1 Introduction

This paper addresses the question of whether, and to what ends, historians of philosophy should employ the concept ‘Western Philosophy.’ Specifically, this paper draws on scholarship that addresses identical questions about ‘Analytic Philosophy’ in order to respond to those applicable to ‘Western Philosophy.’ The outcome, we contend, is that despite problems of coherency and problems arising from the racist and imperialist origins of the concept of ‘Western Philosophy,’ that concept might nevertheless fruitfully be employed in future history of philosophy scholarship. In fact, we will argue that it might have to be employed precisely to undo the effects of these racist and imperialist origins. This puts us at odds with two recent pieces of scholarship (Platzky Miller 2023; Platzky Miller and Cantor 2023) that have argued that the concept of ‘Western Philosophy’ ought to be abandoned altogether. The aim of this paper is to apply the lessons learnt from discussions about ‘Analytic Philosophy’ and to build on Platzky Miller’s and Cantor’s innovative proposals. We argue that the term should not be abandoned altogether and that continuing to use it is in fact required for combating exclusions in the history of philosophy more generally. (For reasons that will become clear in what follows, we distinguish between Western Philosophy, the tradition, and ‘Western Philosophy,’ the concept under whose extension that tradition is supposed to fall—and similarly between Analytic Philosophy and ‘Analytic Philosophy.’) More generally, this paper brings to light much wider questions about the status of certain concepts and categories (like ‘Analytic Philosophy’ and ‘Western Philosophy’) in the history of philosophy.

This paper proceeds as follows. In Section 2, we outline what we call the Cantor/Platzky Miller Thesis (CPM):

CPM: The concept of ‘Western Philosophy’ should be abandoned.

As we will show, close scrutiny reveals that this thesis is ambiguous and can be construed as having both a descriptive (D) and a normative (N) dimension, each of which requires a separate assessment. In Section 3, we analyse the descriptive dimension of the thesis (D) by responding to its three possible interpretations. In Section 4, we apply further scrutiny to D by appealing to important analogies between discussions about ‘Western Philosophy’ and ‘Analytic Philosophy.’ In Section 5, we distinguish ‘realist’ and ‘functionalist’ uses of categories in the history of philosophy, showing that Cantor and Platzky Miller’s arguments only apply to the former. In Section 6, we address N, the normative dimension of the CPM and expand on an undertheorized dimension of the work of Cantor and Platzky Miller that offers real benefits to historians of philosophy setting out to combat various exclusionary mechanisms of the ‘Western’ tradition.

2 The Descriptive and Normative Dimensions of the CPM

The aim of this section is to provide an overview of Cantor’s and Platzky Miller’s extensive catalogue of problems associated with the use of ‘Western Philosophy’—that they assemble in support of the CPM— and by disentangling the descriptive (D) and normative (N) dimensions of the thesis itself. We focus on Cantor and Platzky Miller’s work, first, because they provide an exhaustive list of concerns and, second, because their works we cite here are part of a much wider ongoing and multi-faceted project (including a recent conference in Oxford) targeted at drawing attention to the question of whether talk of ‘Western Philosophy’ should be abandoned. But that is not to say that they are the only scholars who our discussion is relevant to. In fact, we think the discussion that follows applies to much wider questions about how certain categories and concepts in the history of philosophy are employed and how they ought to be employed.

Cantor and Platzky Miller (2023, 28) challenge what they call the “Standard Narrative” in the history of philosophy. They argue that both “the most influential” and “[m]ost influential histories of philosophy” propagate the idea that the most significant kind of philosophy is “specifically ‘Western Philosophy’” (Platzky Miller and Cantor 2023, 28, emphasis added). Features of this narrative include the belief that Western Philosophy has Greek origins, a kind of insularity (i.e., it is sharply separated from other traditions and thinkers), and a purported continuity with contemporary Europe (Platzky Miller and Cantor 2023, 28–29). They argue that this narrative is misleading because ‘Western Philosophy’ cannot be defined, is geographically inaccurate and serves exclusionary purposes. As a result, they argue that “we should abandon the idea of a ‘Western Philosophy’ and re-examine the history of philosophy without its distorting effects” (Platzky Miller and Cantor 2023, 28).¹

This claim is encapsulated by the CPM, which states:

CPM: The concept of ‘Western Philosophy’ should be abandoned.

The most explicit formulation of this thesis can be found in Cantor and Miller’s co-authored article where they argue that “we should abandon the idea of a ‘Western Philosophy’” (Platzky Miller and Cantor 2023, 33). But, based on textual evidence we provide in Sections 3 and 5, the CPM is a blend of two distinct claims, a descriptive (D) and a normative (N) one. These claims are as follows:

D: There is no such thing as Western Philosophy.

N: We should not employ the concept of ‘Western Philosophy’ in the history of philosophy.

D, the descriptive component of the CPM states that the term ‘Western Philosophy’ does not genuinely refer to or pick out some ‘thing.’ But the precise nature of this ‘thing’ is unclear. On the one hand, that ‘thing’ might be the tradition of Western Philosophy, that is, a set of thinkers and ideas with a traceable lineage. In which case, D is equivalent to the denial that Western Philosophy is something like a natural kind existing independently of our conceptions of things.² On the other hand, D might target the coherence of the concept itself (‘Western Philosophy’). It is not immediately clear which of these options best captures Platzky Miller and Cantor’s own position. We disambiguate these in Section 3.

By contrast, N amounts to the claim that historians of philosophy should stop employing the concept of ‘Western Philosophy.’ If N were true, philosophy as a discipline would need to find different historical categorisations. Platzky Miller offers the notion of ‘entangled histories’ of philosophy as an alternative (Platzky Miller 2023, 1249–54). But N presumably opens up conceptual space for an entire host of methodological approaches to the history of philosophy, all of which have in common their avoidance of ‘Western Philosophy’ as a tool, analogously to the way that all modern theories in physics share a methodological avoidance of ‘luminiferous aether.’

Having now identified two different dimensions of the CPM, we further clarify and critically respond to both in the sections that follow.

