A Critical Assessment of Shafer-Landau’s Ethical Non-Naturalism

Abstract

I focus on the ethical non-naturalism of Russ Shafer-Landau. First, I spend a good bit of time specifying the nature of two versions of naturalism and arguing that one is embraced ubiquitously—more importantly, should be embraced—by contemporary naturalists. I do so because if I am right about this, before we investigate the details of Shafer-Landau’s ethical non-naturalism, there will be a significant burden of proof for him to meet. In my view, that burden is strong enough to justify the claim that a critic’s epistemic task is merely to provide undercutting defeaters for Shafer-Landau’s position, and not to proffer rebutting defeaters, though I will attempt to supply both. After presenting a crucial characterization of contemporary naturalism followed by a critique of naturalist emergent properties, I state and critique Shafer-Landau’s ontology followed by the same for his epistemology. Both will be evaluated with a particular focus on their plausibility to support his ethical non-naturalism.

1. Introduction

David Papineau opines that contemporary naturalism is a consequence of the build-up of scientific evidence during the twentieth century. (Papineau 2001) I agree, but regardless of whether he is correct, one thing is certain: We are now confronted with a large and often disparate menagerie of standpoints that claim to represent naturalism. This is particularly true when it comes to naturalist ethical theory. I think it is fair to say that when it comes to ethical theory, most naturalists adopt either some version of non-cognitivism or ethical naturalism (See Moreland and Horner forthcoming).

However, there is a growing number who embrace ethical non-naturalism and, along with it, the existence and knowability of objective intrinsic value and moral duties. In my view, both in their sophistication and accurate representation of this approach, Erik Wielenberg and Russ Shafer-Landau are regarded as the leading advocates of ethical non-naturalism. I have criticized Wielenberg’s position elsewhere (Moreland 2020).

So, in this article I will focus on the views of Shafer-Landau. Accordingly, in what follows, I will spend a good bit of time specifying the nature of two versions of naturalism and arguing that one is embraced ubiquitously—more importantly, should be embraced—by contemporary naturalists. I do so because if I am right about this, before we investigate the details of Shafer-Landau’s ethical non-naturalism, there will be a significant burden of proof for him to meet. In my view, that burden is strong enough to justify the claim that a critic’s epistemic task is merely to provide undercutting defeaters for Shafer-Landau’s position, and not to proffer rebutting defeaters, though I will attempt to supply both. After presenting a crucial characterization of contemporary naturalism followed by a critique of naturalist emergent properties, I will state and critique Shafer-Landau’s ontology followed by the same for his epistemology. Both will be evaluated with a particular focus on emergent properties, and I believe those working in metaethics can benefit from the rich distinctions and assess their plausibility to support his ethical non-naturalism.

Before I provide a characterization of worldview naturalism, structural supervenient properties, and emergent properties, a few preliminary comments are in order. First, philosophers of science and mind have led the way in studying supervenient detailed analyses of these notions. Accordingly, I will turn to philosophy of science and mind in the upcoming section. This and the points to follow are important for the contribution I hope to make about Shafer-Landau’s ethical non-naturalism. In an earlier draft of this article, a very astute anonymous referee raised a number of criticisms of that draft. I have incorporated several of them in this article, but I believe that a cluster of objections were largely due to a misunderstanding of my approach. That misunderstanding was my fault. I was not making my precise approach clear enough. In this and the following preliminary comments, along with their application to specific challenges that arise from the literature in the metaethics of Shafer-Landau’s moral realism, I hope to provide a needed context for how my approach attempts to solve these issues.

Second, while supervenience strictly speaking is merely a doctrine of property (or world) co-variance, it is now ubiquitously understood to include a ontological grounding condition construed as ontological dependency between supervenient properties that are constituted by subvenient entities and sui generis emergent properties whose conditions for their initial exemplification and continued sustaining are grounded in their associated subvenient base.

Third, in philosophy of science during the first half of the twentieth century, emergent properties were defined epistemologically:1 Property P is an emergent property of some particular x at level ln just in the case when P is a property of x at ln, and no amount of knowledge of (or descriptive statements about) entities at subvenient levels below ln would justify a prediction of (or logically entail a descriptive statement about) P at ln. In this sense an emergent property (or a statement about it) is surprising and unexpected relative to knowledge of (or statements about) lower levels.

Since the 1960s, emergent properties have been understood in a straightforward ontological way (see below) with the result that Nagelian and property reduction came to replace linguistic and semantic reduction.

