1 Introduction

There has been much debate in contemporary philosophy of mind on whether Russellian panpsychism, a newly (re-)discovered position of Russellian heritage, is tenable and can be successful in providing a convincing account of mind within a unified conception of nature (Chalmers, 2015; Alter & Nagasawa, 2015) — thus proving better than its more traditional and established competitors (namely, dualism and physicalism above all) (cf. Goff, 2017). As their view relies on the postulation of (phenomenal) categorical properties as the ground of physical reality, one of the major challenges that panpsychists have been presented with in recent years is advanced by dispositionalists — that is, those who believe that (at least some) properties are essentially dispositional and do not need any categories to ground them (Mumford, 2006; Bird, 2007; Ellis, 2013). Now, in her recent production (cf. 2014, 2018, 2020), H. H. Mørch has been arguing that there can be designed an argument for panpsychism which speaks precisely to dispositionalists — even more: an argument that would show that panpsychism actually follows from dispositionalism. In addition to this, she makes a further significant claim: she maintains that, by adopting a particular version of the so-called powerful qualities view (namely, the Identity view of properties), the panpsychist can, at the same time, retain the original subscription to the traditional case for the view (which involves categorical properties), while also endorsing her own new dispositionalism-oriented argument (2014: 55; 2018: 1078; 2020: 281).

This being said, in my paper, I shall address primarily Mørch’s claim that one can hold both panpsychist arguments, if she subscribes to the Identity view (which I shall refer to as Reconciliation), and argue that it fails to obtain — that is, that one cannot achieve the sought reconciliation via the Identity view. Yet, I will show that another view (namely, the Compound view), which is part of the same family of theories of the Identity view (namely, the Powerful qualities view), can indeed carry out the reconciliation that Mørch envisions — I will call such reformulated version of the claim Reconciliation*. However, even though one can reconcile the two panpsychist arguments via the subscription to the Compound view (that is, Reconciliation* is valid), the adoption of such an approach to the metaphysics of properties might render Mørch’s original argument undesirable — for it will expose her project to the risk of epiphenomenalism. This being said, after a careful study of the notions at play, I shall conclude that the particular form of epiphenomenalism that one will have to accept to secure the reconciliation is marginal and certainly well worth the pain. In sum, my proposed version of Reconciliation* (based on the Compound view of properties) will contribute to equip the Russellian panpsychist with an all-encompassing, unitary metaphysical apparatus — which will sustain the view as a positive proposal (and not as a mere alternative to its more established competitors, i.e. dualism and physicalism).

2 Panpsychism and Dispositionalism: Status Quaestionis

Before considering Mørch’s arguments and addressing the reconciliation claim, we ought to say something more on the positions at play in the debate, namely Russellian panpsychism and dispositionalism.

A significant trend in contemporary philosophy of mind focuses on the study of the so-called Hard Problem of Consciousness (cf. Chalmers, 1995). In its essential form, the problem arises from the age-old intuition that mental stuff and physical stuff do not seem to belong together — and thus questions the place of mind within our metaphysical story of reality. Traditionally, two candidates have been put forth to explain the relationship between mind and body, namely dualism, which sees mind and matter as utterly distinct and fundamentally irreducible one to the other (cf. Papineau, 2002); and idealism, which moves from the intuition that the only thing we can be sure about is the reality of mental life (and thus maintains that one should not postulate mind-independent matter). However, both options have significant drawbacks: the dualist picture of the world is radically a fragmented one, while we should aim at a unified conception of nature; the idealist solution is sceptical, while it would be desirable to retain some trust in the observer-independent reality. This being said, in the past century, a third view has gained momentum and has now (arguably) imposed itself as the default position in the analytic circles: physicalism (cf. Field, 1972: 357; Stoljar, 2010: 13). The physicalist roughly believes that the fundamental level of reality is populated only by purely physical entities (cf. Stoljar, 2015) — such that mental phenomena do exist, but are either entirely reducible to physical facts or supervene on the physical. Now, in the past 20 years, due to the rediscovery of Bertrand Russell’s and Arthur Eddington’s early writings (and to the appearance of a number of compelling arguments against physicalism (Jackson, 1982, 1986; Chalmers, 1996, 2009)), a fourth candidate has presented (better: has returned) on the scene and has been receiving increasing scholarly interest: Russellian Monism. In general, Russellian monism can be defined as the view that at the fundamental level of reality there exist some inscrutable properties of a single kind, properties which provide ground for physical entities and consciousness alike (cf. Kind, 2015: 404). In its panpsychist fashion, Russellian monism is often described as having two main components, namely one positive and one negative (Goff, 2017: 17). The negative aspect of the view lies in the claim that physical science provides us only with a partial account of the nature of the physical world (i.e. the extrinsic, dispositional properties of matter), while remaining completely silent about the intrinsic, categorical nature of the material — this is what is supposed to motivate the ‘Russellian’ qualification (cf. Stubenberg, 2015). The positive component consists of the claim that the hidden, intrinsic nature of matter explains phenomenal consciousness — which amounts to say that the ultimate constituents of reality are endowed with phenomenal (or protophenomenal) properties. In sum, Russellian panpsychism is an anti-idealist monistic view which promises to avoid the main difficulties of both dualism and physicalism at once (Chalmers, 2015; Alter & Nagasawa, 2015), while providing a convincing account of mind within a unified conception of nature (Goff, 2017).

