Scientific Perspectives, Feminist Standpoints, and Non-Silly Relativism

Defences of perspectival realism are motivated, in part, by an attempt to find a middle ground between the realist intuition that science seems to tell us a true story about the world, and the Kuhnian intuition that scientific knowledge is historically and culturally situated. The first intuition pulls us towards a traditional, absolutist scientific picture, and the second towards a relativist one. Thus, perspectival realism can be seen as an attempt to secure situated knowledge without entailing epistemic relativism. A very similar motivation is behind feminist standpoint theory, a view which aims to capture the idea that knowledge is socially situated whilst retaining some kind of absolutism. Elsewhere I argue that the feminist project fails to achieve this balance; its commitment to situated knowledge unavoidably entails epistemic relativism (though of an unproblematic kind), which allows them to achieve all of their feminist goals. In this paper I will explore whether the same arguments apply to perspectival realism. And so I will be asking whether perspectival realism too is committed to an unproblematic kind of relativism, capable of achieving scientific goals; or, whether it succeeds in carving out a third view, between or beyond the relativism/absolutism dichotomy.

independent of any perspective. This sounds very similar, both in content and motivation, to feminist standpoint theory. Standpoint theory is also a view on which knowledge is situated (socially situated, due to factors like a knower's race and gender), and which (at least on some framings) has been presented as a third way between absolute scientific objectivism and relativism about scientific claims.
Elsewhere I have argued that feminist standpoint theory fails at its goal of providing a genuinely distinct 'third way'. On closer inspection, it turns out that (a certain kind of) relativism is essential to the view (Ashton forthcoming); although, unlike most anti-relativist critics of standpoint theory, I don't believe that this diminishes its usefulness in any way. In this paper I apply these arguments I've made about feminist standpoint theory to perspectival realism, to show that perspectival realism (or at least some versions of it) are akin to relativism too.
In Sect. 5.1, I introduce perspectival realism and feminist standpoint theory, and discuss the similarities between them in more detail. In Sect. 5.2, I recount my argument that feminist standpoint theory is a form of relativism. In Sect. 5.3, I apply this argument to perspectival realism, to show that it too is a form of relativism. Giere has expressed concerns about his view being mistaken for, or conflated with, what he calls "silly relativism" (2006,13)-the view that all perspectives are as good as one another, and thus 'anything goes'. So, in Sect. 5.4, I spend some time specifying what kind of relativism is present in the version of perspectival realism that Giere puts forward. I make clear that it is an unproblematic, epistemic relativism, which is not "silly" in the way that Giere warns against. 1 I conclude by considering what perspectival realism and feminist standpoint theories can learn from one another.
Before I start, I am going to run through some preliminary definitions which will be useful to keep in mind for the rest of the paper. Some might seem like very basic terms that most readers will have a handle on already, but these are precisely the terms where nuances can be lost and confusion created if special care is not taken.
First, realism and anti-realism. These are terms for metaphysical views about 'what there is'. Realists believe that there is an uncomplicatedly mind-independent world, which exists regardless of whether and what we discover about it. Antirealists believe something other than this: namely, that there is no mind-independent world, or that the world depends in some (important) way on (or is affected by) us and our investigations. One form of anti-realism that Giere talks about is social constructivism. Social constructivists believe that (at least some) objects of inquiry are determined (or 'created') by social processes, namely our investigations and our interpretations of the results of these investigations.
The other key terms that I will be using are epistemic terms; these terms apply to claims about what we believe, know, and understand. Objectivity is the property (of 1 Some people equate relativism with the silliness Giere has in mind, and to them the idea of nonsilly relativism seems like a misnomer -views which aren't silly cannot also be relativism, so if a view is shown not to be silly, it must be something other than relativism (c.f. David Bloor's discussion of "foolishness conditions" (Bloor 2011, 452)). As a precursor to my argument for non-silly (and non-derogatory) relativism, it might be helpful to point out that there are a number of serious, self-described epistemic (or scientific) relativists, including: David Bloor (1976), Lorraine Code (1991), Paul Feyerabend (2010, Martin Kusch (2002), and Richard Rorty (1991). justification or of knowledge claims) of being independent of social or individual factors. It is often associated with absolutism, which is the view that the standards for justification apply universally, regardless of time, place, culture, and so on. 2 These are both typically taken to be in opposition to epistemic relativism, which is the view that there are, or can be, different justificatory standards which apply in different times, places, cultures, and so on.
