscience scholars and critical race theory practitioners [55, 56].”
• Data in New Delhi's predictive policing system (2020, background): “Operationalising findings from previous work by Seaver [23] and Haraway [8], we analyse daily activities of human, institutional and societal actors, and situate our analysis of the system within its local context, drawing from on-the-ground observation and interviews.”
• Easily Accessible Text-to-Image Generation Amplifies Demographic Stereotypes at Large Scale (2022, background): “hiding a specific perspective and set of assumptions under the guise of neutrality, has been long-studied and criticized by sociologists for contributing to the exclusion and ostracizing of those who do not belong to the default group (Rogowska-Stangret, 2018; Haraway, 2020).”
• Epistemic values in feature importance methods: Lessons from feminist epistemology (2021, unknown): “constructed by social processes, we can also see how the boundaries that define those objects “materialize in social interaction” [31].”
• Algorithmic realism: expanding the boundaries of algorithmic thought (2020, background): “reinforcing their status as the only ones entitled to legitimate claims of neutrality [25, 30, 62, 63].”
• Data Science as Political Action: Grounding Data Science in a Politics of Justice (2018, unknown): “Although scientific standards of objectivity account for certain kinds of individual subjectivity, they are too narrowly construed: “methods for maximizing objectivism have no way of detecting values, interests, discursive resources, and ways of organizing the production of knowledge that first cons”
• Entanglement HCI The Next Wave? (2019, methodology): “Guided by feminist philosophy of science, third wave HCI embraced a standpoint epistemology that recognised the centrality of context and a pluralism in knowledge claims—producing situated knowledge [Haraway 1988].”
• Artificial intelligence and illusions of understanding in scientific research (2024, unknown): “Scientists have different standpoints (Box 1) that influence which questions they choose to pursue, how they ask these questions, what they take to be acceptable answers and how they frame the broader implications of those answers for future research 149–151 .”
• ORES (2019, unknown): “Our take on decoupling draws on feminist standpoint epistemology, which often critiques a single monolithic “God’s eye view” or “view from nowhere” and emphasizes accounts from multiple positionalities [53, 54].”
• Toward situated interventions for algorithmic equity: lessons from the field (2020, methodology): “We invoke this use of the term situated from feminist philosopher of science Donna Haraway [27], who describes situated knowledge as the product of embodied vantage points, located historically (see also [28, 29]).”