On the other hand, in the neighbouring field of contemporary metaphysics, philosophers have been debating on the precise nature of properties. Again, two fronts can be isolated. One is generally known as categoricalism — where the categoricalist believes that all extrinsic/relational/structural properties need some sort of categorical ground (Russell, 1927; Armstrong, 1997; Lowe, 2006; Seager, 2006). The other is often referred to as dispositionalism — where the dispositionalist believes that properties are dispositional, such that properties are described in virtue of what they dispose their bearers to do (Ellis, 2013: 11). More in detail, within the broad spectrum of dispositionalism, two further positions can be identified: a radical position (generally known as ‘pure powers view’), that is, the one of those who claim that the essence of a property is defined by the nomic role it plays, which amounts to say that all properties are essentially dispositional (Shoemaker, 1980; Hawthorne, 2001; Mumford, 2004, 2006; Bird, 2007, 2013); and a moderate position (often called ‘dispositional essentialism’), like the one of those who believe that in order to have dispositional properties like causal powers, some more fundamental (categorical) properties need to be posited (Molnar, 2003; Ellis, 2002, 2013).Footnote 1 This being said, even in the moderate’s case, no random property can serve as the categorical base for powers — such that dispositional essentialism is still different from categoricalism, even though they both make use of categorical properties. In Ellis and Molnar’s case, most of the categorical properties that are admitted are dimensions — where “dimensions include all of the usual spatial and temporal relations. [.. .] [T]he dimensions also include the generic categorical properties of things, including shape, size, orientation, handedness, spatio-temporal interval, and so on” (2013: 18). According to the two dispositional essentialists, then, the only properties that can be rightfully called categorical are those that locate, identify, and orient causal powers in space and time — for, given that causal powers are essentially oriented, without spatio-temporal categories, there could be no causal powers at all (cf. Molnar, 2003: 60; Ellis, 2013: 18).

3 H. H. Mørch’s Defence of Panpsychism

Having introduced sufficient background on the positions that are relevant for our purposes, let us now consider Mørch’s treatment of Russellian panpsychism. This will consist of a new formulation of the traditional argument from categorical properties, the proposition and discussion of her new argument from the experience of causation, and the claim about the reconciliation of the two via the Identity view — I shall address the three points in order.

3.1 The Argument for Panpsychism from Categorical Properties

First of all, let us consider the traditional argument for panpsychism, generally known as the argument from categorical properties (in the form it is presented and discussed by Mørch in her recent production (2020: 280)):

  1. 1.

    Categoricalism: All physical things have categorical properties.

  2. 2.

    Mental categoricity: The only categorical properties whose nature we can know, or positively conceive of, are mental properties.

  3. 3.

    Non-skeptical realism: The nature of the categorical properties of physical things is knowable or positively conceivable.

Therefore,

  1. 4.

    Panpsychism: All physical things have mental properties.

As we can see, the traditional case for (Russellian) panpsychism appeals to the notion of inscrutables, defined as the (categorical) properties that ground the physical structure that physics describes, and claims that these are in fact phenomenal properties (cf. Seager, 2006; Alter & Nagasawa, 2015) — thus elaborating on the (already introduced) view that all dispositions must have categorical grounds (Categoricalism).

3.2 The Argument for Panpsychism from (the Experience of) Causation

As anticipated, pure powers theorists reject the argument from categorical properties, as they believe that the essence of a property is exhausted by its causal role (Mumford, 2004: 95) — which amounts to say that all properties are essentially dispositional. Pure powers theorists thus reject Premise 1 of the argument and hold that dispositional properties are irreducible and do not need any categorical properties to ground them. On the other hand, dispositional essentialists also would reject the argument, as they generally believe that, even though some categorical properties are needed to ground causal powers, not all properties can serve as these categorical bases (and certainly not qualitative properties). In sum, as Mørch aptly remarks, both dispositionalism (i.e. the pure powers view) and dispositional essentialism constitute a challenge to the argument from categorical properties (2014: 54; 2018: 1077).

Now, instead of arguing against dispositionalism, Mørch aims to appeal to our acquaintance with the nature of causation in agency to develop an argument for panpsychism that is consistent with the dispositionalist view — showing that, in fact, dispositionalism entails panpsychism. Her argument from the experience of causation runs as follows (2020: 276):

  1. I

    Non-reductionism: All physical things have causal powers.

  2. II

    Mental causation: The only causal powers whose nature we can know, or positively conceive of, are mental.

  3. III

    Non-skeptical realism: The nature of the causal powers of physical things is knowable, or positively conceivable.

Therefore,

  1. IV

    Panpsychism: All physical things have mental properties.

The argument from causation starts with the claim that all physical entities have irreducibly dispositional properties — which is in line with both the pure powers theory and dispositional essentialism, and thus (for our purposes) does not need further defence.

As for Mental causation, which constitutes the most central (and controversial) premise of the argument, Mørch’s strategy is two-fold. First of all, she needs to establish that there are such things as mental causal powers. To this extent, she subscribes to the ‘inconceivability principle for causation’, such that causation implies a necessary connection between cause and effect which renders it inconceivable that causes do not bring about their effects (and they do so in virtue of their essential character) (Mørch, 2014: 102–103; 2018: 1081; 2020: 276). Then, she goes on to consider the relationship between motives and efforts (such as the relationship between ‘feeling pain’ and ‘trying to avoid it’) and claims that such relationship (vis-à-vis the one of voluntary physical action) might qualify as a good candidate to meet the standards of causation that are previously set out via the principle of inconceivability — in sum, Mørch contends that, if no other power interferes, “it is hard to conceive of someone experiencing strong pain, but where this does not make them at least try to avoid it” (2020: 278). To the conclusion that the property ‘pain’ makes its bearer ‘try to avoid’ it in virtue of its nature, that is, of the intrinsic character of the property.Footnote 2 Secondly, to defend the latter part of Premise II (that is, we cannot know or positively conceive of non-mental causal powers) Mørch claims that, in principle, one could ‘abstract away’ from the concept of mental causal power a general concept of (non-mental) causal power, which might then be used to account for the physical causal powers that physics tells us about (2018: 1084). However, by carrying out such a process of abstraction, Mørch argues, it is not clear whether we would be left with any positive conception of power (cf. 2014: 114; 2018: 1084–1086; 2020: 278–279) — such that, if we lose the phenomenal aspect of them, it seems that the only thing that we know about non-mental causal powers is that they are “just like mental powers, but non-mental” (2020: 279). Thus, this leaves us with the conclusion that any experience that does not present causation as mental, and any account of causation that involves non-mental powers, is either deceptive or inaccurate.