When I talk about relativism I mean epistemic relativism, and specifically a view which incorporates three components identified by Martin Kusch (2016, 34-5). 3 The first component is epistemic dependence, which says that a belief or claim has an epistemic status (as justified or unjustified) only relative to an epistemic system or practice. This can be seen as a rejection of absolutism, as defined above. Michael Williams (2007, 94) also identifies this component, though his definition only makes explicit reference to systems, not practices. The second component is plurality: there are (or have been, or could be) more than one such epistemic system or practice. Plurality is widely accepted as a necessary component of relativism (e.g. Williams 2007, 94;Coliva 2015, 140). The third component, non-neutral symmetry, says that there is no neutral (i.e., system-independent) way to evaluate or rank systems or practices (Kusch 2016, 35). It will be worth discussing this component in more detail.
Non-neutral symmetry lines up nicely with the rejection of absolutism that we've seen embodied by dependence. Recall that absolutism is the idea that justification is independent of time, place, culture, and so on. Epistemic dependence rejects this claim, on the ground that justification is dependent on (temporally, geographically, or culturally) contingent systems and practices. Non-neutrality expresses a related point: if justification is system-dependent, then the justification for any evaluation or ranking of a set of systems will be epistemically dependent too. Neutral, or system-independent, rankings are not possible.
As we'll see, discussions of relativism often involve a different version of symmetry, known as equality, or equal validity, which says that all epistemic systems and practices are equally correct (Kusch 2016, 35). At first sight this might appear to be almost indistinguishable from non-neutrality-after all, they both seem to say that it is not possible to rank different systems and practices. But on further investigation they are quite different. Neither of them rules out system-rankings altogether, and, unlike non-neutrality, equal validity is incompatible with the rejection of absolutism. Remember, non-neutrality doesn't say that no rankings at all are possible-only that system-independent rankings are not. This, as we saw above, is in line with the rejection of absolutism as embodied by dependence. Equal validity doesn't say that no rankings are possible either, because it itself is a ranking: it says that all systems and practices rank in the same position, and it seems to presume a neutral point (or nonpoint) of view from which this ranking is made. This directly conflicts with dependence and the rejection of absolutism, but fits naturally within an absolutist picture.
If a view contains the first three components I've mentioned-dependence, plurality, and non-neutral symmetry-then it is, or includes, a form of epistemic relativism. If, instead, (or in addition) it contains equal validity then it is internally inconsistent, and not a form of epistemic relativism. With these preliminaries in place, it's time to properly introduce and compare perspectival realism and feminist standpoint theory.