3 The Concept of ‘Western Philosophy’

This section clarifies D, the descriptive component of the CPM. We will outline several arguments that have been put forward for D and conclude that although many of them fail, Cantor and Platzky Miller’s argument from vagueness is convincing. Section 4 will use a comparison with the concept of ‘Analytic Philosophy’ to demonstrate that the argument from vagueness does not warrant abandoning the concept of ‘Western Philosophy’ altogether. In Section 6 we will emphasise one key benefit of holding onto the concept, suggested by Platzky Miller, that has not received sufficient attention.

The descriptive component of the CPM states:

D: There is no such thing as Western Philosophy.

The most explicit statement of D can be found in Platzky Miller, who approvingly quotes Ben Kies as paving the way for recognising that “there is no such thing as ‘Western philosophy’” (Platzky Miller 2023, 1236, 1249). However, D could mean one of three things:

D1: There is no such thing as ‘Western Philosophy.’

D2: There is no such thing as Western Philosophy.

D3: The concept of ‘Western Philosophy’ is not coherent.

D1 states that the concept of ‘Western Philosophy’ does not exist.³ D2 states that the natural kind Western Philosophy (i.e., the tradition that exists independently of our conceptualisations of it, and that is supposed to fall under the extension of the concept ‘Western Philosophy’) does not exist. D3 states that ‘Western Philosophy’ is an incoherent concept, like a square circle or a married bachelor. We address the arguments for these in turn.

D1 is the least likely interpretation of D. There clearly is such a thing as the concept of ‘Western Philosophy,’ regardless of whether it is coherent, accurate, meaningful, or whether it refers to anything. Platzky Miller, for example, says that ‘Western Philosophy’ is an “intellectual construct” (Platzky Miller 2023, 1237), a “political myth” (2023, 1240) propping up the racist colonial project, or a “recent fabrication” (2023, 1236). Whatever these are, these are. So D1 is unlikely to be what Cantor and Platzky Miller have in mind and we will not discuss it further.

What of D2 and D3? The first argument produced by Cantor and Platzky Miller that seems to support D2 can be called the “recent fabrication argument.” It goes as follows. ‘Western Philosophy’ is a recent category that would have been unintelligible to most philosophers that are supposed to fall under its heading (Platzky Miller 2023, 1249; see also Platzky Miller and Cantor 2023, 29). The Greeks, for example, certainly did not think of themselves as “Western” and the same is probably true of most individuals in the Western philosophical canon. Since the concept arose recently, it is dubious whether the natural kind it refers to actually existed before its “fabrication,” or whether the natural kind was fictionalised as an ad hoc entity (like witches) to justify the colonial project.

However, not only does the “recent fabrication argument” not support D3, but it also does not really support D2 either. This is because, however artificial the fabrication of a concept was, that concept might still be coherent. For instance, the concept of a ‘witch’ is fairly coherent even though it does not refer to a real entity and was also fabricated. Furthermore, it does not follow from the fact that a particular historical category is recent that it cannot refer to the natural kind whose existence predates it. ‘Feudalism’ did not originate as a concept until the eighteenth century and yet the term refers to a real political arrangement existing prior to its baptism. So the fact that a concept was fabricated recently does not mean that the natural kind it refers to did not previously exist. The recent fabrication of ‘Western Philosophy’ does not mean that there is (or has been) no such thing as Western Philosophy.

There is another, closely related, argument which we might call the “argument from dodgy origins.” The emergence of ‘Western Philosophy’ as a historical category coincides (and not coincidentally) with a set of objectionable values and worldviews. Its origins are dodgy. And it does not particularly matter whether these are tied to colonial dominance from the 1600s onwards (Platzky Miller 2023, 1243) or to the way that “[a]fter the Second World War, these racialized narratives were repurposed to build a common identity across Europe, and to justify continued European dominance globally” (Platzky Miller 2023, 1244). As Ben Kies (in Platzky Miller’s reconstruction) argues, “The idea of ‘Western Philosophy’ is the product of a legitimation project for European colonialism” (Platzky Miller 2023, 1234). This part of the argument seems prima facie coherent: ‘Western Philosophy’ is inseparably connected with Western imperial dominance. So we should not use it, because we cannot discard its racist baggage.

But the argument from dodgy origins does not support D2. It simply states that we should not use the concept, because it cannot shed its racist baggage. It is an argument for N, the normative dimension of CPM, not D, the descriptive one. We will return to N in Sections 5 and 6. For the time being, the bigger problem with the argument from dodgy origins is that it not only fails as an argument for D2, but it also has no bearing on D3. It has no bearing on D3, that is, the claim that there is no coherent concept of ‘Western Philosophy,’ since for all we know, the concept might be recently fabricated, and fabricated with racist motivations, but might, after all, be fabricated coherently. Those nineteenth-century racist historians who, as Cantor (2022) argues, fabricated the mythology of Western Philosophy originating with Thales, might have just paid extra attention to the coherence of their fabrication.

The third argument in support of D2 can be called the “argument from obscure origins.” It goes as follows. While we can identify where the concept of ‘Western Philosophy’ originated (at some point within the past three hundred years or so), working out the precise beginnings of the tradition (i.e., natural kind) that is supposed to fall under its extension seems impossible. In other words, we can identify where ‘Western Philosophy’ originates, but we cannot identify where Western Philosophy does. Cantor (2022) suggests that the prevailing view is that this great originator was Thales. But as her recent paper rightly demonstrates, this claim has a shaky truth status and politically dubious motivations: “One of the reasons for [Thales acquiring the status of the first philosopher] was the rise of pseudo-scientific racism” (Cantor 2022, 728). Not only is the endlessly perpetuated story about Thales false, but it was also fabricated with racist intent. Cantor is right about this. But her claim about Thales is mobilised, in the co-authored paper with Platzky Miller, to back up a much more radical implication, namely, that because it is impossible to tell where Western Philosophy originates, there is no such thing. Natural kinds, one might argue, are just not the types of things that should have unclear boundaries. So if Western Philosophy is whatever is supposed to start with Thales and it does not, then there is no such thing as Western Philosophy.

