Russellian monism is a theory in the metaphysics of mind, on which asingle set of properties underlies both consciousness and the mostbasic entities posited by physics. The theory is named for BertrandRussell, whose views about consciousness and its place in nature wereinformed by a structuralist conception of theoretical physics. On sucha structuralist conception, physics describes the world in terms ofits spatiotemporal structure and dynamics (changes within thatstructure) and says nothing about what, if anything, underlies thatstructure and dynamics. For example, as it is sometimes put, physicsdescribes what mass and chargedo, e.g., how they disposeobjects to move toward or away from each other, but not what mass andchargeare. Thus, Russell writes the following about theevents physics describes:
All that physics gives us is certain equations giving abstractproperties of their changes. But as to what it is that changes, andwhat it changes from and to—as to this, physics is silent.(Russell 1959: 18)
Russellian monism can be seen as breaking that silence. It positsproperties that underlie the structure and dynamics that physicsdescribes. Further, according to Russellian monism, those sameproperties are relevant to, and may at least partly constitute,consciousness.
Russellian monism can be seen as combining three core theses:structuralism about physics, which states roughly thatphysics, at least current physics (see section 4.1), describes theworld only in terms of its spatiotemporal structure and dynamics;realism about quiddities, which states that there arequiddities, that is, properties that underlie the structureand dynamics physics describes; andquidditism aboutconsciousness, which states that quiddities are relevant toconsciousness. We will discuss these theses in turn.
To understand the first core thesis, structuralism about physics,consider David J. Chalmers’s description of how physical theorycharacterizes its basic entities:
…physical theory only characterizes its basic entitiesrelationally, in terms of their causal and other relations toother entities. Basic particles, for instance, are largelycharacterized in terms of their propensity to interact with otherparticles. Their mass and charge is specified, to be sure, but allthat a specification of mass ultimately comes to is a propensity to beaccelerated by certain forces, and so on. Each entity is characterizedby its relation to other entities, and so on forever. …Thepicture of the physical world that this yields is that of a giantcausal flux, but the picture tells us nothing about what all thiscausationrelates. Reference to the proton is fixed as thething that causes interactions of a certain kind, that combines incertain ways with other entities, and so on; but what is the thingthat is doing the causing and combining? As Russell (1927a) notes,this is a matter about which physical theory is silent. (Chalmers1996: 153)
Chalmers’s description expresses structuralism about physicsvividly. But his claim that physical theory characterizes its basicentities only relationally could be challenged. Shape is usuallycounted among an object’s non-relational properties (Stoljar2015). For example, contrast a ball’s sphericality, anon-relational property, with its tendency to roll when pushed, arelational property. So, if physical theory characterizes its basicentities partly in terms of their shapes, then Chalmers’s claimmight seem doubtful.
Such considerations need not be taken to indicate that structuralismabout physics is false. But they do seem to indicate that the relevantdistinction between structural and non-structural properties—theone relevant to structuralism about physics—does not align with,and thus should not be explicated in terms of, therelational/non-relational distinction. (Henceforth we omit mention ofdynamics because the relevant issues concern structure.)
Structuralism about physics is sometimes formulated in terms of thedistinction between extrinsic and intrinsic properties (Chalmers 1996:153–155; see the entryintrinsic vs. extrinsic properties). But those formulations have been challenged on similar grounds: aball’s sphericality is usually counted among its intrinsicproperties, as are other properties in terms of which physical theorycharacterizes its basic entities (Ney 2015; Stoljar 2015). (We willhenceforth use ‘extrinsic’ and ‘relational’interchangeably, and likewise for ‘intrinsic’ and‘non-relational’.) It might therefore be preferable toexplicate the structural/non-structural distinction in a differentway.
One approach is to do so by distinguishing betweenrelativelyandabsolutely intrinsic properties (Pereboom 2011:92–97, 2014, 2015, 2016; for an alternative explication, seeChalmers 2010: 20, fn. 17; 2013: 256; cf. Alter 2016: 804–09).Roughly, relatively intrinsic properties are intrinsic properties thatare nothing over and above extrinsic properties, whereas absolutelyintrinsic properties are intrinsic properties that are not relativelyintrinsic. (The relative/absolute intrinsicness distinction derivesfrom Leibniz and Kant; see Leibniz to De Volder April 1702 [Leibniz GII, 240; 1969: 526–527]; Kant 1781/1787: A277/B333; Pereboom1985: 413–23; 1991b; Van Cleve 1988; Langton 1998.)
We can articulate that distinction more precisely in terms ofmetaphysical grounding (or, on some conceptions, constitution; see theentriesmetaphysical grounding andmaterial constitution) whereAgrounds (orconstitutes)B just in caseA metaphysically necessitatesB andB exists in virtue ofA, as follows(Pereboom 2016):
P is arelatively intrinsic property ofXjust in caseP is an intrinsic property ofX andP is grounded in extrinsic properties of eitherX orparts ofX.
By contrast,
P is anabsolutely intrinsic property ofXjust in caseP is an intrinsic property ofX andP is not grounded in extrinsic properties of eitherX or parts ofX.
Thus, for example, it seems plausible that a ball’s sphericalityis grounded in spatiotemporal relations among its parts (as Leibnizargued; Leibniz to De Volder April 1702 [G II, 240; 1969:526–527]; seesection 2.1 below) and is therefore merely relatively intrinsic. A similar pointapplies to other intrinsic properties described in physics. Forexample, although mass would seem to be an intrinsic property ofmaterial entities, in the passage quoted above Chalmers remarks thatin physical theory, “all that a specification of mass ultimatelycomes to is a propensity to be accelerated by certain forces, and soon” (Chalmers 1996: 153). Accordingly, the only aspects of massthat physical theory describes are relatively, not absolutely,intrinsic.
Suppose structuralism about physics is explicated such that structuralproperties are construed as extrinsic properties or relativelyintrinsic properties. In that case, the thesis is consistent with theclaim that physical theory characterizes its basic entities partly interms of its intrinsic properties, as long as those intrinsicproperties are relatively and not absolutely intrinsic. (For furtherdiscussion of structuralism about physics, seesection 4.1 below; and the entriesstructuralism in physics andstructural realism.)