Comparing Perspectival Realism and Feminist Standpoint Theory
In this sect. I will outline both perspectival realism and feminist standpoint theory, and then highlight some central similarities between the two views. In doing this I will make clear why I think that the arguments I am about to make about standpoint theory could also be applied to perspectival realism. I will also point to some important differences which should be borne in mind when attempting to draw out connections between perspectival realism and feminist standpoint theory. Perspectival realism is a view proposed as a middle way between objective scientific realism and a variety of anti-realist views, encompassing relativism and constructivism (Giere 2006, 3;Massimi 2018, 164). It is realist to the extent that it allows that there can be a single way the world metaphysically is, but is perspectival in the sense that it says that our knowledge of the world-including our best scientific knowledge-is historically and culturally situated, and cannot give a single, complete, objective picture of the world. 4 Giere's comparison between his description of representational scientific models, on the one hand, and maps on the other hand, helps to make this idea clearer (2006,(72)(73)(74)(75)(76)(77)(78)(79)(80)(81). Like maps, Giere says that scientific models are partial, limited representations of a particular area or aspect of the world (2006,. In the same way that multiple, different, and even conflicting, maps can be said to accurately represent the world, multiple incompatible models, can accurately represent the world too (2006,(78)(79)(80). This means that we must create, and choose between, models (as we do with maps) based on our interests, and interpret them using existing cultural conventions (2006,. Feminist standpoint theories, which comprise one of the main branches of feminist epistemology, are best explained by making reference to three theses. 5 The situ-ated knowledge thesis, which is common to all branches of feminist epistemology, says that differences in the social situations of inquirers (including things like their race and gender) make for epistemic differences-differences in what they can justifiably believe (Ashton and McKenna forthcoming). Feminist standpoint theory offers two further theses which help to specify the situated knowledge thesis. The standpoint thesis says that justification depends on 'socially situated' perspectives. According to this idea, subjects have different 'social locations', or different statuses as socially oppressed or socially privileged. For example, black women occupy very different social locations to white men, and these different social locations come with different experiences, which have the potential to enable different epistemic perspectives. Finally, the epistemic advantage thesis says that experiencing social oppression can lead to more, or better, justification. The idea is that subjects who are socially oppressed have distinct experiences, and through critically reflecting on these can turn their perspective into a 'standpoint'-an epistemically privileged perspective from which the nature of relevant social relations is visible. Subjects who are not oppressed do not have these experiences, and as a result are less likely to achieve a standpoint. 6 Different versions of standpoint theory can be distinguished by how they flesh out the details of the epistemic advantage thesis, but all of them take this, and the standpoint thesis, as a starting point. Then it is clear that on standpoint theory, as with perspectival realism, knowledge is situated.
Standpoint theory can also be seen as a middle way, between traditional scientific objectivity and what Giere has called "silly relativism" (Giere 2006, 13). Sandra Harding (1991) has explicitly presented it in this way. 7 She rejects the traditional approach to objectivity in science, which she describes an as attempt to eradicate all values from science. She says that this is too ambitious, because some values have benefited science (Harding 1991, 144), 8 but also that these ambitions aren't successfully lived up to. When you try to ignore the effect of social factors on science, you end up allowing unconscious, deeply embedded values-including racist and sexist ones-to thrive unmonitored (1991,143). However, instead of embracing relativism as the alternative to this "weak objectivism", she instead proposes that we conduct science according to the goal of "strong objectivity" (1991,. This means acknowledging that values have a role in science-i.e., that knowledge is situatedand then paying close attention to them, so that we can be aware of and mitigate their effects. 6 Four important caveats accompany the epistemic advantage thesis. The advantage: doesn't depend on essential categories (e.g., of 'woman'); it isn't a necessary or a sufficient condition on membership of a social group; it isn't 'automatic', but earned through critical reflection; and it can be restricted in scope (Ashton and McKenna forthcoming). 7 Most standpoint theorists have disavowed relativism (e.g., Medina 2013), and their views are often interpreted as rejections of objectivity due to their commitment to situatedness. Harding is the only author I know of who has explicitly presented her view as a middle way. 8 C.f. Longino (1997) who argues that some values are necessary when choosing between multiple models or theories that fit the evidence equally well.