Unlike the first two arguments, the argument from obscure origins does indeed lend credence to D2, the claim that there is no such thing as (the natural kind) Western Philosophy. However, it is irrelevant to the truth of D3. The fact that it is impossible to find a clear starting point for a philosophical tradition does not mean that any concept applicable to that tradition is inevitably incoherent.⁴ This tradition might just fall under the extension of a fuzzy concept. Fuzzy concepts do not have sharp boundaries, but neither are they incoherent. ‘Being drunk’ is a fuzzy concept without a rigid boundary which entails that it is difficult to identify its first instance. It is not clear when a given individual S was first drunk: was it when S had their first glass of wine at the age of 16 and felt tipsy? When S finished their first bottle at the age of 18 and could not walk straight? Or when S downed a bottle of vodka at the age of 21 and fell over? It is not only difficult but impossible to say. Yet the fact that we cannot identify the first instance of “S being drunk” does not mean that there is no concept of ‘drunkenness,’ or no coherent concept of ‘drunkenness.’ So while the argument from obscure origins might support D2, it has no impact on the truth status of D3.

However, Cantor and Platzky Miller produce several further arguments that support D3 in a way that they do not support D2. We believe that their arguments for D3 provide crucial insights regarding the status of categories in the history of philosophy.

The first argument might be called “geographical.” It targets the assumption that “‘Western Philosophy’ is a straightforward geographic designator” (Platzky Miller 2023, 1238) and that the concept of ‘Western Philosophy’ can be defined derivatively using the concept of ‘the West.’ On the geographical reading, ‘Western Philosophy’ is philosophy practised in the West. However, the argument goes, it is not very clear what ‘the West’ is. What is it supposed to be west of? More importantly, it is rightly pointed out that philosophy in Australia or New Zealand—both of which are very much east of wherever Plato, Descartes, and Hegel were writing—are standardly counted as forming part of the concept of ‘Western Philosophy’ (Platzky Miller 2023, 1241; Platzky Miller and Cantor 2023, 30). But this is absurd. If the concept of ‘Western Philosophy’ should be defined by appealing to philosophy practised in the West understood purely geographically, and several of its paradigmatic representatives are geographically not located in the West, then the concept is self-contradictory.

This argument is inconclusive. We agree that ‘the West’ does not refer to a unique, isolated, distinct, (let alone culturally superior) entity. We also agree that it is not a geographical designator. However, the geographical argument is irrelevant to the possibility of there being a distinct tradition that has come to be called (under highly suspicious circumstances) ‘Western Philosophy.’ Such a tradition might not exclusively, or even primarily, be located in ‘the West.’ It might be practised by Martians for all we know. It could also be called something else, for example, the “European philosophical tradition” (Rée 2001, 230), “Greek-originating,” or perhaps even “hegemonic.” As Voltaire famously quipped, the Holy Roman Empire was neither Holy, nor Roman, nor an Empire. And yet the phrase denotes a real thing which, importantly for our purposes here, falls under a coherent concept referred to by the phrase “The Holy Roman Empire.” The same might very well be the case with ‘Western Philosophy.’ The name for the concept might contradict its meaning or even contradict itself, but that does not mean that the concept is self-contradictory. The most the geographical argument induces us to do is just to replace Western Philosophy’s name with a different one.⁵ But addressing the exclusionary function of ‘Western Philosophy’ certainly requires more than sticking a different label (“European”) on whatever it is that is responsible for the excluding and continue doing business as usual.⁶ Regardless of whether this purely semantic point is all that the CPM boils down to (and we do not think it is), the geographical argument certainly does not induce us to accept D3.

The second argument for D3 could be called the “argument from vagueness” and is a more encompassing version of the geographical argument. This argument, as we will show in the section that follows, is both the most powerful and the one most relevant to the comparison with similar arguments regarding the concept of ‘Analytic Philosophy.’ It is the claim that any attempt to define ‘Western Philosophy’ (geographical or otherwise) will inevitably fail. Cantor and Platzky Miller (2023, 30) argue that “[w]hen explanations [of what Western Philosophy is] are attempted, these turn out to be implausible, unstable or nonspecific to this supposed ‘tradition.’” This is, in our view, the strongest argument for D3. If it succeeds, it not only provides a convincing case for accepting D3, but by extension and together with the assumption that D3 is the correct interpretation of D, it also supports D. We turn to the argument from vagueness in the next section.

4 ‘Western Philosophy’ and ‘Analytic Philosophy’

As we have seen, the questions raised by D are not so much about Western Philosophy (the tradition), but about ‘Western Philosophy’ (the concept of that tradition). What exactly is ‘Western Philosophy’? A fruitful way of addressing this question is to turn to similar questions that have already been raised concerning a different tradition, namely, Analytic Philosophy. In this section, we will specifically appeal to Glock’s discussion of the concept in What Is Analytic Philosophy? (2008). We will show that his discussion is equally informative for debates about the concept of ‘Western Philosophy’ and that his reasons for thinking the concept of ‘Analytic Philosophy’ ought not to be abandoned apply equally well to the case of ‘Western Philosophy.’

This section has two aims. First, to explore analogies between attempts at defining ‘Western Philosophy’ and ‘Analytic Philosophy.’ Second, to identify what can be learnt from definitions of ‘Analytic Philosophy’ when it comes to generating a coherent definition of ‘Western Philosophy.’ Specifically, we argue that this analogy reveals that the argument from vagueness fails and does not establish the truth of D3. Of course, it might be objected that what is true of a tradition spanning just over a hundred years cannot be true of one that supposedly spans millennia. But as we will see below, although the scale of the phenomena that each concept is supposed to capture is vastly different, the structural problems with defining how such concepts capture them are very similar.

