Three Kinds of Arguments for Panpsychism

Abstract

Panpsychism may be roughly defined as a view that at least some of the properties constituting the fundamental level of reality are mental or proto-mental. Despite its long history, it has been revived in recent discussions as a solution to the problems raised by the mind, especially to the so-called hard problem of consciousness. Contemporary panpsychism differs significantly from incarnations known from the history of philosophy mainly due to the fact that the former is often combined with so-called Russellian monism. According to Russellian monism, the intrinsic properties of physical things remain unknown. This encourages panpsychists to argue that those properties are in fact mental. In my paper, I examine the three most common arguments for panpsychism: the Continuity Argument, the Hegelian Argument, and the Agnostic Argument. I take a closer look at each of them to assess their advantages and weaknesses. As I point out, the way in which one argues implies the version of panpsychism one adopts. This turns out to be especially important with regard to the Hegelian Argument and the Agnostic Argument. Both can be combined with Russellian monism, resulting in Russellian panpsychism. However, I claim that the philosophical consequences of these arguments are different, so it is legitimate to distinguish two kinds of Russellian panpsychism. In conclusion, I hold that there are reasons to prefer panpsychism based on the Agnostic Argument, which is more promising, as it responds to some general problems of panpsychism.

In this paper, I argue that panpsychism can be justified in at least three different ways. The first way consists of an assumption that the chain of sentient beings cannot end abruptly, so consciousness runs all the way down to the fundamental level of reality. I call this argument the Continuity Argument. The second way is based on the assumption that both materialism and dualism—which are usually thought to be the only viable options in the mind–body debate—can be proved deeply problematic. In light of this fact, one should choose panpsychism—“the Hegelian synthesis,” as Chalmers (Chalmers, 2015, p. 247) calls it—which has the advantages of materialism and dualism but at the same time avoids their downsides. Following Chalmers, I call this argument the Hegelian Argument. The third way begins with the claim that, due to our lack of cognitive access to non-conscious phenomena, conscious phenomena are all that we know. It is legitimate, then, to assume that the nature of reality—of which we are ignorant, as the argument says—is (at least partially) the same as the nature of our minds, namely, mental. I call this argument the Agnostic Argument.

The purpose of this paper is to analyze these influential arguments with respect to their strengths and weaknesses. I claim that the Continuity Argument is invalid, so the only viable arguments are the Hegelian Argument and the Agnostic Argument. What I find interesting is that while both can be placed in a wider frame of Russellian monism, they constitute two different versions of the latter. This distinction is both novel, as it does not appear in literature about panpsychism, and important, because—as I argue—the Agnostic Argument offers the most secure position for a panpsychist. Still, it has to face its own, maybe even intractable, problems.

1 Terminological Preliminaries

Within the last 100 years, the analytic philosophy of mind has been dominated by materialism, a view that one can fully account for mental phenomena in purely physical terms, such as behavior or brain processes.¹ This approach supplanted dualism, according to which an adequate explanation for mental phenomena can be achieved only if we accept the existence of non-material substances, for example, Cartesian souls. Materialists often argue that dualism is implausible, mainly due to the supposed incompatibility with science; some of them—for example, Gilbert Ryle—defend even the stronger claim that dualism is logically incoherent. For many decades, the universal belief was that the choice between materialism and dualism leaves no place for an alternative.² In effect, it was thought that the lack of good arguments for the latter makes a case for the former.³

However, at the end of the twentieth century, the widespread consensus was challenged by proponents of the so-called hard problem of consciousness. The problem stemmed from the recognition that the materialistic explanation is incomplete: even the most detailed description of the world in purely physical terms leaves the domain of phenomenal consciousness—conscious experiences—untouched.⁴ This led philosophers to look for new theories that could navigate between Scilla of materialism and Charybdis of substance dualism. One of the previously overlooked alternatives was panpsychism, which views mental phenomena as fundamental, i.e., as a part of the natural domain. At first glance, panpsychism seems to be the best of both worlds: it does not reduce mental phenomena to non-mental phenomena and, at the same time, it rejects the existence of supernatural beings, such as non-physical souls.⁵ Within the last 20 years, panpsychism has gained considerable attention, and today it is one of the most discussed views in philosophy of mind.

