The first issue is about who speaks and with what authority. There is an insight that is crucial to research about social justice that has to do with the privileged perspective of the oppressed. A woman is better positioned to identify and interpret sexism than a man, and a person of colour is more likely to understand racism than is a white person. Men, for instance, may fail to be aware of sexist practices simply because they do not run up against them in daily life. There are two insights here, one about the nature of understanding and the other about the nature of persons.

The first, the epistemological point, has been developed by Marxist theorists as well as by feminists, although it also has important sources in recent philosophy of science. The general idea is that when we understand the world, we rely upon expectations and background beliefs. We understand the world from a particular perspective and if we do not possess certain expectations and concepts, we can look at phenomena in the world and just not see them. Politically, this point implies that if I am privileged in some ways, I do not see the complexities of injustice. I do not see what it means in the lives of those affected, even if I care to. Only someone whose life expectations are different, because she is relevantly different, will be able to identify such practices as discriminatory ones. The specific epistemological point, sometimes referred to as the importance of the "view from below" (Haraway) is that if a society is systemically discriminatory, we may not see some people as existing as persons at all - the problem of invisibility. Anthropologist Levi Strauss wrote in his report "The whole village left, leaving us alone with the women and children in the abandoned houses". Evidently Levi Strauss did not expect the women or children to be people. Yet if someone exists from that invisible position, she can encounter, and identify the error in, the expectation. Thus, Celie in The Color Purple, after she's told that she's nothing because she's poor, black and a woman retorts "I'm pore, I'm black, I'm a woman .... But I'm here".

The second insight has to do with the nature of persons, more specifically, of the selfunderstanding of persons, and is a more deeply seated intuition of liberal philosophy. The idea is that the individual herself has direct access to her thoughts and feelings. I am the one who knows what I experience and I am best able to interpret my experiences. This idea is fundamental to a number of liberal philosophical conceptions of autonomy, respect for rights and of the relation between individuals and the state. For instance, it is taken to be uncontroversial by some that "every life goes better lived from the inside, with true beliefs" and that "no life goes better lived from the outside, according to beliefs that are not endorsed" (Kymlicka). The intuition about the primacy of individuals' perspectives on their own experience is important. People who are living a certain social reality are usually better positioned to identify that reality, and to make decisions about it. It is true of course that, as an individual, I have a certain stake in defining my own path, and in doing so on the basis of my own interpretations, even if I am wrong.

The important insight, then, is the primacy of the individual perspective, in particular, recognition of the privileged perspective of the oppressed. We should, without doubt, acknowledge the importance of these insights. The question is about what follows from these intuitions for researchers investigating the situations of people whose lives and cultures are relevantly different from our own. There is an assumption that, as a privileged researcher, I cannot be justified in criticizing the evaluative perspective of the less privileged person, especially when I do not share her culture. If someone says that a traditional practice is not damaging to her, who am I to dispute that? Thus, it is sometimes assumed that because people possess a privileged perspective on their own experiences, and on those of their own group, there are no good grounds for respectfully criticizing that perspective from outside the group. And it is assumed that if I, as an investigator, suggest, or judge, that someone is mistaken about her experiences, I am patronizing that person and perhaps even robbing her of her autonomy. When the rhetorical response, "Who's to say?" is offered, the suggested implication is that one ought not to say, that one ought not to take authority in such a case. The investigator, correctly aware of the power of her position as investigator, may assume that in order to be acting respectfully toward the person whose situation is of interest, she should not make evaluative judgments about what is being expressed.

There are three reasons for thinking that it does not follow from insights about the privileged perspective of the oppressed and the primacy of the individual perspective that there are not good reasons for thinking that someone can be wrong about the nature and evaluation of their own personal experiences. The first reason is that such a response, by the investigator, is much too easy. It absolves us of making judgments about almost any issue or situation. If we cannot be justified in criticizing a situation in which we do not participate, we absolve ourselves of any responsibility toward acquiring proper understanding of that situation. We are only responsible for that which directly affects us personally. Such a conclusion is counterintuitive. For one thing, if we accept it, we should give up the commitment that motivated the insight about the privileged perspective of the oppressed in the first place. Namely, that it matters that we understand the experience of those whose experiences, because of difference, are invisible to us, moreover, that it matters to us personally.