Finally, Premise III is partly defended on the basis of Premise II. Mørch claims that if there are some causal powers whose intrinsic character we know from our phenomenology of motivation and will (that is, mental powers) then we should not posit further unknown kinds of powers, provided that our account of mental powers already adequately explains our observations. This can be defended by appealing to Leibniz’s principle of the continuity of nature, to the qualitative principle of parsimony, or to the general methodological principle that we should not reject a perfectly adequate positive account of causation on the basis of the negative theory that there may be unknowable powers.

In sum, it is clear that Mørch’s entire argument hangs on the intuition that in our phenomenology of will and motivation we truly experience causation (conceived of on the basis of the inconceivability principle) — and thus on the claim that (at least some) mental properties are essentially dispositional (2020: 277).

3.3 Russellian Panpsychism as a positive proposal: Reconciliation

Given that the “proponents of the argument from categorical properties often argue that all mental properties are categorical” and if “categorical and dispositional properties are indeed opposites” (Mørch, 2020: 280), the argument from causation might seem to clash with the argument from categorical properties. After all, as already mentioned, the advocates of the argument from categorical properties will deny Premise II of the argument from causation, as they would not accept that mental properties are dispositional in nature; while pure powers theorists reject Premise 1 of the argument from categorical properties, as they believe that there are no categorical properties at all, (and dispositional essentialists deny Premise 3 of the same argument, as they tend to resist the idea that phenomenal properties can be categorical).

In this direction, in order to accept both arguments, Mørch proposes, following Marting & Heil (1999) and Strawson (2008), to drop the assumption that categorical and dispositional properties are opposite fundamental kinds, and subscribe instead to the Identity view of the metaphysics of properties (Identity view, for short). In her recent production, Mørch defines the Identity view as the theory that “all properties are necessarily both dispositional and categorical, as opposed to purely one or the other” (2018: 1079), and that “all properties necessarily have both categorical and dispositional aspects, and that categorical and dispositional properties are actually identical” (2020: 281) — we can thus understand the Identity view as follows:

Identity view: the categoricality and the dispositionality of the property are identical, and they exhaust the intrinsic nature of the property — such that, categoricality, dispositionality, and the property are all identical.

Thus, she maintains that, by subscribing to the Identity view, one can elegantly reconcile the two panpsychist arguments without contradiction (2014: 55; 2018: 1078; 2020: 281) — we can thus render the reconciliatory attempt as follows:

Reconciliation: if one subscribes to the Identity view, she can hold both the argument from categorical properties and the argument from causation without contradiction.

The idea behind Reconciliation is that, if we embrace the Identity theory, then we will be able to accept both Premise II of the argument from causation and Premise 1 of the argument from categorical properties; while if we reject the Identity theory and stick to the pure powers view, we will be restricted to the argument of causation and the argument from categorical properties will remain unacceptable. Therefore, if Reconciliation were true, we would be able to provide a unitary defence for panpsychism, which would run roughly as follows:

  • I. Non-reductionism: All physical things have causal powers.

  • 1. Categoricalism: All physical things have categorical properties.

  • II. Mental causation: The only causal powers whose nature we can know, or positively conceive of, are mental.

  • 2. Mental categoricity: The only categorical properties whose nature we can know, or positively conceive of, are mental properties.

  • III./ 3. Non-skeptical realism*: The nature of the causal powers and of the categorical properties of physical things is knowable, or positively conceivable.

Therefore,

  • IV./ 4. Panpsychism: All physical things have mental properties.

Before moving on, let me emphasise the significance of the enterprise of reconciling the two panpsychist arguments into one single, all-encompassing case in support of the position. The challenge here is not only to build an argument for panpsychism which may appeal to dispositionalists (who are traditionally powerfully opposed to panpsychism, due to their beliefs concerning the metaphysics of properties); rather, the point is to widen and strengthen the resources that the panpsychist (who is already convinced by the traditional argument) has to support her view. Thus understood, the upshot of Reconciliation is essentially to give to the traditional panpsychist the possibility to subscribe to a new, strong argument to support her view, without forcing her to give up on the resources that she already has (namely, the argument from categorical properties) — thus producing a comprehensive case in favour of her view.

This appears to be especially relevant if we consider the dialectic of the current debate on the Hard Problem of Consciousness. Even in panpsychism-sympathetic circles, there might be detected a certain tendency to regard the view primarily as a less problematic alternative to the more established (but problematic) dualist and physicalist options; rather than as a positive proposal, which is desirable per se (that is, in isolation from the two competing views) — this trend has been recently registered by Maung (2019). It is no coincidence that, when it comes to presenting their view, Russellian advocates first resort to conceivability arguments against physicalism (showing that the view cannot withstand the challenge), and then claim that panpsychism is a better and more cogent theory, as it is able to maintain the fundamentality of mentality (cf., e.g., Seager, 1995; Chalmers, 2015: 249–252) — thus taking the conceivability argument (and the consequent failure of physicalism) to constitute the key motivation for even considering panpsychism (Maung, 2019: 159).Footnote 3 In this direction, it seems clear that constituting a unitary panpsychist front (which would retain the traditional case for the view and subsume Mørch’s new argument) represents a pivotal objective for the Russellian advocate — and would signal a clear step change for the hopes of the view in the philosophical arena.