To sum up, standpoint theory and perspectival realism have at least two things in common. They both see knowledge as situated, by which I mean they both think that knowledge depends on necessarily partial and interested perspectives. And they have each been framed as a middle way between traditional scientific objectivism on the one hand, and 'silly', relativism and constructivism on the other. Before moving on, there are however two differences between perspectival realism and standpoint theory that I should flag up. The first is due to Kristina Rolin (2009), who argues that standpoint theories have a distinctive focus on power relations, which Giere's view of scientific models as akin to maps does not capture: The map metaphor invites us to see the subject matter of inquiry as a passive object waiting to be represented by someone else. In the map metaphor, scientific knowledge is perspectival in the sense that the mapmakers decide what to represent and to what degree of detail to represent it. In my interpretation of feminist standpoint theory, social-scientific research is perspectival not only in this sense but also in another sense. The term "perspective" refers not only to a social scientist's perspective but also to an informant's perspective, and these two perspectives may or may not meet each other in the inquiry (Rolin 2009, 223). Whilst Giere emphasises the partiality and contingency of scientific theories, Rolin says that standpoint theories go further by showing that the power relations surrounding the construction and testing of scientific theories can suppress or distort evidence. It's not just that there are multiple ways that scientists can choose to go about 'mapping', but that what they are trying to map can present itself in multiple ways-the terrain can move. The second difference between feminist standpoint theory and perspectival realism is that they come from different sub-disciplines with different sets of terminology. Perspectival realism comes from the philosophy of science, and authors here use the language of "theories", "models" and their "fit with the evidence", whilst feminist standpoint theory comes from epistemology and often makes claims about "belief" and "justification". These terminologies don't translate perfectly, and whilst I don't think this will raise particular problems in this paper, I do think it's important to flag it to avoid confusion.

Relativism in Feminist Standpoint Theory
In the previous sect. I showed that there are two key similarities between feminist standpoint theory and perspectival realism. The first is that they both involved situatedness, and so reject traditional objectivity. The second is that, despite the above, their proponents have been keen to distance themselves from epistemic relativism. Elsewhere I have argued that the anti-relativist arguments that standpoint theorists make don't work, and that feminist standpoint theory is best understood as an illustration of how epistemic relativism can be successful (Ashton, forthcoming). In this paper I'm going to argue that the same is true of perspectival realism, but first I'll briefly explain my argument that standpoint theory is relativist.
Recall that there are three components which I take to be jointly sufficient for epistemic relativism: epistemic dependence, plurality, and non-neutral symmetry.
The first two of these follow quite naturally from the standpoint thesis. Dependence follows because the standpoint thesis says that justification is dependent on a perspective. Plurality follows from it because the standpoint thesis says that there is more than one of these perspectives. So standpoint theory uncontroversially has two of the three components that are jointly sufficient for relativism. Accordingly, standpoint theorists hoping to disprove the claim that their views are relativist standardly attempt to do so by denying that they endorse the third component of epistemic relativism, symmetry. They do this by pointing to the epistemic advantage thesis, which says that the standpoints of socially oppressed groups provide epistemic benefits (Tanesini 2018). The idea is that the epistemic advantage thesis shows that some standpoints (those of the socially oppressed) are better than others, that standpoint theory allows standpoints to be ranked, and that standpoint theory is therefore not compatible with symmetry. If it isn't compatible with symmetry then it can't include all three of the components which are jointly sufficient for relativism, and so the argument that it is relativist falls apart.
But remember in the Introduction I distinguished between two versions of symmetry: non-neutral symmetry and symmetry based on equal validity. My (forthcoming) argument that feminist standpoint theory is relativist turns on this point. Or, rather, on standpoint theorist's failure to recognise this point. Standpoint theorists succeed in showing that the epistemic advantage thesis is incompatible with symmetry based on equal validity. Equal validity says that there is no way at all to rank epistemic systems and that they are all equally correct. The epistemic advantage thesis says that some standpoints are better than others, and so is incompatible with this. But this is all beside the point if, as I have claimed, non-neutrality, rather than equal validity, is what makes a view relativist. 9 Non-neutrality makes a weaker claim than equal validity; it doesn't say that there is no way at all to rank different epistemic systems, it only says that there is no neutral, system-independent way to evaluate or rank different epistemic systems. The epistemic advantage thesis isn't incompatible with this version of symmetry, and so its presence isn't sufficient to show that standpoint theory is not a form of epistemic relativism.