Debates about ‘Analytic Philosophy’ are relevant to our present concerns because they display at least four structural similarities with debates about ‘Western Philosophy.’ First, identical claims as the one that there is no such thing as Western Philosophy have been made in the case of the analytic tradition: some scholars have claimed that there is no such thing as ‘Analytic Philosophy’ (see Preston 2004, 462–63). The second point of relevance concerns evaluative judgements about each tradition. At least superficially, it seems like one of the motivations for the CPM is the recognition of the nefarious effects that the concept of ‘Western Philosophy’ has exerted over the centuries. But this recognition, true though it likely is, does not motivate D. We do not have to believe that ‘Western Philosophy’ is a good thing to believe that it is a thing. The same is true of the analytic tradition:

Understanding of the term ‘analytic philosophy’ is tied to the ability to specify certain figures, movements, texts and institutions, and perhaps some of their prominent features. It does not require the belief that analytic philosophy is at any rate a jolly good thing. (Glock 2008, 209)

The belief that there is such a thing as ‘Analytic Philosophy’ might be entailed by its being a bad thing—for example, its being a tradition that often uses “continental” as a derogatory term (Glock 2008, 63–63; Campbell 2001, 343).

The third analogy is an obverse of the second. Several figures in the analytic tradition have made the claim that there is no distinction between analytic and non-analytic philosophy: there is just a distinction between good and bad philosophy. Analytic philosophy just miraculously happens to be the good one—as Buckle (2004, 112) puts it, “an unsurprisingly happy result for the analytic philosopher!” Similarly, it has been assumed by some Western Philosophers that Western Philosophy just is philosophy, with other intellectual traditions being relegated to “thought” or something of the sort (for a critique of this, see Schuringa 2020; see also Connell and Janssen‐Lauret 2023 for discussion of this idea in relation to women in the history of philosophy). Thus, in a surprising plot twist, the staunchest defenders of ‘Western Philosophy’ might turn out to agree with the CPM, namely by insisting that, true enough, there is no such thing as ‘Western Philosophy,’ because the only distinction is between philosophy and non-philosophy, the former being the type of intellectual activity that just happens to be practised in ‘the West.’ Abandoning ‘Western Philosophy’ carries the risk of inadvertently enforcing such viewpoints.

The fourth analogy concerns the claim that the term ‘Analytic Philosophy’ is also a recent fabrication. As Glock (2008, 24) notes, the term ‘Analytic Philosophy’ only takes off after the Second World War. But the same response we proposed in the previous section applies. Just because the concept is recent does not mean that it is incoherent or that it does not pick out the instantiations of the natural kind whose existence predates it.

These analogies suggest that comparing the nature of ‘Analytic Philosophy’ and ‘Western Philosophy’ might provide a useful way to address CPM.

The most important point of comparison is the argument from vagueness. Platzky Miller offers four ways that ‘Western Philosophy’ might be defined but argues that all of them fail. These options conjointly constitute the argument from vagueness:

The first is ‘Western Philosophy’ is [sic] a straightforward geographic designator: it is obviously just Philosophy from the West. A second account is that ‘Western Philosophy’ is distinctive because of some unifying essence or unique, inherent characteristic, such as being reasoned, argued, or secular. A third account suggests that there is a tradition, or cultural/intellectual continuity. . . . A final species of justification includes ethnocentric celebrations of ‘The West’ and ‘Western Philosophy’ which attempt to give it a ‘Greek/Judeo-Christian’ veneer. (Platzky Miller 2023, 1237)

We label these options (1) geographical, (2) essentialist, (3) traditional, and (4) ethnocentric. In what follows, we consider the analogues of the first three of these definitions in the literature on the analytic tradition. (We will not discuss ethnocentric definitions here. Apart from being incoherent, we agree that ethnocentric definitions are morally unacceptable for the reasons highlighted by scholars like Allais, Cantor, Platzky Miller, and Schuringa.)⁷ The purpose of this is to strengthen the case for structural similarities between the concepts of ‘Analytic Philosophy’ and ‘Western Philosophy.’ At the end of this section, we will consider a fifth possible option for coherently defining ‘Analytic Philosophy’ proposed in Glock’s What Is Analytic Philosophy? and assess whether an analogous fifth option for defining ‘Western Philosophy’ can be coherently provided.

(1) The first way of defining ‘Analytic Philosophy’ is by appealing to geography. This is analogous to the geographical attempts to define ‘Western Philosophy’ criticised by Platzky Miller. Just as some have tried to identify Western philosophy with philosophy in ‘the West,’ Analytic Philosophy has frequently been identified as philosophy materialising elsewhere than “The Continent.” But, as Glock summarises, this approach fails because it

is indifferent to geographic variations within continental Europe, the present ascendancy there of analytic philosophy, the importance of non-analytic ways of philosophizing in Anglophone countries, and the fact that continental philosophy is neither the only nor in many respects the major alternative to analytic philosophy. (Glock 2008, 80)

The origins of the analytic tradition seem to be very much on the continent (e.g., in the Vienna Circle)—just as the ‘Greek’ origins of what is typically taken to be Western Philosophy, according to Cantor (2022) and Stone (2017), are with the Egyptians, Babylonians, and Phoenicians. So the geographical approach fails in the case of ‘Analytic Philosophy’ as much as it does in the case of ‘Western Philosophy.’

(2) The essentialist attempt at defining ‘Analytic Philosophy’ identifies the analytic tradition with a distinctive doctrine, method, or style. There are many options for what this unique distinctive feature could be. It might be the linguistic turn. Or it could be a specific approach to history—be this historiophobia or ahistoricism (Glock 2008, 90). Or a specific method or style, typically construed as clarity or focus on argument (Glock 2008, 151–77; Buckle 2004, 111; Trakakis 2012). However, there are counter-examples to all such criteria. Identifying analytic philosophy with the linguistic turn would exclude Russell and Moore, but might allow us to count Foucault and Derrida as analytic philosophers (Vrahimis 2018, 110). Moreover, many of those generally accepted as analytic philosophers appeal to historical thinkers (as evidenced by the presence of Aristotle, Hume, or Kant in contemporary metaphysics). A focus on clarity is in danger of being too narrow a criterion; one that excludes paradigmatic figures. For instance, it would be quite a push to describe Wittgenstein as “clear.” On the other hand, a focus on argument seems too wide or permissive, with the result being that most philosophy ends up being analytic (Glock 2008, 178). In short, essentialism regarding ‘Analytic Philosophy’ fails just as much as essentialism regarding ‘Western Philosophy.’