It would be wrong, however, to think of panpsychism as a novel view. Quite the contrary, it has a venerable history and—despite its changing fortunes—it has actually never left (Skrbina, 2005). Its long tradition and numerous metamorphoses are the main reason why it is so difficult to define. Nevertheless, such a definition is necessary, as the borderline between some versions of panpsychism and competing views—i.e., materialism and dualism—is blurry. What I would like to propose is a definition wide enough to cover most (perhaps all) varieties of panpsychism but that can be narrowed down by answering to further clarificatory questions:

Panpsychism is a metaphysical view according to which at least some of the properties constituting the fundamental level of reality are mental or proto-mental.

Although there are at least five possible questions that can make this definition more specific, I would like to focus on the two that will play a significant role in the next parts of this paper⁶:

• (Q1) What does the mental mean?

• (Q2) What is a metaphysical relationship between fundamental mental stuff and fundamental physical stuff?

What makes (Q1) so important is the fact that the way in which one defines “the mental” determines what panpsychism exactly says. Recent panpsychism answers this question by identifying “the mental” with “the qualitative” or “the experiential.” It is because the view appeared as a solution of the hard problem of consciousness, that is, the problem with explaining the so-called phenomenal consciousness—the property of being a qualitative feeling or a what-is-it-like type of experience—in purely physical, quantitative terms.⁷ For this reason, the recent version of panpsychism is usually called panexperientialism.⁸ Such understanding is new: in modern philosophy—even in some writings of William James and Bertrand Russell—“the mental” was usually understand more broadly.⁹

(Q2) is no less important as different answers to it can lead to opposite metaphysical interpretations of panpsychism. According to the first interpretation, there is only kind of fundamental mental stuff all the way down. If this is the case, then panpsychism is a version of metaphysical idealism and (Q2) is simply ill posed. However, it may be argued that panpsychism does not imply that mental stuff is the only fundamental stuff in the universe.¹⁰ If this is so, then—as the second interpretation says—the fundamental stuff may consist of fundamentally mental stuff as well as fundamentally non-mental stuff. Because this view can be also interpreted as a kind of naturalistic dualism, it suggests that the borderline between it and panpsychism is blurry.

What is the source of disagreement between proponents of these two interpretations? Both parties agree that prima facie there is a distinction between fundamental mental beings and fundamental non-mental beings. Most philosophers think that this distinction accurately describes the fundamental reality, so they believe that both mental and non-mental are equally real. On the other hand, those who sympathize with idealism argue that this distinction is only superficial and stems from the way we acquire knowledge about the world. In fact, any token of fundamental stuff can be known as the mental and as the physical but is nevertheless only the former. Such a view is sometimes called double-access or double-knowledge theory (Feigl, 1975, p. 14).¹¹ The difference between panpsychistic idealism and panpsychistic dualism can be reduced to the question: Do we have any reasons to believe that there is something non-mental in the universe? As we shall see, the way one answers to it changes the interpretation of panpsychism drastically.

2 Kinds of Arguments for Panpsychism

Having a clear definition of panpsychism and a good grasp of its most basic versions, we are ready to follow the three most common and most significant arguments for this view put forward in the history of philosophy as well as in recent debates.

2.1 The Continuity Argument

Historically speaking, one of the earliest arguments for panpsychism is the argument appealing to the obvious observation that the borderline between conscious and unconscious beings is blurry. It is thus plausible that extremely primitive consciousness can still be present at the very bottom of the physical world. We may call this type of reasoning the Continuity Argument. It runs as follows:

• 1.1 Human beings have mental phenomena.

• 1.2 When we go down the ladder, we encounter other sentient beings that also have mental phenomena.

• 1.3 The borderline between sentient beings that have mental phenomena and insentient beings that do not have them is always arbitrary (or impossible to draw for other reasons).

• 1.4 If the borderline between sentient beings that have mental phenomena and insentient beings that do not have them is always arbitrary (or impossible to draw for other reasons), then it is the best to assume that mental phenomena already exist at the fundamental level of reality.

• 1.5 Therefore, it is the best to assume that mental phenomena already exist at the fundamental level of reality.