The epistemic advantage thesis could be made to be incompatible with nonneutrality (and so with epistemic relativism) if it were interpreted such that the ranking(s) it recommends are system-independent. If the claim that (for example) black feminist standpoints are epistemically superior to the perspectives of white men was supposed to be absolutely justified, rather than justified relative to one or more specified epistemic systems, then it would be incompatible with non-neutrality and so avoid epistemic relativism. 10 However, I don't think the option is open to standpoint theorists to interpret the epistemic advantage thesis in this way.
The central and defining component of standpoint theory, the standpoint thesis, says that justification is dependent on socially-situated perspectives. This means that the justification of a ranking of different systems will be dependent on sociallysituated perspectives too, and so standpoint theorists are committed to understanding the epistemic advantage thesis (which is effectively a ranking of different systems) as system-dependent. The epistemic advantage has to be understood as system-dependent if standpoint theorists are going to maintain a consistent view, and so standpoint theory is best understood as a form of epistemic relativism. 11

Relativism in Perspectival Realism
In the last sect. I recounted my argument that standpoint theory is a kind of relativism. I argued that, in addition to having the first two components of relativismdependence and plurality-standpoint theory can also be shown to have the more controversial, third component of relativism, namely non-neutral symmetry. Standpoint theorists can't appeal to an absolutist ranking of standpoints (because of their commitment to dependence) and they (correctly) deny that their view involves symmetry based on equal validity, so the epistemic advantage thesis is best understood in line with non-neutral symmetry. We've already seen that there are significant structural similarities between perspectival realism and standpoint theory, which may suggest that perspectival realism too could contain all three of these components. In this sect. I'll follow through on this, by arguing that Giere's perspectival realism is committed to all three components, and so is a kind of epistemic relativism.
Let's start with the first component. Dependence says that a belief has an epistemic status (as justified or not) only relative to an epistemic system or practice. Giere's perspectivism, like standpoint theory, says that knowledge is situated: scientific claims can only be made using, and evaluated against, scientific models and their associated instruments, methods and interpretive conventions. This means that, 10 Aidan McGlynn raised a similar point in a commentary on an earlier version of this paper: at least some arguments for standpoint theory (those turning on examples of scientific progress made by the introduction of a feminist perspective) seem to justify an absolutist understanding of epistemic advantage. In addition to the response I go on to make above (that this would render standpoint theory inconsistent) I also dispute that McGlynn's is the best way to understand these examples. I believe they can be explained just as well by deploying a relativist understanding of shared progress, which doesn't justify an absolutist understanding of advantage, as they can with an absolute notion of progress. I defend this claim in Ashton (forthcoming). 11 To be clear, I don't consider this to be a problematic result for standpoint theorists because, as I've argued elsewhere (Ashton forthcoming; Ashton and McKenna forthcoming), the most frequently criticised aspects of relativist feminist epistemologies are much less troubling than they are typically supposed to be. according to perspectival realism, any scientific knowledge claim we might make is necessarily understood and evaluated relative to an epistemic (scientific) practice, and so its epistemic status is dependent on that practice. Perspectival realism incorporates the dependence component. 12 The second component is plurality, which says that there are, have been, or could be multiple epistemic systems or practices. This, too, is the case according to perspectival realism, whose proponents point to various different models and practices, such as Aristotelian and Ptolemaic perspectives (models and practices), Copernican and Newtonian ones, and the contemporary perspective of General Relativity (Giere 2006, 94). Of course, these different perspectives share plenty of very general principles, such as principles outlining the most general uses of observation and testimony, but there are other, narrower, principles which are specific to particular perspectives. For example, principles of the form: "If S uses instrument i, and observes result r, whilst conditions c obtain, then S is justified in believing x." These are the kinds of principles which scientists from different perspectives disagree over, and which (in combination with more general principles, as well as certain modelling practices) their different beliefs are justified relative to. So the plurality component is also present in perspectival realism.
