Guidelines for Submissions

Newsletter Editors

Navigation
    Newsletters Index (06:1)
    apaOnline Home Page

APA Newsletters

Fall 2006
Volume 06, Number 1

Newsletter on Feminism and Philosophy

Articles

Previous Article | Index | Next Article

Activist Research and the Objectivity of Science1

Sharon Crasnow
Riverside Community College

1. Introduction

Political activism is often seen as antithetical to the objectivity of science. This is the case even for the social sciences, despite their role in shaping the social world. Although engagement might improve science by increasing its relevance to the needs of the communities under investigation, this usefulness is usually thought to be confined either to the context of discovery or applied knowledge. It is not immediately obvious how activist commitments could be evidentially relevant. To the contrary, it is frequently argued that such commitments impede gathering, interpreting, and evaluating evidence and thus compromise objectivity. As a consequence, activist social scientists face professional criticism and experience conflict between their training and their political commitments.

Traditional conceptions of objectivity identify it as a requirement for good science and locate it in scientific methodology. Methodologies that promote objectivity are impartial, neutral, and autonomous; both the research and the researcher are required to have these characteristics.2 In addition, it is typically thought that an objective methodology will result in an account of the objects and their relationships to each other that is independent of the researchers. I will refer to these two aspects of objectivity as the justificatory and ontological aspects of objectivity.

Activist research challenges at least the first of these and quite possibly the second as well. To do research as an activist is to adopt an explicitly value-laden methodology, and so to challenge the norms of impartiality, neutrality, and autonomy. I sketch an alternative account of objectivity that would not automatically rule out such value-laden activist research as good science. In order to do so, I use resources from a model-theoretic account of science and standpoint theory.

2. Model-based objectivity

In order to explicate "model-based objectivity," I will be using an understanding of "model" akin to Nancy Cartwright’s.3 According to Cartwright, theories do not represent the world directly and models do not constitute theories. "There are not theories, on the one hand, that represent and phenomena, on the other hand, that get represented (though perhaps only more or less accurately). Rather…models mediate between theory and the world."4 Models might be physical, scale models, mathematical, conceptual, representations, analogies, drawings, or even narratives. The most important aspect of a model is that it provides a means for our interaction with the world in order to achieve a particular goal. Since our goals are diverse, it is not surprising that there is diversity of models. Ronald Giere has compared models to maps in order to make this clearer.5 There are many sorts of maps, topological, road maps, trail maps, and others. Each is appropriate for different purposes and yet each accurately captures some key features of the natural world.

How can this model-theoretic approach offer us insight into the objectivity of science and how activism could contribute to that objectivity? Lorraine Daston and Peter Galison offer the following observation: "All sciences must deal with this problem of selecting and constituting ‘working objects’, as opposed to the too plentiful and too various natural objects."6 Daston and Galison suggest a modeling of what they call the "working objects." I will call these the "objects of scientific knowledge." A systematic body of knowledge, such as science, requires that we make choices about which features of the complex and multifaceted natural world we will pay attention to and which we will ignore. The results of these choices are the modeled objects of scientific knowledge.7 Precisely how models are related to the world may vary depending upon the nature of the objects in the world and the goals that we have. It is possible to build models of the objects of scientific knowledge from features of the everyday objects in a variety of ways depending on our needs and interests.
Model building depends on determining which features of everyday objects are relevant. These choices are constrained by previous choices, theory, background knowledge, and interests. We identify characteristics that we believe will allow us to answer the questions before us at any given time. These questions are expressions of our interests, and it is because of those interests that we focus on certain aspects of the world rather than others. The interests themselves are shaped by awareness of particular features of the world, which are, in turn, dependent on background knowledge. The question of whether these features are the "right" ones is a broadly empirical question. Do the models that we construct enable us to do what we want to do in the world? Are we able to intervene as we had hoped? Are we able to successfully meet our goals and address our interests? The answers to these questions constitute further empirical constraint on the objects of scientific inquiry. When models are successful, we have grasped the objects of scientific knowledge in a way that supports our interaction with the world. When models are successful in this way, then they are objective. It is this that I refer to as "model-based objectivity."