The argument had many defenders, especially amongst philosophers inspired by Darwin’s theory of evolution. As William James famously notes, “If evolution is to work smoothly, consciousness in some shape must have been present at the very origin of things. Accordingly, we find that the more clear-sighted evolutionary philosophers are beginning to posit it there” (James, 1890, p. 149; James’s emphasis). Those “clear-sighted philosophers” were John Tyndall, Herbert Spencer, and William Clifford, all of whom adopted the Continuity Argument. For example, Clifford argued that:

as we go back along the line, the complexity of the organism and of its nerve-action insensibly diminishes; and for the first part of our course, we see reason to think that the complexity of consciousness insensibly diminishes also. […] But as the line of ascent is unbroken, and must end at last in inorganic matter, we have no choice but to admit that every motion of matter is simultaneous with some ejective [i.e. conscious—JJ] fact or event which might be part of a consciousness. (Clifford, 1878, pp. 64–65)

One of the recent versions of a similar argument is proposed by Chalmers:

As we move along the scale from fish and slugs through simple neural networks all the way to thermostats, where should consciousness wink out? The phenomenology of fish and slugs will likely not be primitive but relatively complex, reflecting the various distinctions they can make. […] The thermostat seems to realize the sort of information processing in a fish or a slug stripped down to its simplest form, so perhaps it might also have the corresponding sort of phenomenology in its most stripped-down form. (Chalmers, 1996, p. 295)

It is difficult to say whether Chalmers finds it to be an independent argument for panpsychism or just a way “to make the view seem less crazy” (Chalmers, 1996, p. 294). He believes, however, that “a radical discontinuity from complex experiences to none […] seems unlikely” (Chalmers, 1996, p. 294), and in that way he clearly expresses an intuition on which the Continuity Argument rests. Numerous other authors put forward this reasoning, although it is rarely used as the only or main reason for embracing panpsychism.¹² In fact, an intuition behind this argument is very old (Skrbina, 2005, pp. 26–57). In modern philosophy, it is most famously adapted by Gottfried Leibniz, who turns it into a general principle called the “Law of Continuity,” expressed in the Latin phrase natura non facit saltus (nature does not make leaps).¹³ The principle can be also found in premise (1.4).

Unfortunately, while (1.4) is crucial, it is at the same time the most vulnerable to criticism.¹⁴ Perhaps the best way to prove its falsity is to show that the general scheme of the argument is the subject of a reductio. Consider:

• 1.1* Human beings are living beings.

• 1.2* When we go down the ladder, we encounter other living beings.

• 1.3* A borderline between living beings and non-living beings is always arbitrary (or impossible to draw for other reasons).

• 1.4* If the borderline between living beings and non-living beings is always arbitrary (or impossible to draw for other reasons), then it is the best to assume that life already exist at the fundamental level of reality.

• 1.5* Therefore, it is the best to assume that life already exists at the fundamental level of reality.

In that way, we argue for vitalism, which has been proved false by science. This clearly suggests that the argument is flawed. The problem is—critics say—that nature sometimes does make leaps, and we have scientific evidence that this is not uncommon. By virtue of that, one can argue that the imagination of panpsychists is simply too poor to see how mental phenomena can be caused by highly complex non-mental phenomena.¹⁵ The proponents of the Continuity Argument are left with only one viable answer: they may argue that the problem of consciousness has a special status in science, so the reductio is an example of a wrong analogy.¹⁶ However, this moves the discussion only one step further—to the question of whether there are reasons to see the problem of consciousness as a special one. We may expect this debate to be no less heated (but perhaps no more conclusive) than the original one. In any case, panpsychism motivated by the Continuity Argument clearly needs an independent argument. Otherwise, it fails to prove what it is supposed to prove.

2.2 The Hegelian Argument

Although materialism is still a dominant view in the philosophy of mind, for the last 50 years, it has been constantly challenged by counterarguments, such as the epistemic gap argument, the explanatory gap argument, and the inverted spectrum argument. However, perhaps the most powerful of them are conceivability arguments. While they date back to Descartes, two recent versions of them are by Saul Kripke and Chalmers. The latter offers the so-called zombie argument, still thought to be one of the most sophisticated and effective reasonings against materialism. According to the argument, we can conceive a possible world where all the physical states are duplicated but where there are no phenomenal properties. If this is the case, then the relation between physical properties and phenomenal states is not strong enough to make materialism viable. At the same time, we are finding more and more evidence that dualism—the second horn of the traditional mind–body dilemma—is untenable: it not only seems incompatible with our scientific knowledge but also leaves many problems, such as mental causation, i.e., the problem of causality between the mental and the non-mental, unsolved. In effect, as Chalmers admits, “we have a standoff” (Chalmers, 2015, p. 252; see Goff, 2019a, p. 145). This encourages him to search for a Hegelian synthesis that saves the advantages of both views and avoids their downsides. Chalmers believes that this requirement is met by panpsychism. The “Hegelian” argument inspired by his approach runs as follows¹⁷:

• 2.1 Physicalism poses many problems in explaining phenomenal consciousness.