The second core Russellian monist thesis, realism about quiddities,states that there are (instantiated) properties of precisely the sortabout which, according to structuralism about physics, physical theoryis silent: properties that underlie the spatiotemporal structurephysical theory describes. Those properties categorically ground themost basic physical dispositions that physics describes, in the way aball’s spherical shape categorically grounds its disposition toroll when pushed. These underlying properties are often calledquiddities (Lewis 2009; Chalmers 2012). They are also calledinscrutables (Montero 2010). The latter designation is meantto indicate something on which many proponents of realism aboutquiddities agree: we know little about quiddities beyond thetheoretical roles they are supposed to play.
One might reject realism about quiddities even if one acceptsstructuralism about physics. As Chalmers (2013: 254) writes,“There are respectable structuralist or dispostionalist views ofphysics on which physics involves just structure or dispositions allthe way down” (e.g., Ladyman and Ross, with Spurrett and Collier2007; Ney 2015; cf. Shoemaker 1980; Hawthorne 2001; McKitrick 2003).But Russellian monists and others (e.g., Lewis 2009) regard suchstructuralist or dispositionalist views as implausible. That isbecause such views “seem to yield a world devoid of substance orqualities” (Chalmers 2013 [2015: 254]). A world devoid ofsubstance or qualities might or might not be possible. But accordingto realism about quiddities, the actual world is not like that.
The third core Russellian monist thesis, quidditism aboutconsciousness, states that quiddities are relevant to consciousness.More specifically, it states that quiddities are relevant tophenomenal consciousness. For example, consider the feelingone has when suffering from a painful migraine or when seeing abrilliantly red sunset. As it is often put, there issomething itis like to have a phenomenally conscious experience (Nagel 1974;see the entryconsciousness; throughout, all references to consciousness refer to phenomenalconsciousness). There is no consensus among Russellian monists onexactly how quiddities are relevant to consciousness (section 4). But many claim that phenomenal properties areconstituted byquiddities, or by quiddities together with various structuralproperties.
One might reject quidditism about consciousness even if one acceptsstructuralism about physics and realism about quiddities. For example,one might combine the latter two theses with a dualist theory on whichquiddities categorically ground the most basic physical dispositionsphysics describes but are not relevant to consciousness (perhaps onthis theory consciousness is fundamental and not constituted byanything). By contrast, according to quidditism about consciousness,quiddities play both of those roles.
One could identify Russellian monism with the conjunction of all threecore theses, where each thesis could be explicated in different ways,resulting in different varieties of the theory. For present purposes,that characterization will suffice (see below, sections 1.2 and 4; foran alternative characterization, see Pereboom 2011: 89; cf. Alter andNagasawa 2012: 71–72.)
While all Russellian monists maintain that quiddities are relevant toconsciousness (section 1.1), some construe quiddities themselves as phenomenal properties (Maxwell1978; Lockwood 1989, 1992; Rosenberg 2004; Strawson 2003, 2006a,b;Mørch 2014; Goff 2015, 2017; cf. Unger 2005). The result is avariety of Russellian monism known asRussellian panpsychism(Chalmers 2013 [2015: 246–247]). Russellian panpsychism is socalled due to the assumption that phenomenality is ubiquitous: itoccurs everywhere the most basic physical dispositions do. But as aterminological matter, one might allow that a Russellian monist theorycounts as Russellian panpsychism if it specifies either thatatleast some of the most basic physical dispositions are underlainby phenomenal quiddities (Chalmers 2013 [2015: 246]) or thatall such dispositions are so underlain (Strawson2006a,b).)
Panpsychism has a long philosophical history (see the entrypanpsychism; Skrbina 2005), but some find the doctrine incredible (Searle 1997;but see Chalmers 1997b; Alter and Nagasawa 2012: 90). Here it shouldbe noted that Russellian panpsychists typically do not construequiddities asmacrophenomenal properties, that is, asfamiliar phenomenal properties such as those commonly associated withfeeling pain and seeing red. More often, Russellian panpsychistsconstrue quiddities asmicrophenomenal properties, that is,as phenomenal properties of microphysical entities. Microphenomenalproperties might differ radically from the macrophenomenal sort(Chalmers 1996: 293–297; Strawson 2003, 2006a; Rosenberg 2004:95). Some Russellian panpsychists construe quiddities ascosmophenomenal properties, that is, phenomenal properties ofthe whole cosmos, which might likewise differ radically frommacrophenomenal properties (Goff 2017). So, the sorts of phenomenalproperties that, according to Russellian panpsychism, underlie themost basic physical dispositions need not be much like any with whichwe are familiar. This might make panpsychism more palatable to somephilosophers—though, some argue, if such properties differenough from familiar macrophenomenal properties, one might doubt thatthey deserve to be called phenomenal at all (Kind 2006).
Another option for Russellian monists is to construe quiddities aswhat Chalmers callsprotophenomenal properties, which hecharacterizes as follows:
…protophenomenal properties are special propertiesthat are not phenomenal (there is nothing it is like to have a singleprotophenomenal property) but that can collectively constitutephenomenal properties, perhaps when arranged in the right structure.(Chalmers 2013 [2015: 259])
The result is a variety of Russellian monism known asRussellianpanprotopsychism. Russellian panprotopsychism does not entailpanpsychism. Perhaps only entities in a limited class (e.g., somestates of some animals) have protophenomenal properties that arearranged in a consciousness-constituting structure. In that case, onRussellian panprotopsychism consciousness would occur only in thoseentities and not elsewhere. There are also hybrid varieties ofRussellian monism, on which some quiddities are phenomenal and othersare protophenomenal (Holman 2008).