Model-based objectivity provides a way of thinking about how activist research can count as objective, both in the ontological sense and in the justificatory sense. Modeling requires choosing properties of the complex objects in the world (whether they are social or natural). But which properties are relevant depends on interests, values, and background knowledge. The properties of the objects are independent of us (hence ontological objectivity), but their existence alone does not tell us which ones should be important to us. Knowledge in aid of particular political or social goals will be shaped by those interests. The success of the model in achieving the desired goals allows us to determine the objectivity of the model in the justificatory sense.

This brief sketch does not address the vexing issues that arise as we begin to make such judgments about interests and values. In activist research, the interests of the researcher should be aligned with the interests of the community. However, the actual interests of members of a society, culture, or group studied may not be identical with the stated interests of individual members of that group. What should be valued may not be the same as what is actually valued. The question of how to identify interests is a difficult one.

One way to tackle this problem is through standpoint theory. Feminist social scientists advocate designing research projects that begin from the lives of the oppressed. In order to do this, they focus on the experiences and voices of the oppressed, as well as their social location. In addition, the collective statements of the oppressed, such as lawsuits, manifestos, and other political actions, are a means of identifying these interests. Though no one means provides the key, each helps in coming to understand the standpoint of those whom the knowledge serves.

A standpoint is not the same thing as a perspective, and the epistemic privilege that accrues from a standpoint is not automatic.8 Standpoint approaches require not only adopting the perspective of those studied but also recognizing the social and political structures of everyday life that contribute to that perspective. The objects and circumstances are not modeled from the precise perspective of those who are being studied, but the modeling begins in their lives, with their concerns, their work, and their relationships. However, the model must ultimately reveal the power structures through which their lives are shaped and the relevance of these factors to their lives.9 Standpoint theory helps reveal which properties of the world are relevant in this way and how the objects of (social) scientific knowledge should be modeled.

3. An Example: Models of mothering

Nancy Scheper-Hughes describes herself as a "militant anthropologist." She claims that her anthropology is "phenomenologically grounded…an anthropologia-pe-no-chao, an anthropology-with-one’s-feet-on-the-ground."10 Her Death Without Weeping: The Violence of Everyday Life in Brazil is an account of life in a shantytown in Northern Brazil, the Alto de Cruzeiro, Crucifix Hill, in a town that she refers to as Bom Jesus da Mata.11 The book is an account of the ways in which poverty, hunger, and infant mortality are "normalized" in the Alto. Her work is feminist in the following way: "This ethnography…is women centered, as is everyday life in the shantytown marginalized by poverty and set on edge by what I describe…as ‘nervous hunger’. Mothers and children dominate these pages even as they dominate, numerically and symbolically, Alto life."12

Scheper-Hughes began her study with the idea that motherhood is a natural and universal relationship.13 Given a model of mothering that accounts for the bond between mother and infant as natural, much of the behavior of the women of the Alto is seen as a result of the distortion of that natural relationship. Her interaction with and study of these women ultimately leads her not only to reject the model of mothering as natural but, in doing so, to revise her research questions. Ultimately she models mothering as a social/cultural phenomenon rather than a natural one.

Mother love is anything other than natural and instead represents a matrix of images, meanings, sentiments, and practices that are everywhere socially and culturally produced. In the place of the poetics of motherhood, I refer to the pragmatics of motherhood for, to paraphrase Marx, these shantytown women create their own culture, but they do not create it just as they please or under circumstances chosen by themselves. …The following discussion obviously makes no claim to universality.14

Instead of asking how the natural emotional bond between infant and mother is altered in situations of scarcity, Scheper-Hughes asks how those in a culture shaped by scarcity and poverty form the bonds between mothers and children.
If mothering is natural, the primary cause of high infant mortality would be sheer scarcity. Mothers would lose children only because they lack access to adequate food and medical care. It follows that making food and medical care available should decrease infant mortality.15 Infants who were in immediate danger can be "rescued," rehydrated, and fed. But, in practice, such "rescues" turned out to be temporary. Frequently, these same infants returned to homes where they died of dehydration or starvation at a later time, even when food and medical care were available.