• 2.2 Dualism poses many problems in explaining phenomenal consciousness.

• 2.3 Panpsychism poses less problems in explaining phenomenal consciousness than physicalism and dualism.

• 2.4 An explanation that poses less problems is better than an explanation that poses more problems.

• 2.5 Therefore, panpsychism is a better explanation of phenomenal consciousness than physicalism and dualism.

Moreover, if we assume that physicalism and dualism are the only views that can compete with panpsychism—which is far from obvious but let us ignore it for the sake of the argument—we can modify (2.3):

• 2.3* Panpsychism poses the least problems in explaining phenomenal consciousness.

and reach a conclusion:

• 2.5* Therefore, panpsychism is the best explanation of phenomenal consciousness.¹⁸

At first glance, this argumentation is very promising. Nevertheless, it has to deal with several difficulties as most of the premises—perhaps with the exception of (2.4)—are disputable. Against (2.2), one can argue that, despite the fact that dualism has certainly lost much of its popularity in the last decades, it is still logically coherent. Indeed, some of the leading panpsychists, such as Chalmers and Goff, are ready to admit it.¹⁹ One can also attack (2.3) by claiming that in fact panpsychism brings more problems that its competitors. As we shall see, there are reasons (most notably the combination problem) supporting this criticism.

However, the most troublesome is premise (2.1), which states that physicalism is not a viable solution to the hard problem of consciousness. Philosophers have devoted literally thousands of pages to proving it, but the discussion is far from being settled once and for all. Materialism still has powerful proponents and their view is by all means defensible. Most notably, one can maintain that, if one day it becomes possible to dismiss all the arguments against materialism, panpsychism will be no better than its competitor. If this is the case, then the standoff is temporary and needs not new solutions but—at most—new counterarguments.

Panpsychists may try to avoid this conclusion by arguing that materialism not only is not but also will never be a viable solution to the mind–body problem. Such an argument usually rests on an independent view called Russellian monism. The latter consists of two assumptions:

• RM₁. Physics describes extrinsic (i.e., structural, dispositional, behavioral) features of things.

• RM₂. Extrinsic properties have to be grounded in intrinsic properties that cannot be grasped by physics.²⁰

(RM₁) is based on an observation that physics describes its entities abstractly in terms of structure, dispositions, and behaviors. For example, a definition of an electron states what it does (e.g., it attracts particles with positive charges and repels particles with negative charges as well as causes certain readings on the displays of our apparatuses), not what it is. Doubtless, though, as (RM₂) maintains, there must be some intrinsic properties—categorical bases—of these dispositions and behaviors. However, we are ignorant about the intrinsic features as science can say nothing about them.

What follows from (RM₁) and (RM₂) taken together is that, due to our cognitive limitations, we do not know—and probably never will know—the nature of intrinsic properties of the ultimate stuff in which the physical world is grounded. If this is true, then the standoff between materialism and panpsychism will perhaps be impossible to solve, leaving us in a state of permanent agnosticism about the nature of the constituents of reality. In that case, the choice between materialism and panpsychism is arbitrary.

However, there is an argument that seems to tip the scale in panpsychism’s favor: the argument of simplicity. We know that our scientific picture of the world is seriously incomplete. It has two gaps: the first is phenomenal consciousness, which cannot be accounted for in physical terms, and the second is the intrinsic properties of physical things of which we are ignorant. The proponents of panpsychism suggest that the simplest way to eliminate these two mysteries is to solve them in one go by assuming that conscious phenomena are intrinsic properties of physical things. If they are right, it means that physics describes structural or dispositional features of reality that intrinsically are experiences. Such a view—which seems to dominate the recent debates—can be called Russellian panpsychism.

Unfortunately, both (RM₁) and (RM₂) are prone to attack. (RM₁) is challenged by the argument of a British mathematician, M.H.A. Newman, who claims that it is implausible to hold that science tells us nothing about non-structural properties. One can come to this conclusion only by putting unnaturally strong constraints on our knowledge and claiming “that nothing can be known that is not logically deducible from the mere fact of existence” (Newman, 1928p. 144). Moreover, if one accepts the latter claim, one ends up with the conclusion that physics cannot be an empirical science, for everything it is able to say can be deduced a priori.²¹ (RM₂), in turn, is questioned by philosophers claiming that the idea of extrinsic properties without intrinsic properties is perfectly coherent (see, e.g., Ladyman, 1998; Tegmark, 2008).