4 H. Taylor’s Challenge to the Identity View

In a recent paper, H. Taylor (2018) has challenged the common belief that the Identity theory and the pure powers theory are distinct views — just like Mørch, he also takes the Identity theory to be the view that “the property, the dispositionality and the qualitativity of the property are all identical” (2018: 1424), following Martin (1997: 216), Marting & Heil (1999: 46–47), and Strawson (2008). Thus, Taylor argues that if one takes a close look at their proposed accounts of dispositions and qualities, the advocates of the Identity theory and the pure powers theorists turn out to share the very same views on properties — meaning that the two theories “are not distinct” (2018: 1438). To this extent, he holds that this follows from the understanding that the two positions have of the metaphysics of properties, and in particular from a) their analysis of what qualities are, b) their stance on what qualities are not, and c) a shared commitment to the complete powerfulness of properties — let us consider the claims in order.

First of all, Taylor argues that, on most accounts of what a quality is, a) both pure powers and Identity theorists will agree that properties are qualities. Heil and Martin characterise qualities as the “ways things are” (Heil, 2010, 2012: 59), and being qualitative as being “really ‘there’ in the object” (Martin, 1996: 74) (cf. also Jacobs, 2011). The same account of properties is held by the powers theorists: dispositionalists such as Molnar (2003: 99) say that properties are ‘really’ in the object — for example, fragility is an ‘actual’ property of the glass, it is ‘really there’ in the glass. An example might help (cf. Jaworski, 2016). When we consider a diamond, and we focus on its categorical features, we can say that the diamond is essentially hard, or that it has a tetrahedral arrangement of carbon atoms essentially — such that in every possible world a diamond is hard, and has a tetrahedral arrangement of carbon atoms. Conceived of in this way, categorical properties are those features of objects which are always and essentially possessed by the objects in question, independently of the circumstances: they are really there, in the object. Now, if being ‘really there in the object’ and being ‘the way things are’, are the individuating traits of categorical properties, then powers, as conceived of by the pure powers theorists, will also be fully fledged categorical properties, such that, e.g., fragility is always and essentially a property of fragile objects — it is just the way fragile objects are: fragile.

Secondly, Taylor does not only claim that the two views agree on what qualities are and conceive of properties in the same way, but also suggests that b) the Identity theorists and the pure powers theorists both reject David Armstrong’s characterisation of categorical properties. Armstrong defines categorical properties as non-dispositional properties, claiming that qualities and powers are “just different” (2005: 315). However, Identity theorists, Taylor claims, do not characterise categorical properties as non-dispositional, as Armstrong does, but rather they understand the qualitative in terms of “real, actual features of objects” (2018: 1430) — as per a). Consider again the diamond example. When we look at a diamond, we can describe it in various ways: we can say that it is hard, that it would be able to scratch glass it is was pressed onto it, and that it has tetrahedral arrangement of carbon atoms. Now, according to the Identity theorist, these three descriptions are “of numerically one and the same property”, such that “the diamond’s hardness = the diamond’s power to scratch glass = the diamond’s having a tetrahedral arrangement of carbon atoms” (Jaworski, 2016: 54). This is to say that, under the Identity view of properties, categorical properties (such as the diamond’s hardness or its being constituted by tetrahedral arrangement of carbon atoms) and dispositional properties (the diamond’s power to scratch glass), are one and the same thing, just considered under different aspects — thus they are not numerically distinct, separate properties (or parts of properties).

Finally, pure powers theorists believe that the whole nature of a property is powerful, that is, that there is no aspect or part of a property which is not powerful. But Identity theorists must also be committed to this, as they believe that “properties are identical with powers” (Taylor, 2018: 1434) and that dispositionality and categoricality are one and the same thing. Thus, c) both views endorse a characterisation of the nature of properties as entirely powerful, such that there is no part or aspect of a property that is independent of its dispositional nature.

To sum up, Identity theorists and Pure powers advocates share the same view of the metaphysics of properties, such that they both accept

Taylorian metaphysics of properties (TMP): Properties are powers, and they are ‘qualities’ in the sense of real, actual features of the object, i.e. features that are always and essentially there in the object — such that the whole nature of the property is powerful, that is, properties have no nature independent of their dispositional nature.

While they both deny

Armstrongian metaphysics of properties (AMP): Properties are powers, and they are ‘qualities’ in the Armstrongian sense of ‘qualities’, that is as ‘non-dispositional properties’ — such that, it is not the case that the whole nature of the property is powerful, that is, properties have some nature that is independent of their dispositional nature.

5 Taylorian metaphysics of properties and Reconciliation

On the one hand, Mørch claims that only by adopting the Identity view, following Marting & Heil (1999) and Strawson (2008), can one subscribe to both the argument from causation and the argument from categorical properties (i.e. Reconciliation) — while the pure powers theorist must view the argument from categorical properties as unacceptable. On the other hand, it follows from Taylor’s argument that the Identity theorist has the exact same resources as the pure powers theorist — as they both accept TMP, and they both deny AMP. Prima facie, then, it seems that, if Taylor is right, Reconciliation (for how it is defended by Mørch in 2014, 2018, 2020) must be false — as it is not the case that, via the commitment to the Identity view, one can hold both the argument from causation and the argument from categorical properties. Indeed, if Taylor is right and the Identity view is identical to the pure powers view, then the Identity theorist should not be able to accept the argument from categorical properties (just like the pure powers advocate) — and thus Reconciliation fails. Let us investigate the matter in more detail and determine whether this initial indication is verified.