Russellian monism is a distinctive theory in the metaphysics of mind.Nevertheless, it is possible to identify varieties that correspond tomore traditional theories, including physicalism, idealism, neutralmonism, and (despite the name) even dualism (see the entriesphysicalism;idealism;neutral monism; anddualism). On physicalist Russellian monism, both quiddities and structuralproperties are physical (Stoljar 2001; Pereboom 2011, 2014, 2015,2019; Coleman 2012, 2015; Montero 2015; Morris 2016; Brown 2017). Onidealist Russellian monism, both quiddities and structural propertiesare mental (Bolender, 2001; Adams 2007; Schneider 2018; cf. Chalmers1996: 155). On neutral monist Russellian monism, both quiddities andstructural properties are neutral, where neutral properties areneither physical nor mental but underlie both physical and mentalproperties (Nagel 1986, 1998; cf. Stoljar 2015; Wishon 2016; also seethe discussion ofpanqualityism in the entrypanpsychism). Finally, Russellian monism could be combined with the traditional(e.g., Cartesian) dualist view that the concrete world includes twofundamentally distinct sorts of entities, the physical sort and themental sort. This can be done by, for example, identifying structuralproperties with physical properties and quiddities with certain mentalproperties (cf. Chalmers 2010: 135).
Physicalist and neutral monist varieties of Russellian monism are mostnaturally understood as forms of Russellian panprotopsychism (Alterand Coleman 2020b; but see Morris 2016, Brown 2017), whereas idealistand dualist varieties are most naturally understood as forms ofRussellian panpsychism (Chalmers 2013, Schneider 2018). Much of therecent literature on Russellian monism focuses on the physicalistvariety, often calledRussellian physicalism (Montero 2015).Russellian physicalism is usually presented as an alternative to,rather than a version of, traditional physicalism. But that might seemmisleading. On many traditional physicalist theories, there arenon-phenomenal properties that can collectively constitute phenomenalproperties, perhaps when arranged in the rightstructure—microphysical properties, say. On those theories, suchproperties would qualify as protophenomenal properties by thecharacterization quoted above from Chalmers 2013 (Papineau 2002:22–23, fn. 5).
Some traditional physicalist theories conflict with other aspects ofRussellian monism. For example, (physicalist) analytic functionalism(Armstrong 1968; Lewis 1966, 1972, 1980) conflicts with the Russellianmonist’s denial that structural truths alonea priorientail all phenomenal truths (see the entryfunctionalism;phenomenal truths are truths about consciousness). But othertraditional physicalist theories are not similarly disqualified. Forexample,a posteriori physicalists also deny that structuraltruths alonea priori entail all phenomenal truths (see theentryphysicalism). Nevertheless, in their view, there are non-phenomenal properties thatcan collectively constitute phenomenal properties, perhaps whenarranged in the right structure. So, one might wonder whyaposteriori physicalism does not qualify as Russellianphysicalism.
Chalmers recognizes this problem. He therefore makes two furtherstipulations designed to help distinguish Russellian physicalism fromtraditional physicalist theories:
…(i) protophenomenal properties are distinct from structuralproperties and (ii) there is ana priori entailment fromtruths about protophenomenal properties (perhaps along with structuralproperties) to truths about the phenomenal properties that theyconstitute. (Chalmers 2013 [2015: 260])
Given those further stipulations,a posteriori physicalistswould deny that there are protophenomenal properties. Similarconsiderations apply to other traditional physicalist theories.Russellian physicalism is different. (For other ways to distinguishamong varieties of Russellian monism, seesection 4; Alter and Nagasawa 2012; Chalmers 2013; Goff 2017.)
One could find views that resemble Russellian monism in certainrespects throughout the history of philosophy. For example,panpsychism is very old indeed (and not distinctive to the west). Butit is hard to find views approximating full-fledged Russellian monismbefore the early modern period. This is not a coincidence: the theoryis closely tied to modern physics, which was first developed duringthat era. Yet it did not take long for the theory, or something akinto it, to emerge. Arguably, Russellian monism can be ascribed tovarious early modern figures, including at least Gottfried WilhelmLeibniz, Immanuel Kant, and Arthur Schopenhauer.
Leibniz’s Russellian monist theory may be found in his critiqueof René Descartes’ theory of matter (Pereboom 2011:92–97; cf. 1991a,b). In Descartes’ theory, materialsubstance is essentially just extension in three spatial dimensions(DescartesPrincipia Philosophiæ, “Part Two: ThePrinciples of Material Things” 1644 [1984: v. 1, 223–247;AT VIII 40–79]; see the entryDescartes’ physics). Leibniz contends that Descartes’ theory is unsatisfactorybecause, first, the extension of a physical object is (in theterminology introduced insection 1.1, which is Kantian (Kant 1781/1787: A277/B333)) a relatively intrinsicproperty of it, since it resolves into the plurality, spatialcontinuity, and temporal coexistence of its parts, and the extensionof those parts resolves into those extrinsic properties of theirparts,ad infinitum; and, second, any real thing, that is,any actual, concrete entity, cannot possess only extrinsic orrelatively intrinsic properties: it must possess an absolutelyintrinsic property as well. Leibniz writes: “there is nodenomination so extrinsic that it does not have an intrinsicdenomination at its basis” (Leibniz to De Volder April 1702[Leibniz G II, 240; 1969: 526–527]). In his view, a conceptionof the physical world that does not include absolutely intrinsicproperties is in an important sense incomplete. In effect, hiscomplaint is that Descartes’ theory of matter describes onlystructure without describing a quiddistic ground for that structure,or even acknowledging the possibility of or the need for such aground.
In Leibniz’s view, the absolutely intrinsic propertiesunderlying real things are at least connected toprimitiveforce (see the entryLeibniz’s philosophy of physics). Primitive force (which can appear to us phenomenally as physicalforce) generates a progression from one representation to the next ina series, in conformity with a law (Leibniz 1698 [1969: 504; 1989:162–63]; Adams 1994: 352; Jorati 2018). On one interpretation,primitive forces are the fundamental elements, “the soleinhabitants of the ground floor” of Leibniz’s ontology(Jorati 2018); thus, the most basic substances, mind-like monads,“do not have forces, strictly speaking—they areforces” (Jorati 2018; cf. Adams 1994: 265, 378–379). Onthat reading, primitive forcesare the absolutely intrinsicproperties underlying real things. They might be called Leibnizianquiddities.