While one can still defend Russellian monism, it seems clear that the Hegelian Argument for panpsychism depends heavily on other arguments—namely, those against materialism and for Russellian monism—and thus is not self-sufficient. It gives proponents of panpsychism a reason to look for a better argument.

2.3 The Agnostic Argument

The third argument for panpsychism starts with the distinction between knowledge by acquaintance and knowledge by description. Knowledge by acquaintance is knowledge that we are directly aware of and therefore is certainly true. By contrast, knowledge by description is inferred from knowledge by acquaintance, so—like every reductive reasoning—is prone to errors. Although the distinction was popular in modern philosophy (most notably in British empiricism), it became unfashionable in the nineteenth century. It is Bertrand Russell who brings it back to life by making it a crucial point of his epistemology (see Russell, 1912, pp. 72–92).²² His main claim is that any knowledge can be ultimately reducible to knowledge by acquaintance, that is, to certain experiences, such as patches of color, sounds, tactile sensations, and so on.²³ This view is also the starting point of the Agnostic Argument:

• 3.1 Our knowledge is ultimately rooted in knowledge by acquaintance (knowledge of experiences).

In later works—The Analysis of Matter and An Outline of Philosophy—Russell develops this point by claiming that knowledge by acquaintance lets us grasp the intrinsic natures of our experiences. By contrast, knowledge by description is abstract and leaves us ignorant about the intrinsic properties of physical things. Russell spells the view out in an often-quoted sentence: “physics, in itself, is exceedingly abstract, and reveals only certain mathematical characteristics of the material with which it deals” (Russell, 1954, p. 10). One can rightly notice that it is a different expression of a view that I earlier labeled as Russellian monism. However, there is a fundamental difference between these approaches. Russellian monism—especially its first claim, (RM₁)—is usually thought to be an assumption that does not need any further justification. Even if such justification is offered, it almost never refers to (3.1), because such the view instantiated by this premise is often thought to be “too empiricist” (Smart, 2006, p. 159). On the contrary, Russell finds (3.1) vital, and it is thus far from obvious that he would count himself as a Russellian monist.

The acceptance of (3.1) has further consequences. One of the most frequently overlooked ones is a major revision of philosophical concepts. Russell is fully aware that, if knowledge is limited to what is given by acquaintance, then the philosophical dictionary—particularly the Cartesian opposition of material bodies and mental minds—must be revised. Matter understood as a non-experiential intrinsic property of physical things is by definition absent in our experience and is at most a useful fiction, as is mind, which is in fact no more than a set of simultaneous experiences with no underpinning substance (Russell, 1927, p. 148). A natural distinction between mind and matter is thus not metaphysical but purely epistemological. The appearance that this distinction is metaphysical comes from the fact that we know things in two different ways: we call the things that we know by acquaintance as mental and the things that we know by description as physical.²⁴ If this is so, then we are unable to answer the question of what is the nature of the fundamental stuff in which reality is grounded in. What we do know is that it cannot be called either material or mental. It is thus best to call the stuff neutral.²⁵

Now we are ready to run Russell’s Agnostic Argument for neutral monism:

• 3.1 Our knowledge is ultimately rooted in knowledge by acquaintance (knowledge of experiences).

• 3.2 Knowledge by acquaintance (knowledge of experiences) contains knowledge of intrinsic properties.

• 3.3 Knowledge by description (knowledge of things that are not experiences) does not contain knowledge of intrinsic properties.

• 3.4 Therefore, intrinsic properties of things that are not experiences remain unknown.

Of course, this is not yet an argument for panpsychism. However, by adding one premise, we can turn the Agnostic Argument for neutral monism into an Agnostic Argument for panpsychism.²⁶ Galen Strawson (Strawson, 2017, p. 384), who is arguably the closest to the original thought of Russell, makes this step.²⁷ He adds a crucial premise:

• 3.5 There is no radical emergence.

What is radical emergence? According to Chalmers,

a high-level phenomenon is strongly emergent [radically emergent in Strawson’s terms—JJ] with respect to a low-level domain when the high-level phenomenon arises (in some sense) from the low-level domain, but truths concerning that phenomenon are not deducible even in principle from truths in the low-level domain. (Chalmers, 2006, p. 244)

With regard to experiential phenomena, radical emergence means that experiences can arise out of something non-experiential. Strawson believes that such emergence would be unintelligible. He proposes two arguments to support this view. The first states that, if such emergence occurred, the appearance of experiential phenomena would be completely unexplainable. If our scope is the science—not the magic—of consciousness, we should avoid this conclusion (see, e.g., Strawson, 2008, p. 65). In fact, Strawson restates a problem already noticed a century ago by Samuel Alexander (Alexander, 1920, p. 47), who is aware that any instantiation of radical emergence cannot be accounted for, so it would have to be accepted “with a natural piety.”