It seems that if one buys Taylor’s conception of categorical properties as (merely) the “ways things are” (Heil, 2010, 2012: 59), thus attributing categoricality also to dispositional properties (i.e. TMP), one can make sense of Premise 1 of the argument from categorical properties even from the radical dispositionalist’s standpoint. With this first premise, Mørch wants to defend the view that all extrinsic/relational/structural properties need some sort of intrinsic ground (i.e. Categoricalism) — otherwise, physical reality would end up being populated exclusively by abstract mathematical entities (i.e. the ‘Pythagorean reductio’, Mørch, 2018: 1076; cf. also Lowe, 2006; Armstrong, 1997). As already mentioned in Sections 2 and 3.2, the commitment to Categoricalism is traditionally taken as the main point of contention between pure powers theorists, who claim that properties are essentially and entirely constituted by the nomic role they play (Bird, 2007; 2012); and Russellian monists, who believe in (phenomenal) categorical properties that ground all the physical structure of the natural world. In a word, the pure powers advocate believes that there are only dispositional properties while the Russellian monist also allows for categorical properties (and claims these to be phenomenal). Now, if one takes the view that categoricality and dispositionality are, in fact, identical (as in TMP), then it seems that the dispositionalist can agree that all physical structure is instantiated by (individuals with) categorical properties, on the grounds that all dispositional properties have a categorical aspect of some sort — and this would seem to vindicate Premise 1 of the argument from categorical properties.

However, even if this kind of Identity theory approach (i.e. TMP) saves the first premise of the traditional argument, it threatens the second one. As it is presented in Mørch’s most recent works (2018, 2020), Premise 2 of the argument from categorical properties (Mental categoricity) states that “[t]he only categorical properties whose nature we can know, or positively conceive of, are mental properties” (Mørch, 2020: 280. Emphasis added.). Now, Mørch does not offer an extensive defence of Mental categoricity in either 2018 or 2020, but she does so in 2014. Indeed, in 2014, the main argument that Mørch provides for the premise draws on the basic Russellian claim that science tells us only about the logico-mathematical structure of physical reality and remains silent with regard to the intrinsic nature of the physical world — a claim generally known as ‘Structuralism about physics’ (Russell, 1927; Seager, 2006: 136; Alter & Nagasawa, 2015: 425; Mørch, 2014: 28).Footnote 4 This is to say that, while it fully describes the dispositional structure of the world, physical science does not tell us anything about the categorical properties that ground all reality. But if one buys the Identity view (i.e. TMP), then each and every property is at the same time dispositional and categorical — and indeed categoricality, dispositionality, and the property at issue are all identical. Now, if physical science provides us with a full account of the nomic structure of reality, and dispositionality and categoricality are identical, it follows with necessity that physics indeed provides us with a full account of categoricality. And again, as dispositionality and categoricality are also identical with the property at issue, according to the Identity theory, physical science would be perfectly able to disclose everything there is to know about that property. But clearly the latter conclusion is at odds with the basic metaphysical outline of Russellian monism, given that it implies that the (alleged) inscrutables are perfectly well captured by physical science. Further, it implies that scientific enquiry will reveal the categorical aspect of physical properties and of mental properties in just the same way — which means that, under this account, mental properties will not have any kind of privilege in terms of categoricality. To the conclusion that, if the panpsychist adopts the conception of properties which is at the core of the Identity theory (i.e. TMP) (which is the same one of the pure powers view, as per Taylor’s argument), she will be forced to deny Mental categoricity (i.e. Premise 2 of the argument from categorical properties), and thus she will be forced to abandon her subscription to the traditional case for Russellian panpsychism — thus denying Reconciliation.Footnote 5

To sum up, embracing the Identity view’s understanding of categorical properties (i.e. TMP) would compel the aspirant Russellian monist to deny Premise 2 of the traditional argument from categorical properties (Mental categoricity), as it would not be the case that mental properties are the only categorical properties whose nature we can know, or positively conceive of. Thus, in light of TMP, we can say that Reconciliation, as it is defended by Mørch in her recent production, is false — as it is not the case that, by adopting the Identity view, one can hold both the argument from causation and the argument from categorical properties.Footnote 6

6 Salvaging Reconciliation: The Compound View

As we have seen, if Taylor is right about the Identity view’s and the pure powers view’s shared understanding of the metaphysics of properties (i.e. TMP), Reconciliation (as Mørch presents it in her recent production) is false — as the adoption of the Identity view is incompatible with the traditional case for Russellian panpsychism. Now, I believe there might be found a way to salvage Reconciliation — and that is via the adoption of a view which is quite closely related (but not identical) to Mørch’s Identity view, namely, the Compound view.

6.1 The Compound view and Reconciliation*

We have seen how the Armstrong-inspired view of categorical properties as non-dispositional (i.e. AMP) is incompatible with the Identity view that they are completely powerful (i.e. TMP). And, in all fairness, Mørch seems to be resistant to the idea of conceiving of categorical properties as opposed to dispositional properties (cf. 2020: 280) — that is, to AMP. However, I believe that, by focussing on what I previously called the Armstrongian metaphysics of properties, we can elaborate a view which will prove successful where the Identity view failed, that is, in reconciling the argument from categorical properties with the argument from causation.