A second reason for thinking the conclusion does not follow is that people's expectations for themselves can become diminished as a result of the social expectations informing their lives. People can come to accept injustices for themselves if those injustices are considered normal. If it follows from the fact that you experience your life as happy and fulfilling, that your life is in fact happy and fulfilling, then I can justify oppressing you as long as I can be successful in beating you down sufficiently and coercing you psychologically so that you think you are happy and fulfilled. If it is true that as long as you think you are doing well, you really are doing well, then it would follow that to the extent that a state is successful in indoctrinating its people about the benefits of oppressive, repressive policies, it can claim with good reason to be acting in the interests of its people.

A third reason for resisting the conclusion that the privileged researcher should remain critically removed, and listen, has to do with respect, and what it means and requires. When we respect people as equals, we engage with them. If I respect you, for one thing, I assume that you have at least as much intelligence and ability to reason as I do. So if you say something that sounds implausible, I assume that it is not in fact implausible and if it strikes me that way, I must be missing something. So I ask for an explanation because the implausibility, given my expectations about you, is surprising to me. If we respect people whose backgrounds and traditions we do not share, we expect their views also to make sense, to express at least as much intelligence as we ourselves possess. Thus, if the views expressed do not strike us that way, we ask questions, directly of the person or indirectly by researching other sources, expecting that something had been missed. The asking of questions, whether or not they are addressed to the individual, assumes a critical perspective. If we respect people, we engage with them and their situation on the assumption that they, as equals, can provide defense and explanation. When we do not engage critically with a view, it is sometimes because we assume that the view has no defense, that it is something less than a reasoned-out view.

There is an error involved in drawing from premises about the privileged perspective of the oppressed and the primacy of the individual, the conclusion that, as researchers, we should not take evaluative authority over information provided by interviewees. The logical error here has to do with the significance of the origins of beliefs, and the distinction between the origins of a belief and its justification. It is true that individuals possess a privileged perspective as regards their own experience. But it may not be true that this privileged perspective provides adequate justification for the specific views expressed. The philosophical error here is called the "genetic fallacy". It is the error of confusing the origins of a belief with the grounds for its justification: A belief may come from an unreliable source but this does not, by itself, mean that the belief is false. It may come from a reliable source and be false. We should acknowledge the privileged perspective of the oppressed, but that privileged perspective by itself does not mean the beliefs expressed are reliable. This is simplistic and, in some cases, irresponsible.

There are several ethical risks. One is that additional dimensions of an issue are too easily dismissed and we fail to ask appropriate questions about the grounds for positions. At the Summit of the Americas, the Mexican President made the unfortunate remark that the protestors were just well-fed white kids, suggesting that the make-up of the group of people protesting provided sufficient reason to reject the validity of the protests. Even if it were the case that the protesters were mostly white and well-off, and even if it turned out that they were mistaken in their views, further evidence and argument would be required to support the claim that they were mistaken because they were white and well-off. The Mexican President dismissed the protesters on the basis of a judgment about the relevance of differences between North and South. But in doing so, he characterized the situation in a way that obscured the role of class interests that cut across North/South divisions. Importantly he relied upon his "privileged" perspective to dismiss from consideration the substance of the arguments of the protestors.

A second risk is that we do not take responsibility for the position on the basis of which we already attribute greater authority to some views over others. To ask "who's to say?" as if we are not, and cannot be, qualified to judge the relative merits of the positions involved is somewhat ingenuous. For in resisting critical engagement, we have already made such a judgment of authority, leaving unacknowledged and unexamined the role of social expectations in making some views more understandable than others. We sometimes think that, as investigators, we should refrain from making value judgments because we possess power over the people whose situation is being investigated. But when we attribute authority to a position because of some particular feature of the person's position, we are already making an evaluative judgment that structures the investigation and its interpretation. We need to take responsibility for the judgments we make and our relationship to them. We are already involved and we already exercise power. We need to be honest about our involvement and take appropriate moral and theoretical responsibility for it.