The second argument is that, if one wants to defend the radical emergence of experience out of the non-experiential, one has to have at least a vague idea of what the latter could be. However, it follows from (3.4) that any idea of the intrinsically non-experiential can never be made intelligible to us. To be plausible, materialism must fulfill an impossible requirement, that is, to define the non-experiential. Strawson thinks that both arguments clearly prove the unintelligibility of radical emergence.²⁸

One can notice that the arguments against radical emergence are in fact based on the Law of Continuity, which we already found to be implausible. It is true that, in both cases, the basic intuition is that nature does not make leaps, at least with regard to experiences. Note, however, that, in the case of the Continuity Argument the Law of Continuity is an assumption with no further argumentation and is therefore vulnerable to criticism. By contrast, the Agnostic Argument is justified by the fact that we do not know whether the non-experiential exists; we then have no reason to believe that there are any radical jumps in nature between the experiential and the non-experiential.

Now we are ready to put forward the Agnostic Argument for moderate panpsychism (psychism):

• 3.1 Our knowledge is ultimately rooted in knowledge by acquaintance (knowledge of experiences).

• 3.2 Knowledge by acquaintance (knowledge of experiences) contains knowledge of intrinsic properties.²⁹

• 3.3 Knowledge by description (knowledge of things that are not experiences) does not contain knowledge of intrinsic properties.

• 3.4 Therefore, intrinsic properties of things that are not experiences remain unknown.

• 3.5 There is no radical emergence (therefore, experiences cannot emerge out of something non-experiential).

• 3.6 Therefore, at least some intrinsic properties are irreducibly (fundamentally) experiential.

Sometimes Strawson adds another premise—the monistic assumption. While it is not necessary, it may be accepted in virtue of simplicity:

• 3.7 There is only one kind of the fundamental stuff that makes up reality.

In that way, Strawson reaches the so-called pure panpsychism:

• 3.8 All intrinsic properties are irreducibly (fundamentally) experiential.

The crucial premise of the Agnostic Argument is of course (3.1). One may challenge not only the premise but also its assumption that the distinction between knowledge by acquaintance and knowledge by description is legitimate.³⁰ However, there is a further difficulty—an unwanted consequence of (3.1) that a proponent of the Agnostic Argument has to face: the premise deprives us of all (or almost all) non-structural knowledge of the world. The problem is raised by Immanuel Kant: if metaphysics is supposed to transcend our experiences and say something about the world, in light of (3.1), it simply turns out to be impossible. Some philosophers will happily welcome the conclusion and agree that metaphysics is a pile of nonsense. Panpsychists, however, should be quite suspicious about it. After all, what they hope for is to answer the question about the nature of things that are not our actual experience, and (3.1) means that there will never be a conclusive argument for their view. Russell is perfectly aware of this: he notes that the stuff reality is made out of “may be just like the events that happen to us, or […] [it] may be totally different in strictly unimaginable ways” (Russell, 1975, p. 13). The transition from knowledge by acquaintance to knowledge of the intrinsic properties of physical things is prima facie the greatest problem that may disincline philosophers from approving of the Agnostic Argument for panpsychism.

3 Russellian Panpsychisms?

It is generally known that there are many different arguments for panpsychism, yet it is seldom—if ever—noticed that the way of argumentation determines the kind of panpsychism that one accepts. This is especially the case with regard to the view called Russellian panpsychism, which in fact embraces two slightly different views. Both of them are justified by the same reasoning, which runs as follows:

• RP₁. Consciousness is real and cannot be explained in purely physical (non-experiential) terms.

• RP₂. We do not know the intrinsic nature of physical entities.

• RP₃. Therefore, it is legitimate to assume that the consciousness and the intrinsic nature of physical entities are the same.