Leibniz might be the first Russellian monist. In his view, all theproperties physics presents are either extrinsic or relativelyintrinsic. This corresponds to structuralism about physics. Hecontends there are absolutely intrinsic properties, primitive forceson one reading, in which those extrinsic properties and relativelyintrinsic properties are grounded. This corresponds to realism aboutquiddities. Those quiddities ground all of the other features ofreality, including consciousness generally and macrophenomenalproperties specifically. This corresponds to quidditism aboutconsciousness. Thus, Leibniz seems to endorse versions of all threecore Russellian monist theses. And because on his view all basicentities, including Leibnizian quiddities and the extrinsic andrelatively intrinsic properties they ground, are mental, LeibnizianRussellian monism is a version of Russellian idealism.
Like Leibniz, Kant can be seen as a Russellian monist. Kant endorses aversion of structuralism about physics: “All that we cognize inmatter are nothing but relations. What we call the intrinsicdeterminations of it are intrinsic only in a relativesense…” (1781/1787: A285/B341; Pereboom 1985:413–23, 1991a,b, 2011: 100–101; Van Cleve 1988; Langton1998; cf. Holden 2004: 236–263). In material objects we discoveronly extrinsic properties and relatively intrinsic properties, neverany properties that are absolutely intrinsic. Indeed, for Kant, matteritself does not have absolutely intrinsic properties. That is becausehe regards matter as “mere appearance”. Thus, he writes,“what it [matter] itself consists in is the mere relation ofsomething in general to the senses” (1781/1787: A285/B341). Ifmatter were instead a “thing-in-itself”, he implies, thenit would need to have absolutely intrinsic properties. Nevertheless,in his view all features of appearance, including both matter andmacrophenomenal consciousness, are grounded in things-in-themselves(or a thing-in-itself) and thus in absolutely intrinsicproperties—a position that suggests both realism aboutquiddities and quidditism about consciousness.
Although Kant agrees with Leibniz that there must be absolutelyintrinsic properties for the world to exist as we know it, he differswith Leibniz about forces. In Kant’s view, forces are relationaland thus not absolutely intrinsic (1781/1787: A265/B321; 1786:498–491). He does not say whether there are Leibnizian primitiveforces. And he denies that we know much about the nature of theabsolutely intrinsic properties that, he thinks, must exist. Suchknowledge would be tantamount to substantive knowledge ofthings-in-themselves, and famously he contends that we lack suchknowledge. He could therefore be described as endorsing anepistemological doctrine that many contemporary Russellian monistsendorse (section 1.1): we know little about quiddities beyond the theoretical roles they aresupposed to play.
Schopenhauer can be seen as another pre-twentieth century Russellianmonist. InThe World as Will and Representation, he affirmswhat might be described as versions of structuralism about physics andrealism about quiddities. Natural causal explanation, he writes,
…really does nothing more than indicate the orderly arrangementaccording to which the states of matter appear in space andtime… But it affords us absolutely no information about theintrinsic nature (das innere Wesen) of any one of thesephenomena. (1818: Vol. 1, Bk. II, sect. 17, par. 5 [1961: 113])
And he implies that those phenomena do indeed have intrinsic natures,which suggests quidditism about consciousness.
So far, Schopenhauer’s view is similar to Kant’s. ButSchopenhauer does something Kant would not do. He makes a furtherpositive proposal about the nature of the absolutely intrinsicproperties that underlie the orderly arrangement indicated by naturalcausal explanations. Specifically, in his viewthe will playsthe key role: “indeed the answer to the riddle is given to thesubject of knowledge who appears as an individual, and the answer iswill” (1818: Vol. 1, Bk. II, sect. 18, par. 1 [1961: 115]). Theabsolutely intrinsic properties in which the orderly arrangement ofthe world of appearance is grounded are volitional—and hence thetitle of his famous bookThe World as Will and Representation(Schopenhauer 1818: Vol. 1, Bk. II, sect. 18, par. 1 [1961: 115]; seethe entryArthur Schopenhauer).
Some find Russellian monism in works by others in the modern period aswell. Rebecca Copenhaver (2020) argues that George Berkeley can beseen as a Russellian idealist. Cole Mitchell (in correspondence)reports that there is a version of Russellian panprotopsychismdiscussed by Anthony Collins, in the 1707–08 Clarke-CollinsCorrespondence.
Contemporary Russellian monists often trace their view toRussell’s 1927The Analysis of Matter, which they readas developing a structuralist account of physics. Indeed, some find inthat book suggestions of not only structuralism about physics but theother main components of Russellian monism too (Lockwood 1992 [2015:144–145]; Chalmers 1996: 166). On this interpretation, Russellthere implied that what are now called quiddities might be identifiedwith what he calledpercepts (Russell 1927a: 402). It hasalso been argued that he more explicitly endorsed the eponymous theory(though not by its current name) in some of his later writings,includingHuman Knowledge: Its Scope and Limits andMyPhilosophical Development (Wishon 2015; but see Stubenberg2015).
Some argue that Russellian monism has historical roots not only inRussell’s writings but also in early twentieth century works byArthur Eddington (Kadić 2017; cf. Strawson 2006a). InEddington’s view, science does not describe what he calls thenature of the entities it posits, which he also refers to asthebackground of “pointer readings and similarindications” (Eddington 1928: 252). For instance, he contends,physics—modern physics, as opposed to the physics of the“Victorian physicist” for whom “Atoms were tinybilliard balls…”—describes atoms not in terms oftheir natures but rather in terms of pointer readings (Eddington 1928:259). Thus, like Russell, Eddington seems to affirm a version ofstructuralism about physics (or something in the vicinity). Further,Eddington proposes that atoms, and perhaps everything else physicsdescribes, might have “something of spiritual nature of which aprominent characteristic isthought” (Eddington 1928:259). In support of that proposal, he writes,
But in one case—namely, for the pointer readings of my ownbrain—I have an insight which is not limited to the evidence ofthe pointer readings. That insight shows that they are attached to abackground of consciousness. Although I may expect that the backgroundof other pointer readings in physics is of a nature continuous withthat revealed to me in this particular case, I do not suppose that italways has the more specialized attributes of consciousness.(Eddington 1928: 259–260)
Eddington’s reflections can be seen as suggestive of eitherRussellian panpsychism (Kadić 2017: 46), Russellianpanprotopsychism, or perhaps a hybrid of the two.