(3) The third attempt is to identify Analytic Philosophy with a particular intellectual continuity starting with, for example, the logical positivists, the Vienna circle, or Russell and Wittgenstein. But just as attempts to identify the first representative of Western Philosophy end up running into contradictions, the same is true with attempts to identify where Analytic Philosophy started. Just as it is hard to say why precisely Thales or Parmenides should come first, the same is true of many figures in the case of the analytic tradition: did it all start with Bolzano? Frege? Russell? Moore? (for a discussion, see Glock 2008, 224–30). These questions seem just as difficult to answer as those that Cantor (2022) raises regarding the supposed “first philosopher” Thales.

So Platzky Miller and Glock are both right. Geographic, essentialist, and continuity-based approaches to defining philosophical traditions fail to generate a coherent concept as much in the case of ‘Analytic Philosophy’ as they do in the case of ‘Western Philosophy.’ At this stage, it might seem that the argument from vagueness cannot be defeated, that no definition of ‘Analytic Philosophy’ can be given and similarly—as D3 states—that no coherent concept of ‘Western Philosophy’ can be provided either.

But that would be jumping the gun. Observe that people continue talking about ‘Analytic Philosophy’ even though most of those who do so would probably be unable to provide a coherent definition of it resistant to counter-examples. The same is true of ‘Western Philosophy,’ as evidenced by the stark contrast between the serious problems with satisfactorily defining ‘Western Philosophy’ highlighted by Allais, Cantor, Platzky Miller, and Schuringa on the one hand, and entire series dedicated to histories of Western Philosophy on the other. Even though “so far no consensus has emerged as to what precisely constitutes ‘analytic’ philosophy,” (Campbell 2001, 341, emphasis added) consensus does generally arise about when something (usually a paper) is an instance of ‘Analytic Philosophy.’ This consensus exists even if we cannot precisely define ‘Analytic Philosophy.’ Analogously, someone writing a book on Zhu Xi is unlikely to submit it to one of OUP’s “History of Western Philosophy” series, even though they might not be able to provide reasons why that are resistant to counter-examples. Similarly, though they might struggle to justify this recognition, most philosophers would probably agree that they can recognise paradigm cases of ‘Analytic Philosophy’ (Wittgenstein and Russell yes, Heidegger and Lacan no) just as they could recognise paradigm cases of Western Philosophers (Descartes and Locke yes, Ramanuja and Zhuangzi no). They might be able to recognise borderline cases (Ibn Sina in the case of ‘Western Philosophy,’ Whitehead in the case of ‘Analytic Philosophy’) without necessarily being able to specify precisely why these cases are borderline. Of course, according to Cantor and Platzky Miller, this unassumed and uncritically perpetuated ‘consensus’ amongst philosophers about who gets to count as ‘Western’ is precisely the problem here; getting rid of the concept of ‘Western Philosophy’ should be the first step in changing this status quo. We think that Cantor and Platzky Miller are right about this. But this problem falls under N, the normative dimension of the CPM, which we will put aside for the time being. All that we are interested in now is whether this concept—which might have to be ditched for moral or political reasons—cannot be coherently articulated. We think it can be.

The best way to start to unravel the argument from vagueness is by noting that it seems to rely on the idea that the only way to coherently define a concept is to do so using necessary and sufficient conditions. The conclusion that there is no coherent concept of ‘Western Philosophy’ is inferred from the fact that one can always find notable counter-examples to any single criterion which is associated with it. These criteria are quite diverse and include things like the geographical, essentialist, or continuity criteria we discussed above. As we have seen, each of these criteria on its own is faced with a plethora of counter-examples. If anything that is a specimen of ‘Western Philosophy’ has to meet all of these criteria to count, then nothing counts. If anything that is a specimen of ‘Western Philosophy’ has to meet one of these to count, then almost everything counts. Both options are unworkable.

However, Glock’s critique of geographical, essentialist, and traditional definitions of ‘Analytic Philosophy’ is not purely destructive. Rather, it aims to clear the ground for a constructive approach. His suggestion is that we define ‘Analytic Philosophy’ through a combination of two elements: (i) family resemblance and (ii) historical tradition. In the case of family resemblance, what holds ‘Analytic Philosophy’ together is a similarity relation to certain paradigmatic instances and paradigmatic features (Glock 2008, 214). While the historical-tradition condition means that

[A]nalytic philosophy is first and foremost a historical sequence of individuals and schools that influenced, and engaged in debate with, each other, without sharing any single doctrine, problem, or method. (Glock 2008, 220)

The notion of mutual influence is as important as the claim that the individuals did not share any single doctrine, problem, or method. The fact that someone influenced the analytic tradition is not enough to count them in (otherwise the Greeks would count as analytic philosophers). And being influenced by it is not enough either (otherwise, Glock claims, we would get Lyotard in). Influence is also not equivalent to a mere dialogue: Derrida is not an analytic philosopher just because he debated Searle (Glock 2008, 223).

The specifics of Glock’s approach need not concern us here. But what is important for our present concerns is that his two criteria for defining ‘Analytic Philosophy’ can equally be applied to ‘Western Philosophy.’

First, Glock’s emphasis on family resemblance does away with the futile attempt at searching for a single unifying feature (or a single set of unifying features) that could serve as necessary and sufficient criteria for belonging to Western Philosophy. The ‘Analytic’ and ‘Western’ traditions are just too diverse in terms of geography, methods, and historical influences, for any of these (or a single set of these) to serve as a criterion. The concept of ‘Western Philosophy’ does not need to have a single distinct unifying feature recognizable across two millennia for it to be coherent. There will inevitably be borderline cases and some instances will be more paradigmatic than others, depending on precisely how many of the defining features they tick off. Platzky Miller is right that on their own, the geographical, essentialist, and traditionalist accounts fail. But conjointly they are members of a set of criteria such that, the more of them an instance of philosophy ticks off, the more it counts as being an instance of ‘Western Philosophy.’ Cluster concepts might be fuzzy, but they are coherent.