This being said, if we elaborate on AMP, we get to an understanding of the nature of properties such that

Compound view: each property has separate, purely dispositional and purely categorical parts — such that each property has categorical parts essentially, and dispositional parts essentially, but neither the categoricality nor the dispositionality exhaust the full nature of the property.

To illustrate how the Compound advocate conceives of properties, one can picture the dial of a watch (also known as ‘face’): half of the dial is the semi-circular section that goes from the 12 h marker to the 6 h marker (from now on, just 12 and 6), the other half is the section from 6 to 12, and the two parts together form a compound, namely, the dial — which cannot be conceived of without any of the two sections (that is, take away the 6–12 section, and your watch does not have a face anymore, and the same goes for the 12–6 section). Accordingly, under the Compound view, each and every property is divided into two separate parts, one is purely dispositional, the other is purely categorical, and these two separate parts, taken together (i.e. the compound of the two parts) make up the intrinsic character of the property at issue — such that both parts are essential to the property, but none of them exhausts the nature of the property.

Now, the compound view is clearly different from the Identity view (that is, the view based on TMP) as it holds that categoricality and dispositionality are not identical — it also denies the complete powerfulness claim (that is, no part of a property is non-dispositional) (i.e. c)). However, the view is still different from categoricalism, as it claims that a property has some dispositional parts essentially, and also from the pure powers view, as it claims that a property has some categorical parts essentially. Having introduced the Compound view, let us now reformulate Reconciliation accordingly:

Reconciliation*: if one subscribes to the Compound view, she can hold both the argument from categorical properties and the argument from causation without contradiction.

Now, by claiming that properties have both irreducibly dispositional and categorical parts (that is, the Compound view), one can retain both Premise 1 of the traditional argument, as it is true that every property has a categorical part essentially, and Premise I of the argument from causation, as it is true that every property has a dispositional part essentially. Further, prima facie, it also seems that, by adopting the Compound view, one can hold both Premise II of the argument from causation, and retain the reasoning behind Premise 2 of the traditional argument — and thus Reconciliation* seems to be vindicated. Yet, things are not as straightforward as they might appear.

6.2 The Compound view and the Threat of Epiphenomenalism

In Section 5, we have seen that Reconciliation fails because the Identity view is incompatible with the traditional case for Russellian panpsychism. Thus, in Section 6.1, I have advanced a new formulation of the claim (i.e. Reconciliation*), that is, basing the reconciliation attempt not on the Identity view (which elaborates on TMP) but on the Compound view (which endorses AMP). Now, even though the Compound view does not encounter the same difficulties of the Identity view (that is, it is not incompatible with Premise 2), it might still entail some undesirable consequences for Mørch’s argument from causation — and, in particular, for Premise II. Specifically, adopting the Compound view might make Mørch’s argument from causation entail epiphenomenalism — which would (arguably) make the subscription to the argument altogether undesirable.

Let us consider again Mørch’s reasoning for Mental dispositionality, that is, Premise II of the argument from causation. She claims that the relationship that holds between ‘feeling pain’ and ‘trying to avoid it’, a paradigmatic example from our phenomenology of will, can qualify as a genuine relation of causal necessity (which is defined via the principle of inconceivability) — such that it seems inconceivable for a subject in pain not to try to avoid the pain (ceteris paribus). In particular, Mørch contends that the fact that pain makes us try to avoid it seems to be due to the nature of the pain, that is, to the intrinsic character of the property (i.e. pain).Footnote 7

Let us now consider how Premise II may fare when the Compound view is applied to the view. (Before moving on, let us bear in mind that, according to the Compound advocate, the intrinsic character of the property ‘pain’ is constituted by two parts, one purely dispositional, and one purely categorical (i.e. the ‘feeling of pain’) — such that, each property is essentially dispositional and essentially categorical at the same time.)

At first glance, the claim that it is the intrinsic nature of pain that disposes us to try to avoid it, such that phenomenal properties have an essentially dispositional nature, might seem controversial. For instance, Goff believes that if there were a dispositional aspect to the intrinsic character of phenomenal properties, this should have been introspectively apparent (2018: 1092). Indeed, he maintains that there seems to be nothing incoherent with the claim that “there could be certain odd creatures whose pain produces attraction compulsions and whose pleasure produces avoidance compulsions” (Goff, 2018: 1090) — such that Mørch’s proposed relation (between ‘feeling pain’ and ‘trying to avoid it’) does not meet the standards of the inconceivability principle for causation. In this direction, Goff suggests that one way Mørch can salvage her proposal is to say that “there are some non-phenomenal features of pain that, as it were, activate its causal capacity” (2018: 1091). Now, this is precisely what the Compound view allows her to do. In fact, the Compound view lends support to Mørch’s claim that it is the intrinsic nature of pain that is responsible for the bearer’s trying to avoid it, since it allows that the intrinsic character of pain is composed of two separate parts, one categorical and one dispositional, only the latter of which disposes us to try to avoid pain (along with Goff’s suggestion). Again, we should bear in mind that the main difference between how the Identity advocate and the Compound theorist conceive of properties (that is, the contrast between TMP and AMP) is that, while in the Identity view the whole property is powerful (such that there is no part of the property which is independent of the dispositional nature), in the Compound view this is not the case — and the categorical part of the property is causally inert.