In the late twentieth century, Grover Maxwell (1978) and MichaelLockwood (1989, 1992) each endorse a theory that they attribute toRussell and is often interpreted as Russellian monism (see, forexample, Chalmers 1996, 2013). Maxwell presents his theory in thecourse of defending the mind-brain identity theory (Place 1956; Smart1959; Lewis 1966; see the entrythe mind/brain identity theory) against Saul Kripke’s (1972) influential challenge.Kripke’s challenge concerns how to reconcile that identitytheory with the “apparent contingency of the connection betweenthe mental state and the corresponding brain state…”(Kripke 1972 [1980: 152]). Maxwell rejects an assumption on which, heclaims, that challenge relies: the assumption that “we know fromcommon sense, from physics, from neurophysiology, etc., what brainevents are like” (Maxwell 1978: 134). In his view, those sourcestell us about structural properties but not about brain events’underlying non-structural properties—a line of reasoningsuggestive of both structuralism about physics and realism aboutquiddities. He also seems to endorse something like quidditism aboutconsciousness. In the case of brain events that are, in his view,identical to macrophenomenal experiences, he implies that theirunderlying non-structural properties might be phenomenalproperties—a claim suggestive of Russellian panpsychism.Lockwood (1989, 1992) develops a theory similar to Maxwell’s,remarking that it “seems to me the only approach to thephilosophical mind–body problem, currently on offer, that holdsout the slightest promise” (Lockwood 1992 [2015: 145]).
More recent discussions of Russellian monism have followed Maxwell andLockwood’s lead. The view is advanced as offering a plausibleresponse not only to Kripke’s challenge but also to relatedanti-materialist arguments (section 3). Daniel Stoljar’s “Two Conceptions of the Physical”(2001) is widely considered the classic expression of that idea (seealso Stoljar 2006, 2009, 2014, 2015). The idea is also developed byChalmers (1996, 1997a, 2003, 2010: ch. 6, 10) and others (Rosenberg2004; Alter 2009, 2016; Montero 2010, 2015; Pereboom 2011, 2014, 2015,2016; McClelland 2013; Brown 2017; Goff 2017). Chalmers (1997a)introduced the name “Russellian monism,” and his work onthe topic has been highly influential. Another main source of thecontemporary interest in Russellian monism is Galen Strawson’s(2003, 2006a,b) seminal work on consciousness and physicalism.
Russellian monists usually support their theory by arguing that it hassignificant theoretical benefits. We will describe three sucharguments, which appeal to closely related considerations. (For otherarguments for Russellian monism see, for example, Rosenberg 2004;Strawson 2006a; Goff 2017; Schneider 2018.)
The first argument is based on the claim that Russellian monismcombines theoretical virtues of traditional dualism with those oftraditional physicalism, while avoiding their main drawbacks (Chalmers2013). The reasoning behind that claim could be stated as follows.Traditional physicalism has the virtue of allowing consciousness to beintegrated into physical causation. But this comes at the cost ofdisregarding or distorting the distinctive features of consciousness.Traditional dualism avoids that cost. But by separating consciousnessfrom the physical world, traditional dualism prevents consciousnessfrom being integrated into physical causation. The latter cost isfamously articulated by Princess Elisabeth of Bohemia, in hercriticism of Cartesian dualism (Elisabeth to Descartes, 21 May, 1643[1984: v. 3, 218; AT III 665]; see the entryElisabeth, Princess of Bohemia), and by related causal arguments for physicalism such as thefollowing:
Many effects that we attribute to conscious causes have full physicalcauses. But it would be absurd to suppose that these effects arecaused twice over. So the conscious causes must be identical to somepart of those physical causes. (Papineau 2002: 17; cf. Kim 1989,1998).
By contrast, Russellian monism seems to avoid those costs of bothtraditional physicalism and traditional dualism. On Russellian monism,the properties that constitute consciousness—quiddities and(perhaps) structural properties—play indispensable roles inphysical causation. Consciousness —or at least the components ofconsciousness —is thus integrated into physical causation. Yetthe distinctive features of consciousness are neither disregarded nordistorted. Those features are attributed to quiddities and theirspecial (e.g., constitutive) relationship to consciousness.
A second argument for Russellian monism is based on the claim thatthis theory offers an elegant, unified solution to two distinctphilosophical problems (Lockwood 1989, 1992; Chalmers 1996, 2013[2015: 254]; Rosenberg 2004; Goff 2017; cf. Russell 1927a, 1927b:116). One of those problems is how to provide a foundation for thespatiotemporal structure physics describes. The other is how tointegrate consciousness into physical causation. When considered fromthe perspective of a Russellian monist, these two problems seem madefor each other. The requisite foundation of spatiotemporal structureis provided by (proto)phenomenal quiddities. Because those samequiddities (partly or wholly) constitute consciousness, consciousnessthereby plays a distinctive role in physical causation—or atleast the components of consciousness play that role. Thus, thereasoning runs, both problems are solved at once.
A third argument for Russellian monism is based on the claim thatRussellian monism provides a plausible response to anti-materialistarguments such as the conceivability argument and the knowledgeargument (see the entrieszombies andqualia: the knowledge argument and Alter 2023). The conceivability argument is often formulated interms of azombie world (Chalmers 1996): a physical andfunctional duplicate of the actual world but without consciousness. Itis first argued that a zombie world is ideally conceivable, whichmeans roughly that such a world cannot be ruled out as incoherent bya priori reasoning (Chalmers 2002). It is then argued thatthe ideal conceivability of such a world entails its metaphysicalpossibility. Finally, it is argued that the metaphysical possibilityof such a world entails that physicalism is false. The knowledgeargument begins with Frank Jackson’s (1982, 1986, 1995) case ofMary, the brilliant scientist in the black-and-white room, who comesto know the complete physical truth by reading black-and-white sciencebooks and watching science lectures on a black-and-white monitor. Ifphysicalism were true, Jackson reasons, then Mary would thereby knoweverything there is to know about seeing in color: she could deducethat information from her physical knowledge. But, he claims, this isnot the case. On the contrary, when Mary leaves the room she learnstruths about what it is like to see in color – truths she didnot already know.