Second, a crucial element of Glock’s definition is that influence must be mutual. This approach has two benefits. The first is that it gets around the issue of ‘Western Philosophy’ necessarily having to be an insular phenomenon. Just because Schopenhauer drew on Indian philosophy and was a Western Philosopher, that would not lead most members of the philosophical community to the counter-intuitive conclusion that Indian philosophy is part of ‘Western Philosophy’ too. (Once again, perhaps it should. But this, again, is a normative question. We address it in Sections 5 and 6.) But, secondly, the mutual influence criterion can also be used in the opposite direction, to re-engineer the concept of ‘Western Philosophy’ itself. For example, as Platzky Miller (2023, 1242) rightly notes, “philosophy in the Islamic world . . . [was] responding extensively to ancient Greek philosophers—and impacting philosophy in Europe for centuries thereafter.” Provided that the family resemblance condition is met (and considering the links between topics, methods, and approaches shared by Medieval philosophy in Europe and in the Islamic world, it seems like it is), then Glock’s second criterion can be mobilised to make a convincing case for Islamic philosophy being part of ‘Western Philosophy,’ even if that goes against current intuitions. In short—mutual influence can serve both to block counter-intuitive implications of the concept, but also to change our conception of ‘Western Philosophy.’ Both of these moves strengthen the concept’s coherence.

Glock’s approach abandons the idea that a philosophical tradition must be defined using necessary and sufficient conditions (his definition of the concept ‘Analytic’ is not conceptually analytic). This gets around the descriptive problem raised by the CPM. So it is possible that D, the descriptive dimension of the CPM, is false since D3 is false. There is a coherent concept of ‘Western Philosophy’: one defined using family resemblance incorporating conjointly the geographical, essentialist, and traditional criteria—and mutual influence. This concept is coherent even if no set of necessary and sufficient conditions can demarcate what falls under it.

This section has established that ‘Western Philosophy’ need not be abandoned on the grounds of vagueness or coherency. But the fact that we need not abandon it does not on its own entail that we should keep it. Our argument so far has been drawn up without considering N, the normative dimension of the CPM, which states that we should not employ the concept of ‘Western Philosophy’ in the history of philosophy. We address this in the following sections.

5 Realist and Functionalist Approaches to ‘Western Philosophy’

This section clarifies N, the normative dimension of the CPM, and its connection to D, the descriptive dimension. We will demonstrate that considering N reveals important lessons about the way categories are used in the history of philosophy. We distinguish two such approaches, which we call ‘realist’ and ‘functionalist.’ This exploration of the use of categories paves the way for our discussion of positive motivations for keeping the concept of ‘Western Philosophy’ that we provide in Section 6.

Specifically, we respond to two different interpretations of N that we label as follows:

N1: We should not use the concept of ‘Western Philosophy’ in the history of philosophy because there is no such thing as Western Philosophy.

N2: We should not use the concept of ‘Western Philosophy’ in the history of philosophy because of its immoral origins.

First, let us look at N1. One reason Cantor and Platzky Miller provide for thinking that the concept of ‘Western Philosophy’ should not be employed is that it does not pick out a genuinely existing tradition. For example, Platzky Miller claims that “[m]ost authors simply take for granted that the term [‘Western Philosophy’] makes sense and picks out something real” (2023, 1235, emphasis added). Later he criticises this assumption and describes ‘Western Philosophy’ as a category “to which reality never quite corresponds” (2023, 1255). In other words, at least one motivation for the normative dimension of the CPM depends on a specific interpretation of the descriptive dimension, namely:

D2: There is no such thing as Western Philosophy.

This kind of criticism of ‘Western Philosophy’ is premised on the assumption that when historians of philosophy use the concept, they intend to pick out a really existing and sharply delineated set of thinkers and ideas. In other words, this criticism assumes that the concept ‘Western Philosophy’ is being used in what we might call a “realist” sense—that for the concept of ‘Western Philosophy’ to be coherent, it should pick out a natural kind. And if whatever the concept is picking out cannot be defined analytically (using necessary and sufficient conditions), then the thing that’s being picked out is not a natural kind. The world is not fuzzy so the natural kinds that we find in it cannot be fuzzy either.⁸ And, the assumption goes, if the concept does not pick out a natural kind, then it is an incoherent concept. This assumption can be stated as follows:

D2 => D3

In other words, if there is no natural kind Western Philosophy, then there is no coherent concept ‘Western Philosophy’ either. And since, according to Cantor and Platzky Miller, it seems likely that D2 is true (for example, due to the argument from obscure origins discussed in Section 3 that directly supports D2), then, by modus ponens, there is no coherent concept of ‘Western Philosophy.’ This conclusion is then, in the next step, used to justify N1:

D3 => N1

For scholars such as Cantor and Platzky Miller, it might be this kind of reasoning that leads to the conclusion that it is significant that a plausible set of necessary and sufficient conditions for membership cannot be drawn up. It might also explain why they do not consider the approach to philosophical traditions suggested by Glock, who dispenses with the requirement to provide a rigid set of necessary and sufficient criteria altogether. If one is thinking about the concept of ‘Western Philosophy’ with the “realist” assumption outlined above in mind, then it is understandable that one would want to be able to identify necessary and sufficient criteria for membership. For if Western Philosophy really is (or was) a natural kind out there in the world, then for any historical philosopher X it has got to be true or false that X was a Western Philosopher.

Yet, there are reasons to doubt the initial plausibility of this “realist” approach to philosophical traditions. For one thing, it looks as though the lack of necessary and sufficient conditions does not just apply to ‘Western Philosophy’ and ‘Analytic Philosophy,’ but to pretty much any philosophical tradition, and probably to ‘philosophy’ itself. Thus, if ‘Western Philosophy’ goes, so does every other philosophical tradition. Assuming that we want to hold on to some concepts that pick out philosophical traditions, this constitutes a prima facie case against this “realist” assumption.

What’s more, when historians of philosophy talk of ‘Western Philosophy,’ it does not seem to us that they are trying to pick out a natural kind, i.e., some really existing group of ideas and thinkers ‘out there.’ It seems to us that most historians of philosophy, instead of using the term realistically, use the concept of ‘Western Philosophy’ functionally.⁹ (Whether they should use it at all is a different question which we will address shortly.) A functionalist understanding of ‘Western Philosophy’ is more pragmatic than descriptive, in that it serves a specific practical purpose rather than denoting a natural kind (i.e., a tradition or school of thought). For the functionalist, the emphasis is on the use to which the concept is put, as opposed to any correspondence between the concept and the world.