Scheper-Hughes develops an alternative account that is both consistent with and supports her more radical political ideology, an ideology that is grounded in liberation theology and begins from the standpoint of the women of the Alto. Her analysis identifies a culture of "bad faith" in which none of the participants accepts full responsibility for their part in the "everyday violence" of the society, including, but not limited to, high infant mortality.16 There is an understanding in the community that many infants will die and that these deaths are necessary so that others may live. This culture of bad faith is one of the causal mechanisms through which infant mortality persists.

Scheper-Hughes had come to see the high infant mortality rate as "normal" when she first lived in the Alto in the 1960s while in the Peace Corps. She came to accept it through participating in the daily lives and struggles of the women of the Alto. When she left, it took her several years to re-establish her sense of outrage. Her subsequent training as an anthropologist made the dual vision of the insider/outsider available to her when she returned to the Alto in the 1980s. Scheper-Hughes writes that after her first year of fieldwork she was confronted by a group of Alto women who informed her that they would not cooperate with her on any future work unless she also joined them in their political struggles. "The women gave me an ultimatum: the next time I came back to the Alto I would have to ‘be’ with them—‘accompany them’ was the expression they used—in their luta, and not just ‘sit idly by’ taking field notes. ‘What is this anthropology anyway to us?’ they taunted."17 When she returned to continue her fieldwork, she returned as a campanheira anthropologist, both an insider and outsider.

Activism contributed to Scheper-Hughes’s account in a variety of ways. At the most basic level, it provides both Scheper-Hughes’s entrance into the problem (through her early Peace Corps activism and later relationship with these women) and her ability to have access to her "data," the lives of the women of the Alto, when she returns as an anthropologist. Second, her involvement in the lives of these women (through her activism) enables her to design her project so that it addresses their needs in the actual context in which they have those needs, both understanding their adaptation to the circumstances of scarcity and, at the same time, recognizing the factors that give rise to this adaptation through the dual vision of standpoint. The details of how the women see their children, how they distinguish those they expect to live from those who are just "visitors," is crucial to Scheper-Hughes’s model of mothering. There is a third way in which activism works here, and that is in the testing of the model. The model is 1) accepted by those it describes and 2) more effective in meeting their goals than a model that holds mothering as natural and so focuses entirely on addressing scarcity.

Scheper-Hughes’s account is not objective according to the standards of impartiality, neutrality, and autonomy. It is not autonomous because it is developed with a particular set of values in mind and the model is intended to serve those values. It is not neutral in that it has value consequences. Whether or not it is impartial is not as clear, however. According to Lacey, an account is impartial if it is assessed in terms of cognitive or epistemic values only. The choices that go into building a model depend on values, the ends for which the model is constructed. But these choices also must be empirically adequate. The epistemic requirements are only met in a way that already incorporates other sorts of values. Attempting to distinguish contextual, non-epistemic criteria from epistemic criteria would be misleading, though if one is clear about the specific features in particular contexts there is a sense in which one might do this. However, to argue that only epistemic criteria are being used to make judgments about which theories are better is misleading because all epistemic moves are predicated on holding particular values constant. The model is deeply contextual and it is the context itself that feeds our understanding of how to use the epistemic values in that context.

4. Conclusion

The discussion of the example above is only a suggestion for how standpoint theory could be understood in conjunction with model-based objectivity. Activism is one of the means through which standpoint can be achieved. In the case of Scheper-Hughes’s research, activism was a necessary condition for achieving standpoint because access to at least some of the evidence would have been impossible otherwise. But her activism also motivates her to seek an alternative model of mothering, and it is only in light of that model that particular phenomena become evidence. Additionally, her activism contributed to a better understanding of the role of culture in maintaining the undesirable status quo.

If we hold the model constant, we can still distinguish a context of discovery and a context of justification, but we must be clear that this distinction is conceptual only—not a distinction between two separate moments in the production of scientific knowledge. Consequently, to think of activism as confined to the context of discovery is to fail to recognize the interactive nature of the model as a tool and the role it plays in enabling the researcher to determine which phenomena are to be considered as evidence and which are not.

I have argued that a focus on determining which values, epistemic or non-epistemic/contextual, play a legitimate role in theory choice is misplaced. Values of all sorts play a role in model construction. Determining which models are good cannot be accomplished through determining whether values are used or which ones but, rather, through which models are successful at achieving legitimate goals. Worries about the legitimacy of values and hence the science shaped by them should be directed toward the scrutiny of goals.