However, as I have pointed out, premise (RP₁) may be argued for twofold. In the case of the Hegelian Argument, it is justified by (2.1) the falsity of materialism (and dualism) backed up by independent arguments. Because the most significant amongst the latter are a priori conceivability arguments, we can call this view rationalistic Russellian panpsychism (RRP). In the case of the Agnostic Argument, (RP₁) is justified by (3.1) regarding a special status of knowledge by acquaintance that is an irreducible foundation for any other knowledge. Because the argument is rooted in empiricism, we can call it empiristic Russellian panpsychism (ERP).³¹ Although both RRP and ERP seem to head in the same direction, the differences between them are substantial and have an impact on the current debates. These differences are the following:

• 1. The vocabulary of RRP and that of ERP differ significantly. RRP usually sticks to traditional, Cartesian terminology of matter and minds. At the same time, it is a priori assumed that those terms exclude themselves (minds are immaterial and matter lacks mind) and that both describe—prima facie accurately—the most fundamental distinction of kinds of entities in the world.³² RRP is thus usually based on dualistic conceptual scheme that is shared both by proponents of materialism and their opponents. (The only difference is that materialists come to the conclusion that some terms of this scheme are inaccurate, while their adversaries think otherwise.) ERP, by contrast, rejects the Cartesian conceptual scheme due to the fact that its terms are in fact empty—they do not refer to anything that is directly given—and appear in philosophical language due to “voluntary and conscious economy of mental representation and designation, as expressed in ordinary thought and speech” (Mach, 1914, p. 3). ERP thus requires a replacement of the old scheme with a new one, consisting of terms referring to what is directly given—to “events,” “perspectives,” “experiences,” and so on. Such reconceptualization of the mind–body problem is supposed to dispel confusion and to help find its solution (Strawson, 2003, p. 51).³³

• 2. RRP allows for more reductive views, while ERP excludes them. Proponents of RRP can—and sometimes actually do—adopt more reductive views, such as panprotopsychism (which may be seen as “a borderline case of physicalism” (Montero, 2010, p. 77))³⁴ and neutral monism. In fact, the Hegelian Argument does not involve any premise that would rule out such interpretations. Without further assumptions, proponents of RRP argue not that consciousness cannot be reduced to anything but that it cannot be reduced to the physical stuff. Hence, because panprotopsychism and neutral monism are usually thought to be alternatives to physicalism, both are still viable options for a defender of RRP. It is an unwanted consequence as, in that way, the view loses one of its most important advantages over standard physicalism, namely, the irreducibility of consciousness. It happens, because RRP allows for postulated (i.e., inferred and not directly given) properties, such as the proto-mental and the neutral. By contrast, proponents of ERP explicitly find the conception of the non-experiential or even the “almost experiential” to be unintelligible. Although they cannot rule out the logical possibility that some of the intrinsic properties of things in the world are non-experiential, they argue that we do not have cognitive access to these properties, so we do not have any reason to claim that they exist. Moreover, even if they existed, we would have to accept radical emergence in order to explain how those non-mental properties give rise to mental properties. However, it seems that such a relation does not appear anywhere in the natural world. It gives us two reasons to reject any view claiming that fundamental properties are not mental.

• 3. Contrary to ERP, RRP can be interpreted as standard materialism. Because RRP sticks to the distinction between the mental and the physical, and finds intrinsic properties to be inscrutable, the view allows for the possibility that those inscrutable intrinsic properties are ultimately physical (non-experiential). In fact, this is a popular argument against RRP put forward by Howell (2019), Montero (2010), and Stoljar (2001). By contrast, ERP avoids this problem because it requires additional proof for the existence of non-experiential properties (such as special physical properties, neutral properties, proto-experiential properties) that has not been—and maybe will never be—offered. It is thus very difficult (or perhaps even impossible) to justify physicalism on the grounds of ERP.³⁵

• 4. RRP relies on other arguments, while ERP is (mostly) self-sufficient. One of the biggest problems with RRP is that it relies on other arguments, such as Chalmers’ zombie argument that may turn out to be false. Moreover, if RRP is to be true, not only materialism but all similar theories must be ruled out. It is certainly bad news for proponents of RRP, as they argue not for their view but against competitive views. Even if they succeed, it is still possible that someday such a competitor, immune to any counterarguments, will appear. RRP is thus at most a working hypothesis, particularly prone to falsification. ERP does not face this problem: any competitive view based on the conception of the non-experiential is thought to be less plausible than panpsychism. Of course, ERP may turn out to be false (for example, because of internal incoherence or because some day we will find the cases of radical emergence) and perhaps will never be verified. Still, because it is more self-sufficient, it has to face one fewer problem—and therefore is more plausible—than RRP.