The final component is symmetry, and specifically non-neutral symmetry, which says that there is no neutral way to rank different epistemic practices. We can see this reflected in Giere's view even more clearly than in feminist standpoint theory. Like standpoint theorists, Giere believes in scientific progress and doesn't judge all perspectives to be on a par. This might be taken to undermine symmetry; where standpoint theorists draw on the epistemic advantage thesis, Giere can appeal to a Kuhnian picture of progress on which perspectives differ in their puzzle-solving power. 13 But, also in common with standpoint theorists, Giere is committed to situated perspectival knowledge. This means that we should expect him to reject absolutism and recognise that the scientific claims he makes, including claims he makes about the ranking of different perspectives, are perspectival too. And he does this; he explicitly rules out the possibility that perspectivism itself is absolutely justified and perspective-free, instead acknowledging that it is developed "within the framework of contemporary science", and that his "own claims must be reflexively understood as themselves perspectival" (Giere 2006, 3). This is precisely the move that I encouraged standpoint theorists to make in the previous section: to take situatedness (or perspectivality) seriously enough to apply it to one's own claims, and to recognise that those claims are situated too. In making this move, Giere accepts non-neutrality. He accepts that there's no neutral way to rank different perspectives, because any such ranking would itself be situated, and dependent on a scientific perspective. Like feminist standpoint theory, Giere's perspectival realism incorporates dependence, plurality, and non-neutral symmetry, and so is a form of epistemic relativism. However, unlike standpoint theory, perspectival realism's commitment to non-neutrality (at least in the work of Giere) is made much more explicit.

Non-Neutral and Non-silly Epistemic Relativism
I've just argued that perspectival realism, like standpoint theory, is a form of relativism. Relativism has a negative reputation and describing a view as relativist is typically seen (and often also intended) as being critical of that view. This is not my intention however. I think the form of relativism found in perspectival realism is unproblematic. In this section I will specify exactly what sort of relativism is present in perspectival realism, making reference to Massimi's account which builds on Giere's work, and-perhaps more importantly-I'll show what kind of relativism is not present in perspectival realism. The relativism present in perspectival realism is one that perspectival realists should embrace. The first specification I want to make is that perspectival realism is a form of epistemic, not metaphysical, relativism. Taking a closer look at Giere's taxonomy of different views will help to make this point more clearly.
Giere frames his perspectival realism as an alternative to existing views: scientific realism, and social constructivism. It will help to separate what each of these views have to say about metaphysics (how the world is) and epistemology (what we can legitimately say about the world). 14 Giere seems to categorise scientific objectivism as metaphysically realist (i.e., there is a single way the world is) and epistemically objectivist (i.e., there is a single correct description of that world that science strives to obtain, see 2006, 4-5). Massimi's perspectival realism takes the first (metaphysical) component whilst rejecting the second (epistemic) one. Giere then explains that social constructivism is metaphysically anti-realist (i.e., it denies there's a single way the world is), 15 and epistemically perspectival (i.e., it says there are multiple legitimate ways of describing and understanding the word). 16 Massimi's perspectival realism takes the latter (epistemic) component of this and rejects the former (metaphysical) one. The So, on this understanding, perspectival realism combines the metaphysical view that there is a single way the world is with the epistemic view that there are multiple, legitimate ways that we can view or conceptualise the world. Its perspectivism is only present in the epistemic domain, making it an epistemic relativism. This means that perspectival realism avoids any worries associated with metaphysical or alethic relativism. I take this to be Massimi's (2017) point when she considers the compatibility of perspectivism and realism. She emphasises the distinction between metaphysical and epistemic relativism, saying that the problem in the past has been people conflating the (epistemic/scientific) rejection of objectivity with relativism about truth (2017,170). Even with this specification in place, worries about epistemic relativism might still remain. Giere considers two in particular, which I therefore need to address. Both of these turn out to be unproblematic because of the second specification that I will make: perspectival realism is a form of non-neutral relativism, rather than an equality-based relativism.