Traditional philosophies of science base evaluation of theories on an understanding of justification that takes theories to be linguistic entities. A model-theoretic approach reframes the issue so that the success of the model is evidence for accepting it. Evidential relevance is linked to choices that determine model construction. A model that does not get at the properties of the world that are relevant to achieving goals will be less successful at achieving those goals. Engagement with and commitment to such goals may well be necessary in order to identify relevant properties. Activism can be an important avenue through which such meaningful engagement can occur.
But activism also provides a route through which the effectiveness of the model can be examined. When model-building is in aid of specific goals, the failure to achieve those goals will reveal a lack of fit between the model and the world. Working back and forth between the model and the world in order to achieve desired goals, producing knowledge in aid of those goals, not only supports activism but reveals how activism can be part of objective social science.

Endnotes

1. A version of this paper was presented at the 2006 Philosophy of Social Science Roundtable in Santa Cruz, CA. Thanks to the participants for valuable feedback, particularly Paul Roth, Joseph Rouse, Warren Schmaus, and Alison Wylie. Thanks also to Sandra Harding, Sally Scholz, and an anonymous reviewer for helpful comments.

2. H. Lacey. Is Science Value Free? Values and Scientific Understanding (New York: Routledge, 1999). There are many ways to characterize objectivity. Lacey’s terminology provides one useful way. Other approaches are provided by Heather Douglas, "The Irreducible Complexity of Objectivity," Synthese, 138:3 (2004): 453-73; Elizabeth Lloyd, "Objectivity and the Double Standard for Feminist Epistemologies," Synthese, 104, (1995) 351-81; and Marianne Janack, "Dilemmas of Objectivity," Social Epistemology, 16:3 (2002): 267-81.

3. N. Cartwright. The Dappled World (Cambridge: Cambridge University Press, 1999).

4. Ibid., 179. Other similar understandings of "model" include R. N. Giere, Science Without Laws (Chicago: University of Chicago Press, 1999); M. C. Morrison, "Models as Autonomous Agents," in Models as Mediators, edited by M. S. Morgan & M. C. Morrison (Cambridge: Cambridge University Press, 1999); and D. M. Bailer-Jones, "Models, Metaphors, and Analogies," in Blackwell Guide to Philosophy of Science, edited by P. Machamer & M. Silberstein (Oxford: Blackwell, 2002): 108-27. A model-theoretic account is not the same as the "semantic view" of theories on which models are formal or mathematical; however, the semantic view is a kind of model-theoretic account (i.e., B. C. Van Fraassen, The Scientific Image [Cambridge: Cambridge University Press, 1980]).

5. Giere 1999.

6. L. Daston, and P. L. Galison. "The Image of Objectivity," Representations, 40, Special Issue: Seeing Science (1992): 85.

7. It is with trepidation that I use "construct" here. Let me just be clear that I am not claiming that we construct the world. It is the model that is constructed, and the scientific objects are part of the model.

8. S. Harding. "A Socially Relevant Philosophy of Science? Resources from Standpoint Theory’s Controversality," Hypatia, 19:1 (2004): 25-47; A. Wylie. "Why Standpoint Matters," in The Feminist Standpoint Reader, edited by Sandra Harding (New York: Routledge, 2004): 339-51.

9. This modeling does not require that we assume that the interests of everyone within a group to be served are the same. Whether and where there are commonalities is an empirical matter. The success of the model helps us determine this.

10. N. Scheper-Hughes. Death Without Weeping: The Violence of Everyday Life in Brazil (Berkeley: University of California Press, 1992), 4.

11. The name of the town is a pseudonym; however, the shantytown (Alto do Cruziero) is referred to by its actual name.

12. Scheper-Hughes, 1992, 25.

13. Scheper-Hughes (1992) identifies this idea with Sara Ruddick’s work on mothering.

14. Scheper-Hughes, 1992, 341-42.

15. This is a simplification in order to highlight the general structure of the argument.

16. She adopts an explicitly existentialist ethics and cites Sartre as a source for "bad faith." Scheper-Hughes, 1992, 209-10.

17. Ibid., 18.

Previous Article | Index | Next Article