• 5. ERP encounters one less difficulty than RRP in solving the combination problem. RRP is justified by conceivability arguments, which are very useful when we argue for panpsychism but turn out to be fatal when we try to solve the so-called combination problem.³⁶ The latter appears in multiple versions but may be summarized in a question: “How do the experiences of fundamental physical entities such as quarks and photons combine to yield the familiar sort of human conscious experience that we know and love?” (Chalmers, 2016, p. 179). As Chalmers argues, combination problems are subject to the conceivability argument. Just as we can conceive a duplication of microphysical properties of a brain that is not followed by conscious phenomena, we can conceive a duplication of microphenomenal properties that is not followed by an appearance of macrophenomenal properties. It effect, the conceivability argument makes the combination problem extremely difficult to solve. Unfortunately, the most efficient solution—the rejection of conceivability arguments—deprives the proponents of RRP their principal reason to dismiss physicalism and accept panpsychism. By contrast, ERP also faces combination problems but is able to avoid the constraints imposed by the conceivability arguments that seem to make them intractable. A proponent of ERP can hold that conceivability arguments fail because it is not possible to duplicate physical properties without phenomenal properties as physical properties are phenomenal properties.³⁷ By making this step, ERP does not need to worry about its justification, as the reasons to accept ERP are independent of Chalmers’s argument. Of course, it is only a small step forward, as the general solution of the combination problem is still unavailable both for RRP and ERP. However, proponents of the latter can look for such a solution without being worried with the conceivability arguments.

• 6. The consequences of ERP are much more radical than those of RRP. So far, I have suggested that ERP seems more promising than RRP. Why then has the latter and not the former dominated recent discussions about panpsychism? I think that the biggest obstacle to the acceptance of ERP is its far-reaching philosophical consequences, particularly the Kantian problem that I mentioned above. Due to the fact that the question about the intrinsic natures of physical things remains unanswered, the view may seem explanatorily futile and metaphysically unjustified. Indeed, without further details, ERP is no more than a wild speculation based on our ignorance. However, the conclusion of the Agnostic Argument can be seen not as a finishing line but as the starting point for a more positive argumentation. In the history of philosophy, many philosophers—such as German neo-Kantians and American critical realists (most notably C.A. Strong³⁸)—when defending views similar to ERP, try to bridge the gap between the realm of our experiences and the external world of things-in-themselves. A thorough assessment of their efforts exceeds the scope of this paper, but they certainly deserve a careful analysis. It is also true that ERP implies idealism as the most plausible metaphysical view. Many philosophers find this conclusion to be unacceptable and speaking in favor of RRP. However, upon reflection, it is not clear at all whether they are right. ERP-inspired idealism is different from George Berkeley’s view that the act of perception constitutes the object of perception and definitely does not imply any kind of solipsism.³⁹ Quite the contrary: the proponent of this view agrees “the physical world really exists out there, independently of our observations; it just has a surprising nature” (Chalmers, 2020, p. 354). Overall, the view seems to be no worse than panpsychism itself.⁴⁰

4 Conclusion

It is clear that panpsychism is not a single view but a family of views. By contrast, it is seldom noted that the differences between varieties of panpsychism concern not only particular claims but also their justifications. In this paper, I aimed to show that the way in which one argues for panpsychism determines the version of panpsychism that one will defend. Particular arguments can be assessed differently. I argued that the Continuity Argument alone is untenable, so we should focus on the Hegelian Argument and the Agnostic Argument. Both can be embedded in a wider framework of Russellian monism, but they lead to different conclusions. This outcome allows us to recognize that the similarities between various kinds of panpsychism may be only superfluous. I think that one of these apparent similarities appears between the Russellian panpsychists using the Hegelian Argument and Bertrand Russell, a proponent of the Agnostic Argument. It seems historically and dialectically important to bring this distinction to light. I was also suggesting that the view I called ERP is easier to defend than RRP and can deal with some problems haunting panpsychism better than its competitor. On the other hand, it brings about consequences many panpsychists may find unacceptable. Be that as it may, I think that the choice between ERP and RRP is ultimately a matter of personal taste—especially how far one is ready to depart from what is called the scientific picture of the world.

To be clear, I do not claim that panpsychism is defensible at all. Despite all its merits, it is challenged by combination problems that need to be answered to make it plausible.⁴¹ I also do not rule out the possibility of other serious problems with it. However, Russellian panpsychism is still in its early stage of development and conceptual work needs to be undertaken to establish clearly what this view really is.