According to Russellian monism, those arguments might show thatstructure alone does not fully determine (metaphysically necessitate)the nature or even the existence of consciousness. But it does notfollow that physicalism is false. At most, it follows thattraditional physicalist theories are false, because thosetheories limit the physicalist’s resources to the structural.Russellian physicalism, on which the class of physical truthsincludes not only structural truths but quiddistic truths as well, isnot threatened (Stoljar 2001). Further, this anti-structuralist thesiscomports well with the claim that consciousness consists at leastpartly in non-structural quiddities—a claim that is central toRussellian monism generally, not just the physicalist variety (section 1).
That response can be developed in different ways, such as rejectingepistemic premises on which the anti-materialist arguments rely (forother ways, see Chalmers 2010: 134–135; Pereboom 2011; Alter andNagasawa 2012: 83–86). Regarding the conceivability argument,Russellian physicalists can argue as follows. On reflection, a zombieworld is not ideally conceivable. If such a world initially appearsconceivable, this is because we mistake a zombie world for astructural zombie world: a consciousness-free world that(minimally) duplicates all of the actual world’s structuralfeatures. If there is an ideally conceivable world in the vicinity,then that world is a structural zombie world. There is no good reasonto accept the ideal conceivability of a full-fledged zombie world,which would lack consciousness despite duplicating not only the actualworld’s structural features but its quiddistic features as well.Regarding the knowledge argument, Russellian physicalists can arguethat because Mary’s black-and-white science lessons leave outquiddistic information, those lessons teach her only part of thephysical truth. If so, then the truths she learns when she leavesmight be physical. They would be non-structural quiddistic truths butphysical truths nonetheless.
One objection to Russellian monism concernsthe combinationproblem: how do (proto)phenomenal quiddities (or quiddities andstructural properties) combine to constitute macrophenomenalproperties (Seager 1995; Chalmers 2003)? Suppose someone is having aphenomenally blue experience. Ifconstitutive Russellianmonism is true, then the phenomenal blueness of that experienceconsists wholly in certain quiddities, perhaps structured in a certainway (for the distinction between constitutive and non-constitutiveRussellian monism, see Section 4.2). But on many varieties ofRussellian monism, quiddities are micro-level features. On the face ofit, it seems possible that those (or any) micro-level quiddities couldbe instantiated without anyone having that (or any)experience—and that seems possible no matter how they arestructured. If that is possible, then how could phenomenal bluenessconsist wholly in (structured) quiddities, as constitutive Russellianmonists hold? Essentially the same problem seems to arise for othervarieties of constitutive Russellian monism, such as those on whichquiddities are construed as cosmophenomenal features (Goff 2009;Pereboom 2011, 114; Chalmers 2013, 2017).
That is one version of the combination problem, but there are others.Consider one raised for a version of constitutive Russellianpanspsychism, according to which microphenomenal quiddities, perhapstogether with structural properties, combine to constitutemacrophenomenal properties. The problem can be stated as an objection,as follows. The constitutive Russellian panpsychist is committed tothe following two claims. First, corresponding to each instance of amicrophenomenal quiddity underlying a physical disposition there is aunique conscious subject: a microsubject distinct from themicrosubjects that correspond to other such instances ofmicrophenomenal quiddities. Second, multiple microsubjectscorresponding to microphenomenal quiddities combine to constitute asingle macrosubject. But it is unclear how that second claim could betrue. As William James (1890: 162) suggests, given any group ofsubjects and any further subject, it seems possible for the group toexist without that further subject existing. This is knownthesubject-summing problem (Goff 2009, 2017; Chalmers 2013, 2017;Coleman 2014).
Some respond to such combination problems by appealing to a Leibnizianunderrepresentation thesis (Pereboom 2011: 114–115), on whichintrospective awareness of phenomenal states fails to represent themas having microfeatures they do in fact have (Leibniz,Discourseon Metaphysics, §30 [G IV, 458–59; IV, 574–575];Pereboom 1991b, 2011: 9, n. 1; cf. Stoljar 2001: 276). Consider theversion of the combination problem stated two paragraphs back,involving the constitution of phenomenal blueness. On the Leibnizianunderrepresentation thesis, even if one introspects one’sexperience as featuring onlynoncomplex phenomenal blueness,phenomenal blueness might in fact be constituted by an unrepresentedarray of (proto)phenomenal quiddities. To support that claim, onemight draw an analogy to a macrophenomenal property that isconstituted by othermacrophenomenal properties. For example,a taste experience might appear simple and unconstituted even if infact it is constituted by a combination of sweet, sour, salty, bitter,and umami (as Louis DeRosset has suggested in conversation; seePereboom 2011: 115). Likewise, the response continues, perhaps allmacrophenomenal features are constituted by (proto)phenomenalquiddities, or by quiddities and structural features, whichintrospection fails to reveal.
However, that response is only partial. It might show thatintrospection alone does not rule out the possibility thatmacrophenomenal properties are constituted by quiddities or byquiddities and structure. But the response does not explain howmacrophenomenal properties might result from such constituents.Although some explanations have been proposed (e.g., Rosenberg 2004,2015, Mørch 2019), many regard the combination problem forRussellian monism as unsolved and serious (see the entrypanpsychism; Brüntrup and Jaskolla 2017; Chalmers 2017).
Several other objections to Russellian monism have been raised in therecent literature, of which we will describe four. The first isrelated to the combination problem and applies only to Russellianpanprotopsychism. It is argued that Russellian panprotopsychism facesthe following sort of “revenge” problem. Russellian monismis motivated partly by its ability to provide plausible responses toinfluential anti-physicalist arguments, such as the conceivability andknowledge arguments (section 3.1). Those arguments usually begin by attacking allegedly implausibleepistemic commitments of certain traditional physicalist views (suchas analytic functionalism), such as the claim that structural truthsalonea priori entail all phenomenal truths. However, someargue, parallel considerations undermine the Russellianpanprotopsychist’s claim that protophenomenal truths andstructural truths togethera priori entail all phenomenaltruths (Strawson 2006a,b; Goff 2015, 2017). Thus, for example, it isargued that while still in the black-and-white room Mary can come toknow all the protophenomenal truths just as easily as she can come toknow all the structural truths, and still not thereby be positioned toknow what it is like to see red (for related objections to Russellianmonism, see Cutter 2019, Botin 2023, and Mendelow forthcoming).