The first worry is that a relativist version of perspectival realism would run afoul of 'the reflexive question'. Giere first mentions this question in connection with social constructivist theories. It asks whether or not views' conclusions about contingency apply to themselves. If they don't, then holders of that view (in that case constructivism, in this case perspectival realism) grant their own conclusions exceptional objectivity-a surprising position, which requires further justification. And if they do, then holders of the view are forced to admit that their own conclusions are merely contingently (or relatively) justified (Giere 2006, 11).
Giere seems to consider this question to be problematic for social constructivists. He says that they have a "way out", but cautions that it means aligning themselves with radical critics of science (2006,11), whereas he claims that "no such problems" arise for the perspectival realist (Giere 2006, 95). However, my suggestion in the previous section, that perspectival realists embrace non-neutrality, might be thought to introduce these problems. It seems to commit perspectival realists to something like the second answer to the reflexive question, forcing them to admit that their own claims are 'merely' relatively justified.
Whilst I do think that perspectival realists are committed to the second response to the reflexive question-i.e., to saying that their conclusions themselves are perspectival, and so are contingently or relatively justified-I don't think that this should be considered costly.
The apparent cost of relative justification is that it is 'mere', or a somehow lesser form of, justification. But lesser than what? This worry presumes that there is some other, bigger, better form of justification which relative justification falls short of. This is an absolutist view. On any relativist view (including perspectival realism) which denies both the existence and possibility of absolute justification, contingent, relative, perspectival justification is the best and most legitimate form of justification available. So if perspectival realists acknowledge and accept the relativism in their view, then they can endorse the second answer to the reflexive question without undermining themselves, and without casting aspersions on science.
Massimi's (2018) account shows us one possible way to flesh out this idea. She proposes that scientific perspectives play a "double role", where they function both as a context of use, which fixes its own standards of performance adequacy, and as a context of assessment, from which other perspectives can be evaluated (according to how well they meet the standards of performance adequacy which they have set 2018: 353-7). On this view, perspectives aren't the sole measure of their own success, because other perspectives functioning as a standard of assessment are important too, and so worries about lesser or redundant forms of justification are avoided. But no absolute, or perspective-independent, evaluation is required either, and so both non-neutrality and reflexivity are present. 17 The second worry is that epistemic relativism "can be pushed to the absurd extreme that every perspective is regarded as good as any other" (Giere 2006, 13). This view is sometimes summarised using the phrase 'anything goes' (because all options are equally epistemically permissible). Giere denies that this conclusion follows from his view, making a distinction between perspectivism and "silly relativism". I agree with the letter of this claim, but perhaps not the spirit. 18 Regardless, I will explain why the 'anything goes' conclusion doesn't follow from non-neutrality in particular, and why relativism based on non-neutrality is therefore non-silly.
It's easy to see why this worry arises. The 'anything goes' conclusion can be derived from equal validity, which we've seen is commonly (though mistakenly) associated with relativism. Equal validity says that there is no way to rank epistemic systems whatsoever, and so (from an absolute, independent non-perspective) they are all as good as each other. On this sort of view, any perspective is legitimate, and so: anything goes. But as equal validity is not a component of perspectival realism, this source of the worry is not a concern for us.