This being said, if by positing a purely dispositional part of the intrinsic character of properties (which activates their causal capacities), Mørch’s proposal (read as an instance of the Compound view) might respond to Goff’s scepticism (and respect the inconceivability principle of causation), it also brings about a major problem. In fact, if it is only the purely dispositional part of the intrinsic character of the property that activates the causal capacities of the property (in the case of pain, the ‘trying to avoid’), this now means that the qualitative (that is, the purely categorical) part of pain is irrelevant to our trying to avoid it — as the causal work is entirely done by the dispositional part of the property. On this account, then, one might worry that we would try to avoid the pain even if its phenomenology were quite different from what it is, or even if there were no ‘what it is like’ to pain at all. Again, this follows from the fact that, under the Compound view, it is not the case that the whole property is powerful (vis-à-vis the Identity view); rather, by definition, the purely categorical part of the property is causally inefficacious. To the conclusion that it is indeed inconceivable, as Mørch wants to say, that when we bear the property ‘pain’ we do not ‘try to avoid it’, but this is not due to the qualitative part of the property (i.e. the ‘feeling of pain’), but to the dispositional part of the property.

If this is reasonable, prima facie, it is hard to see how one can avoid the threat of epiphenomenalism: if the dispositional parts of (the intrinsic character of) properties are sufficient to fully explain the physical facts, then it is hard to see why phenomenality is not just ‘along for the ride’.

In this direction, Robert Howell (2015) has similarly argued that Russellian Monism (in general) entails epiphenomenalism. He argues that “even if phenomenal properties cause things on the Russellian Monism picture, they do not cause things in virtue of their phenomenal nature” (Howell, 2015: 28). Let us go through Howell’s modified causal exclusion argument (2015: 32) — which, interestingly, seems to be assuming (without acknowledging it) the Compound view as the default metaphysical position for Russellian Monism:

  1. A.

    There are two distinct and separable aspects of Russellian properties, those that ground phenomenal resemblance relations and those that ground resemblances between causal profiles;

  2. B.

    All physical events have sufficient causes in virtue of those aspects that ground resemblances between the causal profiles of RM properties.

Therefore,

  1. C.

    The aspects of RM properties that ground phenomenal resemblances make no unique causal contribution to the physical world.

Now, Howell’s Premise A seems perfectly in line with the general overview of the Compound view, that is, that properties have separate, purely dispositional and purely categorical parts, with their respective prerogatives. Then, Premise B outlines what I have previously argued, that is, that, under the Compound view, properties dispose their bearers not in virtue of the totality of their intrinsic character (bearing in mind that Compound properties are not wholly powerful), but in virtue of the purely dispositional parts of their intrinsic character — such that ‘pain’ disposes its bearer to ‘try to avoid’ not in virtue of the ‘feeling of pain’, but in virtue of its purely dispositional part. Finally, if Premise A and Premise B obtain (as they do in Mørch’s proposal read as an instance of the Compound view), then the phenomenal parts of properties (e.g. the ‘feeling of pain’ for the property ‘pain’) are entirely inefficacious towards the physical world (i.e. C) — thus qualifying as epiphenomenal.

In conclusion, we have seen how Reconciliation is entirely off the table (as the Identity view is incompatible with the traditional case for Russellian panpsychism), while Reconciliation*, on the other hand, is not incoherent. However, it seems that if we pursue Reconciliation*, that is, if we try to reconcile the traditional argument from categorical properties and the argument from causation by adopting the Compound view, we might end up with a (prima facie rather undesirable) picture where the feeling of pain (i.e. the purely categorical part of the property) is completely inefficacious in disposing the bearers to try to avoid pain — in a word, a picture where phenomenal (parts of) properties are epiphenomenal.

6.3 Two Strategies against Epiphenomenalism

We have seen how Reconciliation* may constitute an undesirable solution for the Russellian panpsychist, as, under the Compound view, phenomenal qualities turn out to be epiphenomenal. Now, the challenge is to determine whether there exist some strategies that can be adopted by the panpsychist to retain Reconciliation* (and thus the Compound view), without incurring epiphenomenalism. In this direction, Alter and Coleman (2020: 233) argue that there are essentially two ways in which the Russellian monist can reply to the epiphenomenalist objection. These are the necessitarian strategy and the compatibilist strategy — let us now consider the strategies and evaluate whether the advocate of Reconciliation* is metaphysically allowed to pursue them.

First of all, the necessitarian strategy consists of denying that the two aspects (read: parts) of a property, as posited by Howell and by the Compound view advocate “are modally separable in the way the modified exclusion argument requires” (Alter & Coleman, 2020: 236). Specifically, the necessitarian Russellian panpsychist argues that metaphysical necessitation from the phenomenality to dispositionality ‘does not exhaust the picture of the intimate relationship that ties the two aspects together’, and claims that the metaphysical necessitation functions also in the opposite direction. It is clear that a metaphysical necessitation from the dispositional aspect to the phenomenal aspect of a property dodges Howell’s modified exclusion argument and thus avoids epiphenomenalism — but it comes with a high price, as Howell himself remarks (2015: 36–37). In fact, positing such a necessitation rules out all the conceivability, zombie-style arguments against physicalism: if there is an entailment from dispositional to phenomenal, then a zombie world is not metaphysically possible. In sum, I do not think that a such conceived picture (where the purely dispositional part of the property metaphysically necessitates the purely categorical one) is one that the traditionalist Russellian panpsychist would easily accept.Footnote 8 (Moreover, I suspect that postulating a metaphysical necessitation from the dispositional aspect to the phenomenal one would make the position ultimately collapse into the already discussed Identity view — with all the related difficulties.)Footnote 9