In response, some deny that the twoa priori entailmentclaims are on a par (Alter 2018). Mary could come to know anystructural truth while still in the room, the response runs, becauseany structural truth can be fully conveyed in the colorless languageof objective science. But protophenomenal properties arenon-structural. So, perhaps some protophenomenal truths cannot befully conveyed in such colorless language. If so, then the Russellianpanprotopsychist’sa priori entailment claim might notmeet the same fate as the one associated with certain traditionalphysicalist theories.
A second recent objection, developed by Amy Kind (2015), is thatRussellian monism is unmotivated. The theory is often advanced asproviding insight into the mind–body problem: insight that haseluded traditional dualists and traditional physicalists (section 3.1). However, Kind points out, Russellian monism fails to resolve the mostcentral issue in the debate: whether consciousness is a fundamentalcomponent of the universe. On Russellian panpsychism consciousness isa fundamental component of the universe, whereas on Russellianpanprotopsychism it is not. Thus, Russellian monism leaves us“essentially back where we started” (Kind 2015: 420; cf.Liu 2021, which compares the view to Thomistic hylomorphism and claimsthat Russellian monism has similar problems). In response, it has beenargued that this charge misunderstands Russellian monism’s aims(Alter and Coleman 2020a). Russellian monism does not purport tosettle whether consciousness is a fundamental component of theuniverse. Rather, the theory is meant to provide a framework in whichthat issue (and others) can be more productively addressed—aframework provided by neither traditional physicalism nor traditionaldualism.
A third recent objection is that Russellian monism, despite its aims,fails to adequately integrate consciousness into physical causation.One version of this objection begins with the premise that, onRussellian monism, a single type of physical disposition might havebeen grounded by a distinct quiddity from the quiddity that actuallygrounds it (Howell 2015; cf. Robinson 1993). For example, if in theactual world quiddity Q1 grounds negative charge, then there is ametaphysically possible world in which a distinct quiddity Q2 groundsnegative charge; the grounding relationship between quiddities andphysical dispositions is in this way metaphysically contingent. If so,the argument continues, then quiddities (even if they exist) do notmake any distinctive contribution to physical causation. With respectto physical causation, quiddities are “just along for the causalride” (Howell 2015: 34). On another version of the objection,even if quiddities contribute to physical causation, consciousnessdoes not thereby inherit any physical efficacy, contrary to whatRussellian monists claim (Robinson 2018; cf. Chan 2020, 2021).
In response to the first version of that objection, on whichquiddities lack physical efficacy, some reject the assumption that anysuch conclusion follows from the premise that a single type ofphysical disposition might have been grounded by a distinct quiddityfrom the quiddity that actually grounds it (Hawthorne 2002, Alter andColeman 2020a, 2021). Causal relations, it is argued, do not ingenerally hold with metaphysical necessity. So, why must therelationship between quiddities and dispositions hold with suchnecessity, if quiddities are to be physically efficacious? Anotherresponse is to reject the premise that on Russellian monism thatrelationship is metaphysically contingent (Chalmers 2013 [2015:264–265]; Mørch 2014, 2019; Gundersen 2015; Kadić2017; Alter and Coleman 2020a, 2021). In response to the secondversion of the objection, on which consciousness does not inherit anyphysical efficacy from that of quiddities (even if quiddities partlyor wholly constitute consciousness), one might argue that this chargeis just an instance of a general worry concerning micro-to-macrocausal exclusion, which does not concern consciousness or Russellianmonism specifically. (See the entrymental causation; Goff 2017: 153–158)
A fourth recent objection to Russellian monism targets the distinctionbetween the structural and the non-structural. Some argue that thisdistinction cannot be explicated in the way that Russellian monismrequires. As noted above (section 1.1), some have made proposals for how to explicate that distinction. Thoseproposals render structuralism about physics consistent with the claimphysics describes some intrinsic properties, thus safeguardingstructuralism about physics from one line of attack. But whether anyof those proposals is fully adequate for the purposes of Russellianmonism is disputed (Stoljar 2015). Other, related objections have beenraised, e.g., that the dispositional/categorical distinction does notapply to physical properties in the way that Russellian monism implies(Hiddleston 2019; but see Alter and Pereboom 2023; for otherobjections concerning the Russellian monist’sstructural/non-structural distinction, see Stoljar 2006, 2009, 2014,2015; Ney 2015; for responses, see Pereboom 2011, 2014, 2015; Alter2009, 2016).
As noted above (section 1.1), Russellian monism can be identified with the conjunction of the threetheses: structuralism about physics, realism about quiddities, andquidditism about consciousness. But one might question whetherRussellian monists are committed to structuralism about physics, atleast if that thesis is understood as saying that physical theory iscompletely silent on the existence and nature of quiddities.Given Russellian monism, terms for the basic entities in physicaltheory, such as “mass” and “charge”, perhapsrefer to quiddities, to entities that have quiddities. Onemight take that point alone to belie the structuralist’scomplete-silence claim. Further, one might wonder, why shouldRussellian monists rule out the possibility of physical theoryexpressing substantive information about the referents of its terms(Hawthorne 2002)?
In response, one might argue that physical theory itself is neutralbetween a Russellian monist interpretation according to which termslike “mass” and “charge” refer to quiddities,and a structuralist interpretation according to which such terms donot refer to quiddities (Ladyman and Ross, with Spurrett and Collier2007). That is, one might argue that the commitments ofphysicaltheory itself, by contrast with ametaphysical interpretationof physical theory, do not go beyond the structural.Structuralism about physics could thus be viewed as a thesis onlyabout physical theory itself. In that case, the considerationsmentioned in the preceding paragraph do not challenge the claim thatRussellian monists are committed to structuralism about physics.
Alternatively, one might give up the claim that Russellian monists arecommitted to structuralism about physics (Alter and Pereboom 2023; butsee Brown 2023). What is at the core of Russellian monism, one mightcontend, is only that there are quiddities that both underliestructural features physics describes and are relevant toconsciousness. So, of the three claims we have described as coreRussellian monist theses, perhaps only the second and the thirdqualify as such. Or instead, Russellian monists might retainstructuralism about physics but explicate it in a way that isrelatively uncontroversial (or at least less controversial than itoften is), e.g., as a view aboutcurrentphysics.