Perspectival realism is only committed to non-neutrality. Non-neutrality comes from the idea that all justification is dependent, or perspectival, and so views incorporating non-neutrality aren't able to state absolutely that one perspective is better than another. This might seem like another reason to worry about the anything goes conclusion besides equal validity. However, it's not. It is possible to say on non- 17 Massimi denies that her account is relativist, framing it as an alternative to both relativism and scientific objectivism. Whilst I agree that her account (2018) is not a form of alethic relativism (as discussed above), I believe it is a version of (non-silly) epistemic relativism, i.e., it incorporates dependence, plurality, and non-neutrality. In conversation Massimi has suggested that her view avoids non-neutrality (and so relativism) because it is able to account for scientific progress, however I have argued (forthcoming) that scientific progress and non-neutrality are compatible. 18 It's not clear to me whether Giere thinks that all relativism is silly, or whether he accepts (as I do) that perspectival realism is a form of non-silly relativism and was merely distinguishing this from a subset of relativist views which are silly. neutral relativism that some perspectives are better or worse relative to a particular perspective. In other words, it is possible to provide a non-neutral ranking of different epistemic systems. As long as one is taking a particular perspective-and according to perspectival realists this is always the case-it is possible to rank and evaluate epistemic systems, and so it is not the case that 'anything goes'. I don't expect the promise of this relativized ranking to be enough to satisfy everyone. I've found there to be a common and strong absolutist intuition that some views are just better (or worse) than others, in an unrestricted and unmitigated sense, and so if perspectival realism can't account for this it might leave some people disappointed. I am sympathetic to this intuition, but it is an absolutist intuition, incompatible with views like perspectival realism and standpoint theory which are based on the situatedness of knowledge. Perspectival realists cannot consistently allow scientists to say that practices other than their own are absolutely unjustified. What they can do is show that non-neutral, perspectival rankings satisfy the criteria for a useful and desirable philosophy of science. This is what Giere and Massimi already claim to do: they argue that perspectival realism reconciles a useful scientific realism with a plausible understanding of knowledge from a human point of view. Acknowledging that the relativist label applies to this view clarifies, rather than undermines, this work.

Conclusion
I began this paper by pointing to some similarities between perspectival realism and feminist standpoint theory. I said that they both see knowledge as situated, and that they have each been framed as a middle way between scientific objectivism on the one hand, and anti-realist or relativist views on the other. I then argued for a further similarity. The standpoint thesis which is central to feminist standpoint theory commits them to dependency, and ultimately to a kind of relativism, and the perspectivism in perspectival realism does the same for their view. I have now clarified that the version of relativism present in perspectival realism is unproblematic. It is an epistemic relativism, which avoids the complications of metaphysical versions of relativism, and a non-neutral, non-silly relativism, which avoids the 'anything goes' conclusion. In closing, I will suggest two places where perspectival realism and feminist standpoint theory can learn from each other.
First, I think that perspectival realism has an opportunity to learn from feminist standpoint theory by expanding the range of factors that constitute a perspective. I haven't argued for this here, but feminist standpoint theorists have argued that decisions about which instruments to use, which methods to deploy, and which interpretative conventions to rely on aren't just affected by factors that scientists are consciously aware of. They argue that they (and other aspects of the scientific process) are also affected by social factors like race and gender, that scientists might not be consciously aware of. Feminist standpoint theorists have had a hard time convincing most 'mainstream' epistemologists that this is the case, I think largely because philosophers who are committed to absolutism and objectivity are primed to see views which don't have these commitments as attacks on science and its goals, rather than suggestions for how to achieve those goals more effectively. I suspect that perspectival realists won't have this problem, as their project is accepted as part of 'mainstream' philosophy of science and will thus be more amenable to the kinds of arguments that standpoint theorists make.
Second, I think that perspectival realists have something to teach standpoint theorists, namely that to be consistent they need to embrace reflexive perspectivality (or situatedness). Giere readily accepts that his own claims are as situated as those that he is attempting to theorise about, but standpoint theorists have been reticent to do this. This is a problem, because if they don't recognise that their own claims (in particular the epistemic advantage thesis) are situated too, then they are not fully internalising the situated knowledge thesis and are allowing a tension to creep into their view. Standpoint theorist's reticence to do this has, as far as I can tell, has been motivated by the goal of avoiding epistemic relativism, in an attempt to be taken seriously by 'mainstream' philosophers of science. My hope is that seeing this move made by perspectival realists, who are more mainstream, will ease these concerns and help feminist standpoint theorists too to embrace (non-silly) epistemic relativism.
Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.
The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.