Philosophy of mind followed the same path. From the 1920s through the late 1950s. linguistic reduction (e.g., as in Analytic Behaviorism) or semantic reduction were replaced by the two ontological reductions mentioned above. For example, in the 1960s, Type Identity Physicalism employed non-synonymous, co-referring expressions to flank the identity symbol in contingent identity statements (e.g., painfulness = being a C-fiber firing). Thus, the issue of whether properties of consciousness were emergent or reducible was taken to be an a posteriori issue; except for rare exceptions, gone was the importance of analytic/synthetic physicalism or naturalism.

To be sure, conceptual analysis continued to be used as one tool among others, but the resulting conceptual equivalencies or lack thereof were examples of use and not mention—they were means to get at the relevant features of the mind-independent world. Consequently, a failure to achieve a conceptual analysis that establishes synonymity between two concepts or terms is not taken to show non-identity of the associated referents. The important issue in any purported reductive claim regarding properties and their instances was whether one is faced with a genuine emergent property or with a case where the property turns out to be identical to one employed in the hard sciences.

This shift away from the analytic/synthetic focus was facilitated even more with the publication of Saul Kripke’s Naming and Necessity.2 Beginning with the early modern empiricists, those with an empirical bent (with apologies to Kant and his synthetic a priori propositions) held to two types of propositions each of which was characterized by three clustered kinds: (1) Analytic, necessary, a priori; (2) Synthetic, contingent, a posteriori. Type (1) propositions were analytic truths, and thus, they were a priori and necessary. Type (2) propositions were synthetic truths and, thus, were a posteriori and contingent. Kripke famously showed that there were contingent a posteriori and necessary a priori truths. The synthetic/analytic distinction became less important, and the debate about property reduction vs. emergence became more directly focused on the ontology involved. Properties, not property-talk, or entities in the world, not semantics, now constitute the primary focus, even if conceptual analysis is a tool among others to get at that ontology. Some still focus on synthetic/analytic physicalism, but this is often associated with debates about a priori/a posteriori physicalism.

To summarize, I take this statement from leading physicalist Andrew Melnyk to be canonical:

The claim that a particular mental property is identical with a particular physical or functional property is not advanced as any kind of a priori claim, e.g., as a conceptual or analytic truth. It must be supported by empirical evidence. For example, it is argued that the identification of mental property M with physical property P would provide the best explanation of an observed correlation between instances of M and instances of P. Physicalism is claimed to be analogous to such a posteriori scientific hypotheses as that genes are segments of DNA or that chemical elements are systems of physical particles (Melnyk 2019).

It is time to apply these insights to a set of issues that must be addressed before I can proceed with my approach to Shafer-Landau.

(i)

Do issues about emergent properties generalize to other areas of philosophy, especially those in metaethics? For two reasons, the answer is yes. First, discussions of emergence began in the philosophy of science, and debates about property emergence vs. Nagelian or property reduction surfaced in a wide range of topics, e.g., biology, systems theory, chemistry, and philosophy of mind. Second, Frank Jackson’s defense of staunch naturalism (see below) addressed alleged emergent properties in four areas of philosophy: indexical properties associated with the first-person point of view, properties of consciousness, secondary qualities, and normative, e.g., moral properties (Jackson 1998; Sturgeon 1986).

(ii)

Some claim that Shafer-Landau seldom, if ever, talks about emergent properties in developing his ontology for moral realism. Thus, critiques of emergent properties (see below) are beside the point. I acknowledge his lack of explicit reference to emergent properties. However, I think that is because his discussion fails to appropriate insights from philosophy of science and mind that are relevant to his views. Discussions about ethical non-naturalism within metaethics seldom bring issues about emergent properties into their discussions.

More importantly, emergent properties are implicit in his ontology. I take his approach to be a specific version of what Bernard Linsky and Edward N. Zath call Platonized Naturalism (Linsky and Zath 1995). This is a view according to which (a) when viewed from the perspective of abstract objects (e.g., properties construed as abstract universals), various concrete objects or events exemplify abstract objects. (b) When viewed from within the Standard Mereological Hierarchy and the perspective of the associated objects or events, when the right physical conditions obtain, the coming-to-be and continued ontological grounding of the exemplified property are understood as emergent properties ontologically dependent on the right subvenient bases.

Shafer-Landau explicitly states that moral principles and specific moral properties are exemplified to form moral facts construed as instances of the principles and properties. He cites specific cases of the right kind in which non-moral facts regularly exemplify these properties (Shafer-Landau 2004). It is clear that the coming-to-be-exemplified and the continued ontological sustaining of these properties are grounded in specific kinds of non-moral facts. If this is correct, then moral properties and their fact-instances meet the conditions for being emergent prope rties.