On the functionalist approach, which we think captures the way most scholars use the term ‘Western Philosophy,’ necessary and sufficient criteria are not required for a coherent definition of the concept of ‘Western Philosophy.’ There are several reasons for this. First, on a functionalist account, there need not be any set of ideas and thinkers that, independently of our historical interests, constitute Western Philosophy. Just what, exactly, Western Philosophy is (or whether it ‘really’ exists) is not an important question for a functionalist. D2 could be true for all we know. It does not matter. There can be ‘Western Philosophy’ without Western Philosophy. This is similar to the way that for a scientific instrumentalist, there need not be any ‘true’ way that reality is organised.

Consider the following analogy from the domain of natural (or ordinary) language. In a biological sense, a tomato is a fruit. Yet were I putting together a fruit salad, adding tomatoes would be an odd (unfruitful) choice. Someone might quite reasonably object that that’s not what they meant when they asked for a fruit salad. The realist (biological) use of ‘fruit’ is different from the functionalist use of ‘fruit’ (for example, when its desired function is to be a component in a fruit salad). Perhaps tomatoes are a borderline instance of ‘fruit,’ or perhaps one might feel compelled to conclude that there are no strict, objective criteria for membership in the category ‘fruit.’ But it would be quite a strong move to then suggest that, in light of such problems of definition, one ought to abandon the concept of fruit entirely.

To apply this to the connection between the descriptive and the normative claims, those who use the concept ‘Western Philosophy’ in the functional sense would deny that

D2 => D3

D2 might be true (for example, due to the argument from obscure origins from Section 3), without that entailing that D3 has to be true. The truth of D2 becomes irrelevant since D3 can be true or false independently of it. We think D3 is false due to the reasons suggested in Section 4: There is a coherent concept of ‘Western Philosophy.’ It just does not come with necessary and sufficient conditions. And if D3 is false, then the following inference breaks down too:

D3 => N1

N1, the claim that “we should not use the concept of ‘Western Philosophy’ in the history of philosophy because there is no such thing as Western Philosophy,” is false because the non-existence of Western Philosophy has no bearing on the existence or coherence of ‘Western Philosophy.’

But what about the truth of N2? That is the topic of the final section.

6 Against Abandoning ‘Western Philosophy’

This section considers N2. We first consider the most obvious motivations that scholars might have for not wanting to abandon ‘Western Philosophy.’ We will show that these motivations are subject to Cantor and Platzky Miller’s critique. However, building upon a comparison with the category of ‘race’ and a key suggestion offered in Platzky Miller’s paper, we will demonstrate that there are, nevertheless, other positive reasons for keeping ‘Western Philosophy.’ In fact, we demonstrate that the process of ‘questioning’ Western Philosophy suggested by Cantor and Platzky Miller, as well as more general efforts at decolonising philosophical practice, require that the concept be held onto.

One reason why the philosophical community might want to hold onto ‘Western Philosophy’ is practical: one might worry that, without it, we would be obliged to take a view of philosophy’s history that resembles the Humean mosaic. Without categorisation via schools or traditions, the history of philosophy would simply be a mass of individual thinkers with individual claims or bibliographies. But tackling a history of philosophy that looked like that, one might think, would be overwhelmingly impractical. It would be nearly impossible to know where to start with one’s research, how to situate one’s research with regard to others, how to put non-arbitrary boundaries on what one covers and, perhaps more importantly, how to trace the development of ideas. Similarly, one might use the concept of ‘Western Philosophy’ to set parameters for one’s own research or a particular research project. The same might be true of organising conferences, establishing scholarly journals, or designing teaching programmes. So, the defender of ‘Western Philosophy’ might say, one of the main motivations for keeping ‘Western Philosophy’ is purely practical.

The question raised by Cantor and Platzky Miller, however, is whether this state of affairs is a good thing. Given that we accept the evidence that the concept of ‘Western Philosophy’ (however coherent or incoherent it might be) has immoral origins rooted in imperialist and racist values, should we really continue holding onto it? ‘Western Philosophy’ is not like ‘fruit’—it is morally permissible to use the concept of ‘fruit’ in a functionalist way because its desired function (making fruit salads, for example) is morally neutral. ‘Western Philosophy’ is not a morally neutral term. This brings us directly to the second motivation for the normative dimension of the CPM:

N2: We should not use the concept of ‘Western Philosophy’ in the history of philosophy because of its immoral origins.

The assumption driving this claim is that if a concept has immoral origins, it cannot be employed in a moral (or at least a morally permissible) way. In this case, the claim is that ‘Western Philosophy’ cannot drop its racist baggage and that its use for the purposes sketched above (categorising philosophy’s past, organising conferences, constructing undergraduate modules, deciding that, say, Heidegger is ‘Western,’ whereas Kitaro Nishida is not, etc.) just confounds the problems attached to the hegemonic function of Western Philosophy.

Before we address this question, it is worth noting that N2 is a view not unique to Cantor and Platzky Miller. It is shared by a whole host of scholars attempting to undo the historical damage that ‘Western Philosophy’ has done, such as the perpetuated false claims made about it during its fabrication (e.g., the claim about Thales). N2 is completely independent of N1 and, by extension, of D2-D3. N2 can be true regardless of anything to do with the coherence or existence of ‘Western Philosophy’ denied by D. However, we will argue that insisting on D2-D3 (i.e., insisting that there is no such thing as ‘Western Philosophy’ and that we should therefore not use it), which is a view unique to Cantor and Platzky Miller, could actually severely undermine the practical motivations driving N2.

Observe that there are innumerable cases of terms and concepts that have immoral origins, but which have been co-opted or reclaimed, and subsequently either discarded their negative connotations or in some cases were used to dismantle precisely those political problems that brought them into existence in the first place. The term ‘queer’ was historically used as a derogatory term for gay people. But the concept of ‘queerness’ was later co-opted and is now a crucial and positive part of the way members of the LGBTQ+ community define their own identities. This example demonstrates that a concept’s oppressive origins are not necessarily a barrier to its being used, subsequently, in a non-oppressive way. This provides an initial reason to think that the use of ‘Western Philosophy’ might be permissible and Platzky Miller’s claim that “no history that adopts the category of ‘Western Philosophy’ can avoid its baggage” (2023, 1254, emphasis added) is false.