As noted above (section 1.1), many Russellian monists claim that macrophenomenal properties areconstituted by quiddities, or by quiddities and structure.But not all do. Instead, some claim that macrophenomenal propertiesemerge from quiddities, or from quiddities and structure(Goff 2015), where emergence is taken to be a (diachronic orsynchronic) causal relation.
However, some object that the causal-emergence construal underminesthe claim that on Russellian monism consciousness is integrated intophysical causation (Alter and Nagasawa 2012: 81; Chalmers 2013 [2015:253–259]). The objection could be stated as follows. On thecausal-emergence construal, there are causal laws connectingquiddities (or quiddities and structural properties) tomacrophenomenal consciousness: laws that are neithera priorientailed nor metaphysically necessitated by the structural truths thatphysical theory describes. In this respect, such laws would be justlike psychophysical laws posited by traditional interactionistdualism. And it is precisely because of the need to posit suchpsychophysical laws that traditional interactionist dualism is chargedwith failing to integrate consciousness adequately into physicalcausation (Smart 1959). Thus, on the causal-emergence construal, thatcharge would apply equally to Russellian monism (but see Goff 2015:294–97; 2017: Part IIpassim).
Much (though by no means all) of the recent discussion of Russellianmonism focuses on Russellian physicalism, on which quiddities arephysical properties. But there is no well-developed conception ofphysical quiddities, or at least none that is widely accepted. Thishas prompted speculation as to what such properties might be like.
Derk Pereboom (2011: ch. 5) identifies some candidates for physicalquiddities in the history of philosophy. One comes from theAristotelian tradition, on which matter is made up at least partly ofprime matter. Prime materiality could be understood as aphysical quiddity that underlies matter, which is structural. OnThomas Aquinas’s (1252–56) Aristotelian view, forinstance, matter consists in extension in three dimensions, i.e.,quantity. But quantity is a form, and in Aquinas’s viewa form must inhere in something. That something, he thinks, is primematter. Unlike quantity, prime materiality is not (in the terminologyofsection 1.1) a relatively intrinsic property of matter. Instead, it is anabsolutely intrinsic property of matter, and it could be seen as aphysical quiddity (Pereboom 2011: 85, 113). Admittedly, Aquinas heldthat prime matter is unintelligible. But other Aristotelians did not(Pasnau 2009).
Another candidate for a physical quiddity that Pereboom identifies isdue to John Locke (1690). In Locke’s conception,solidity is the categorical basis for impenetrability, whichis dispositional. Lockean solidity is “that whichhinders the approach of two bodies when they move toward oneanother” (Locke 1690: II, iv, emphasis added). Lockean solidityis also what differentiates matter from space, and is a definingproperty of matter. On one plausible interpretation, Pereboom (2011:97–100) suggests, Locke regards that property as an absolutelyintrinsic property of matter and as a physical quiddity.
A third candidate for physical quiddities, which is similar to Lockeansolidity, may be found in contemporary metaphysics. Many contemporarymetaphysicians accept that there exist non-Humean causal powers (seethe entrydispositions). Humeans identify causal powers with mere tendencies, which can befully explicated by means of subjunctive conditionals such as,“If an electron were in the vicinity of a proton, it wouldattract that proton” (Jacobs 2011). By contrast, non-Humeansidentify causal powers with properties that categorically groundtendencies—much as Lockean solidity grounds impenetrability.Such categorical properties are truthmakers for, and not explicable interms of, subjunctive conditionals. For example, the causal power ofan electron is a categorical property that makes subjunctiveconditionals such as the one just specified (“If anelectron…”) true (Jacobs 2011; cf. Heil 2003, Jaworski2016). Such truthmakers could be construed as physical quiddities(Gundersen 2015, Pereboom 2016, 2019).
So, there are at least some candidates for what physical quidditiesmight be like (for others, see, e.g., Chalmers 2013 [esp. 2015:270–274]; Rosenberg 2004; Coleman 2012, 2015). Still, one mightwonder if any of the candidates just described—Aristotelianprime materiality, Lockean solidity, or non-Humean causalpowers—can serve all of the Russellian physicalist’spurposes. Russellian physicalists argue that positing physicalquiddities safeguards their theory from the conceivability andknowledge arguments (section 3.1). But suppose those quiddities are construed as non-Humean causalpowers of physical entities such as electrons. Wouldn’t Mary bein a position to fully understand such powers while she is still inthe room? (Pereboom 2011: 113–114). If so, then the proposedconstrual would seem to undermine the Russellian physicalist’sbasis for denying that her pre-release physical knowledge isincomplete. Similarly, one might argue that the proposed construalwould undermine the Russellian physicalist’s contention thatwhile a structural zombie world is ideally conceivable, a full-fledgedzombie world is not. Parallel concerns arise for Aristotelian primemateriality and Lockean solidity. Even if those properties could beconstrued as physical quiddities, they would not appear to facilitatethe Russellian physicalist’s objective of providing a plausibleresponse to the conceivability and knowledge arguments (cf. Botin2023).
Chalmers writes:
If we can find a reasonable solution to the combination problem foreither [Russellian panpsychism or Russellian panprotopsychism], thisview would immediately become the most promising solution to themind–body problem. (Chalmers 2013 [2015: 274])
Whether or not Chalmers is right about this, many philosophers of mindbelieve that Russellian monism merits serious attention.
How to cite this entry. Preview the PDF version of this entry at theFriends of the SEP Society. Look up topics and thinkers related to this entry at the Internet Philosophy Ontology Project (InPhO). Enhanced bibliography for this entryatPhilPapers, with links to its database.
Descartes, René: physics |dispositions |dualism |Elisabeth, Princess of Bohemia |functionalism |grounding, metaphysical |idealism |intrinsic vs. extrinsic properties |Leibniz, Gottfried Wilhelm: philosophy of physics |mental causation |mind/brain identity theory |neutral monism |panpsychism |physicalism |physics: structuralism in |Schopenhauer, Arthur |structural realism
View this site from another server:
The Stanford Encyclopedia of Philosophy iscopyright © 2023 byThe Metaphysics Research Lab, Department of Philosophy, Stanford University
Library of Congress Catalog Data: ISSN 1095-5054