Secondly, the compatibilist can resist the inference from two different categorical properties playing the same grounding role — “either across or within worlds” (Alter & Coleman, 2020: 235) — to the conclusion that they are not causally efficacious. To do this, the compatibilist can appeal to the contingency of the laws of nature, and say that these laws fix which properties do the grounding work in which worlds — and thus the contribution that categorical properties make to physical causation. However, if the advocate of Reconciliation* (and thus the Compound view advocate) were to adopt this compatibilist strategy, then she would inevitably distance herself from the dispositionalists — and from Mørch’s argument from causation. In fact, the compatibilist strategy explicitly builds upon a Humean picture of reality (where causal regularities are viewed as contingent), while dispositionalism, in both its radical (i.e. pure powers view) and moderate (i.e. dispositional essentialism) form, is generally opposed to Humeanism. In this direction, if the advocate of Reconciliation* adopted the compatibilist strategy, she will be (arguably) forced to renounce Premise I of the argument from causation, which is accepted by dispositionalists but is strongly at odds with the regularity theory (Hume, 1739–40; Lewis, 1973) — and, for this very reason, Mørch herself rejects the Humean account of the laws of nature in defending her view (2017: 302; 2018: 1073; 2020: 276–277).

6.4 Is Reconciliation* Worth Epiphenomenalism?

We have seen in the previous section that, if the advocate of the Compound view adopts the necessitarian strategy (to avoid epiphenomenalism), she will distance herself from the traditional Russellian panpsychist, as she will be forced to posit a metaphysical necessitation from the dispositional part to the phenomenal part of the property — and thus Reconciliation* fails. On the other hand, if she goes for the compatibilist strategy, she will distance herself from the dispositionalist (and thus from the argument from causation), as she will be forced to accept that causal powers are not in things but are determined by laws of nature — and thus Reconciliation* fails. Therefore, there seems to be no obvious way the advocate of Reconciliation* (i.e. the Compound theorist) can avoid epiphenomenalism. At this stage, then, one question arises: given that Reconciliation is not viable (that is, the approach based on the TMP), is the entailment of epiphenomenalism so undesirable to make the panpsychist renounce Reconciliation* — and thus abandon the enterprise of reconciling the two panpsychist arguments altogether? To this, my answer is: no, she should not abandon Reconciliation* — there are two reasons for this, one conceptual and another dialectical.

First of all, one might point out that the specific form of epiphenomenalism that is seen to follow from Reconciliation* is not a particularly concerning one. In particular, given that the phenomenal parts of properties may still be said to do some grounding work, the fact that they are not strictly causally efficacious should not be too concerning. That is to say that, while the qualitative parts are certainly epiphenomenal (as they are causally inert and ‘epiphenomenalism’ is a causally-relevant notion), it still follows from the view that the property ‘pain’ disposes its bearers to ‘trying to avoid’ in virtue of its intrinsic nature (where the qualitative part ‘feeling of pain’ is inefficacious but grounds the overall property ‘pain’ while the dispositional part, in turn, is responsible for the causal work). An example might help illustrate the point. When we explain why gravity diminishes as 1 over distance squared, we note the mathematical fact that the surface area of a sphere grows as the square of radius. Now, this is not a causal power of spheres, but it does affect causal powers — or better, it grounds how they change. Thus, the surface area of the sphere is not just ‘along for the ride’.

In sum, the phenomenal part of the property is responsible for grounding the property, while the dispositional part of the property is responsible for activating the causal capacity of the property — thus both parts are essential to the property, and both parts play equally relevant roles (even though different roles) towards the contribution of the property to the structure of physical reality. Again, this does not mean that the phenomenal parts of properties are not epiphenomenal, for they are indeed (as they are causally inert); it means that this specific kind of epiphenomenalism is rather a benign one — phenomenal parts still contribute to the structure of physical reality (by grounding the properties they are parts of).

Secondly, a dialectical reason. As we have seen in Section 3.3, the argumentative importance of the reconciliation attempt should not be underestimated: if we managed to find a way to allow the panpsychist to subscribe to a unitary position that would integrate the traditional case from categorical properties and Mørch’s original dispositionalism-oriented argument, the overall case for panpsychism would be much strengthened — also in view of the alternative answers to the Hard Problem of Consciousness (as presented in Section 2). This is to say that, even though the entailment of (even a rather innocuous form of) epiphenomenalism might be perceived as expensive (from a metaphysical point of view), I believe it is well worth the price.

7 Concluding Remarks

The main objective of this paper has been to question Mørch’s claim that the Identity theory allows us to subscribe both to the traditional argument for panpsychism and also to her own argument from causation — which I referred to as Reconciliation. To this extent, I argued that if one takes the Identity view of properties, which is built around an understanding of properties which I called the Taylorian metaphysics of properties (TMP) (on the basis of Taylor, 2018), then the panpsychist will have to renounce the traditional case for her view — as the resulting conception of the categorical properties is incompatible with Premise 3 of the argument from categorical properties. Then, I proposed to elaborate on an Armstrong-inspired conception of categorical properties (i.e. the Armstrongian metaphysics of properties (AMP)), which led to the consideration of the Compound view of properties. In this direction, I have demonstrated that if one buys the Compound view, one can reformulate the reconciliation strategy (i.e. Reconciliation*) such that the objective can be achieved — that is, one can consistently hold both panpsychist arguments without contradiction. Yet, even though Reconciliation* has been proven successful in its aims, it might entail the (apparently undesirable) consequence that the phenomenal parts of properties are epiphenomenal. However, given that specific kind of epiphenomenalism that Reconciliation* brings about is not a threatening one (and seen the relevance of the overall enterprise), I concluded that one may well bite the bullet. Therefore, if my reasoning is correct, via the Compound view of properties, a strong, unitary panpsychist position is constituted — one by which Panpsychism can be defended as a positive proposal towards the solution of the Hard Problem of Consciousness (and not merely as an alternative to dualism and physicalism).