It might be objected that the example with homophobic slurs is misleading. Slurs are reclaimed by people slurred by them, not by the dominant group assessing superiority over them. So while it might be permissible to continue using the category ‘Western Philosophy,’ much depends on who exactly is using it. But this just illustrates that it is permissible to continue using it, if only by philosophical traditions marginalised by ‘the West.’¹⁰ Of course, this does not imply that critics attacking ‘Western Philosophy’ from the margins should continue using the term. But they might. In other words, while Platzky Miller and Cantor’s position (we take it) is that abandoning the concept of ‘Western Philosophy’ is obligatory, there are reasons to think that continuing to use the concept is permissible.

However, we think that the most relevant reason for continuing to use the category of ‘Western Philosophy’ lies in the very motivations that underlie Cantor and Platzky Miller’s arguments for the CPM. This use of ‘Western Philosophy’ is in fact suggested by Platzky Miller himself, albeit only in passing in a footnote. Platzky Miller says that we could write a ‘History of Western Philosophy,’ but only “as a history of how this misleading idea emerged” (Platzky Miller 2023, 1249 n 16). Writing a History of ‘Western Philosophy’ (as opposed to writing yet another History of Western Philosophy) in the sense that Platzky Miller here suggests is an example of how it might be fruitful to continue using ‘Western Philosophy’ to bring to light and dismantle its nefarious function. Strictly insisting on D, as the CPM instructs us to do, might make this impossible.

To better appreciate the significance of Platzky Miller’s suggestion, a really useful analogy may be drawn with the case of so-called ‘colourblind’ approaches to race. The case of race is particularly instructive here since its origins are located in the very same Western colonial project that gave rise to the idea of ‘Western Philosophy.’ (And race’s being grounded in something like a biological natural kind is even more dubious. Biological realism about race is as much tied to the colonial project as it is philosophically incoherent and morally suspect.) To take a colourblind approach to race is, as the name suggests, to act, in theory, as though one does not “see race,” that is, to act as though there were no visible differences between individuals of different races. At first glance, this might seem like an effective way to address racial inequalities, by dispensing with the exclusionary category of “race” altogether. But, as critics of colourblindness have argued, this approach just compounds the problem (see Crenshaw et al. 2019). Pretending that there are no existing inequalities is not in fact an effective route to remedying them. Race must be acknowledged to exist as a concept before it is addressed as a political problem. It would be highly politically suspicious for a white person to insist that since race is not a natural kind, we ought to stop talking about it. And in a similar vein, we think it is important that we continue to talk, critically, about ‘Western Philosophy’ even if, as it turns out, it does not pick out a natural kind.

Just as discussions of race can be used to combat exclusion and racism, there is potential for discussions of ‘Western Philosophy’ to do so too. For example, there is evidence that the concept of ‘Western Philosophy’ can have a positive impact on pedagogy in the history of philosophy. Many institutions have experienced increased student demand for ‘non-Western philosophy.’ This provides departments with an incentive to cover figures who were not white Europeans. This seems like a clear example of the concept ‘Western Philosophy’ being put to use to combat the kinds of exclusionary practices that worry Cantor and Platzky Miller, despite the fact that the same problems about coherence and vagueness apply to the concept of ‘Non-Western Philosophy.’

To draw one last analogy with the existing literature on the analytic tradition, the same concern has been raised with regard to suggestions that the only way to overcome the analytic/continental divide is by abandoning the category of ‘Analytic Philosophy’:

One problem with the claim that there is no such thing as Continental (or analytic) philosophy—especially when made by those of us who have achieved some measure of success in the philosophical profession—is that the claim risks degenerating into the same kind of move as the wealthy, white, heterosexual, cis-gendered, Anglo-male who claims that class, race, gender, sex and ethnicity are of no philosophical importance. . . . Even if one is . . . nobly trying to bring about a future in which such alleged differences are no longer grounds for sociological exclusions within the professions . . ., the problem remains that such futures do not arise simply by choosing henceforth to ignore an established legacy of differences, however arbitrary their historical formation and unphilosophical their contemporary perpetuating might be. (Thomson 2019, 573)

The same applies verbatim to ‘Western Philosophy.’ Acknowledging the existence of a set of influential norms, values, thinkers, research questions, and motivations (including politically suspicious and immoral ones) at least allows one to identify what it is that is suspicious and immoral. Given that Western Philosophers (whoever they are) are unlikely to stop doing Western Philosophy (whatever it is) overnight, there is a further danger that by ceasing to acknowledge its existence, we might invertedly lend credence, once again, to the idea that Western Philosophy just is philosophy. If most academic philosophers continue doing Western Philosophy but stop talking about ‘Western Philosophy,’ it might be easy to slip into thinking that they are just doing philosophy per se. It is precisely the histories of philosophy that avoid using the term ‘Western Philosophy’—but just provide the usual Plato-to-NATO ‘standard narrative’—that are the real culprits.

7 Conclusion

We have argued that the CPM has two dimensions: a descriptive (D) and a normative one (N). We have distinguished three possible interpretations of the descriptive dimension and showed that the first (D1) and third (D3) are false, and the second (D2) is irrelevant to N. We have also distinguished two interpretations of the normative dimension. We have shown that the truth of the first (N1) depends on the truth of D3. Since D3 is false, N1 is false too. By appealing to analogies with the concept of ‘Analytic Philosophy,’ we have also shown that the truth of N2 is independent of D, but that there are nonetheless convincing reasons to believe that it is not true and that historians of philosophy do not need to reject the concept of ‘Western Philosophy’ entirely. In fact, we have suggested, there are important positive reasons not to abandon it.

Acknowledgements

For comments on earlier versions of the paper, we would like to thank Rebecca Bamford, Alexander Douglas, Christoph Schuringa, and participants at the Joint Session of the Aristotelian Society and the Mind Association, University of Birmingham, July 2024.