There is an additional consideration that supports Shafer-Landau’s employment of emergent properties, viz., his constitutive non-naturalism according to which the rightness or goodness of anything will be brought about by the descriptive facts that constitute it in a particular instance.3 It is the particular arrangement of descriptive features that causes something to have moral status. How does this happen? While a conjunction of moral principle/property and non-evaluative, descriptive facts explain the moral status of a moral fact, only the latter are causal. Moral properties are epiphenomenal. But moral properties supervene and are not type identical to their subvenient purely descriptive properties and property-instances. Thus, moral properties are irreducibly emergent epiphenomenal properties that emergently supervene on the appropriate factual subvenient bases. This is exactly what those in philosophy of science and mind mean by emergent properties.

In an important article, Nicholas L. Sturgeon defends a view quite similar to Shafer-Landau’s with an important difference. Sturgeon’s main goal is to provide defeaters to two claims made by Gilbert Harman: Alleged moral facts are irrelevant to explaining our moral beliefs and observations because (a) moral explanations are of a kind that does not make it possible to test moral claims empirically in ways scientific claims can be tested; (b) moral claims can be tested empirically if we have in hand a natural reduction for moral claims.

Among his responses, Sturgeon tweaks one of Harman’s thought experiments and uses it against him. Harman invites us to consider the following: Albert is beating a cat who should have known better than to stray into his yard. Jane sees him and believes that what he is doing is wrong. Now, says Harman, it may initially seem plausible to hold that part of the reason Jane has this belief is due to the moral fact that Albert’s action is wrong. But this would be mistaken because there is nothing about the occurrence of Jane’s belief that requires anyone else to believe in the postulated moral fact. Thus, the moral explanation provides no useful test of whether Albert’s action is wrong. Irreducible moral facts (if such there be) are epiphenomenal and explanatorily useless.

Among his responses, Sturgeon presents an altered version of the thought experiment. Everything is kept as it was except now another person—Mary—is introduced into the picture. Mary has never thought that there is anything wrong with beating animals. However, based on observational evidence, Mary has come to believe that Jane’s moral judgments are quite good and correct in most cases. She also holds that Jane is a good person with a deep moral life and her moral sensitivity is reliable. Based on this observational evidence, Mary comes to believe what Jane believes about Albert’s actions. Mary did not accept this as true to begin with. Observational evidence changed her mind. Mary did not have a reductive analysis in hand, and her moral beliefs were caused to change by empirical evidence just as scientific beliefs are. Thus, Harman’s two theses are false.

Harman replies that irreducible moral properties/facts are epiphenomenal and, thus, can never be the cause of one’s moral beliefs. Perhaps descriptive wrong-making features of certain acts can explain what we observe, but the wrongness of those acts is explanatorily irrelevant.

Sturgeon digs his heels in and argues that epiphenomenalism in these cases is wrong. The supervenience of moral facts upon non-moral facts does not render the former epiphenomenal. Rather, the supervenience of moral facts is like the supervenience of biological facts upon chemical and physical ones, and this is a kind of “causal constitution of the supervening facts out of the more basic ones, which allows them a causal efficacy inherited from that of the acts out of which they are constituted.”

There are two crucial mistakes in Sturgeon’s final argument that informs Shafer-Landau’s project. First, as noted above, philosophers of mind have written volumes on constitution-supervenience which are exactly like Sturgeon’s characterization of them. Supervenient properties are constituted by their subvenient base entities and a new structural arrangement. Above, I called this structural supervenience. Unfortunately, if all supervenient properties are structural ones, then one’s ontology will be that of the hard sciences and bereft of emergent properties, including moral ones. In this case, sans emergent moral properties/facts, ethical non-naturalism reduces ontologically to ethical naturalism irrespective of the absence or presence of linguistic or semantic reduction.

Second, absent causal overdetermination, if there is either an irreducible mental or value description or a genuinely mental or value property that is an aspect of an object/event that also has a physical description or ontological aspect, then the mental component is causally epiphenomenal and does no work in explaining one’s action. For example, if a state of feeling thirsty emerges on a brain state of certain neurons firing, and one is thereby caused to get a drink, it was the neurons firing that caused the action. The feeling of thirst was epiphenomenal. Or so most naturalists would argue. If this is so, then Sturgeon’s employment of his views of supervenience, etc., is hard to harmonize with his claim that epiphenomenalism is false.

(iii)

Objection: Employment of an emergence theory for moral properties would be a form of naturalism. Since Shafer-Landau in an ethical non-naturalist, this is something he simply would not do. We have already seen reasons for thinking that Shafer-Landau does, in fact, quantify over emergent moral properties, and I will not rehearse those. But what should we make of this objection?

I think a simple distinction is sufficient to defeat this claim. We need to distinguish ethical non-naturalism from worldview naturalism as it applies to genuinely emergent properties in many areas of philosophy. Ethical non-naturalism amounts to the rejection of ethical naturalism and its claim that the ontology of the hard sciences is all there is: alleged emergent entities must be reduced to that ontology or eliminated. As we will see below, ethical non-naturalism is an expression of staunch naturalism applied to moral entities.

By contrast, ethical non-naturalism embraces the ontology of the hard sciences but goes on to adopt a richer ontology that includes, among other things, properties of consciousness and intentional action, along with moral/value properties and principles. The mistake made by the objection under consideration is to think that adopting ethical non-naturalism amounts to an abandonment of worldview naturalism, but this is simply a mistake. As we will see, Faint-Hearted Naturalism is a widely accepted version of worldview naturalism according to which the ontology of the physical sciences provides an inventory of all the fundamental entities in reality, but genuinely emergent properties (e.g., being painful, being good) also exist and are ontologically dependent on or grounded in physical reality. Thus, Shafer-Landau’s ethical non-naturalism is a version of Faint-Hearted Naturalism and in no way an abandonment of naturalism tout court.

A consideration of the following statement by Shafer-Landau in light of the next section on worldview naturalism makes this point, at least as far as I can see.

In effect, what the moral skeptic is giving us is a universe exclusively regulated by scientific laws, containing only the things that are ratified by science. Ethical objectivists give us a fuller, more expansive ontology. This includes all that science does, but adds an additional layer—moral principles and facts.

2. The Nature of Naturalism as a Worldview: Preliminary Remarks

2.1. Staunch vs. Faint-Hearted Naturalism

In this section, I will take both versions of naturalism to be robust, ontological doctrines. I have already explained above why most philosophers of science and mind are not particularly interested in analytic or synthetic naturalism, though a small minority remain. My purpose here is to take insights from philosophy of science and mind about the nature of naturalism, emergence, and property reduction, and use them to assess Shafer-Landau’s ethical non-naturalism.

Staunch naturalism is the view that the cosmos and all entities within it—e.g., particulars, properties, relations, events, processes—are and only are physical. Strong physicalism adopts the same ontology concerning consciousness, the human person, and other entities studied in the philosophy of mind. While he does not use the term, Galen Strawson’s understanding of standard physicalism and his claim that it is the ubiquitously adopted view is consistent with my claims about staunch naturalism: “I take [standard] physicalism to be the view that every real, concrete [i.e., spatio-temporally located] phenomenon in the universe is …physical” (brackets mine).4 According to Strawson, standard physicalism entails (NE): “Physical stuff is, in itself, in its fundamental nature, something wholly and utterly non-experiential.” (Ibid., p. 11.)

Staunch naturalists/strong physicalists eschew sui generis, irreducible, ineliminable emergent properties.5 Faint-hearted naturalism and weak physicalism differ only in that they embrace emergent entities, usually properties, but in a few cases, systems and even substances.

Strictly speaking, neither version of naturalism makes any reference to the existence or non-existence of abstract entities. However, with supplementation, we get Platonized Faint-hearted Naturalism = _def. a version of Platonized naturalism according to which certain abstract entities—relevant properties, laws/standards, and relations—are exemplified, and when exemplified, are understood as emergent properties—more plausibly, emergent property-instances—located in the spatio-temporal, physical, causal realm. Such properties, laws/standards, and relations can be known by human cognizers.

2.2. Emergent vs. Structurally Supervenient Properties

There is an important distinction between emergent and structural properties/supervenience. Emergent properties (EPs) are completely unique, new kinds of properties different from those that characterize their subvenient bases. Moreover, EPs are simple, novel properties different from and not composed of subvenient entities. Being painful is an example. Keep in mind that in the first half of the twentieth century, emergent properties were defined epistemologically:6 Property P is an emergent property of some particular x at level l_n just in the case where P is a property of x at l_n, and no amount of knowledge of (or descriptive statements about) entities at subvenient levels below l_n would justify a prediction of (or logically entails a descriptive statement about) P at l_n. In this sense an emergent property is surprising, unexpected, and utterly brute. Thus, the prior probability of emergent properties, given complete knowledge of lower levels, would be zero.

By contrast, a structural property (SP) is constituted by a new configurational pattern of subvenient parts, properties, relations, and events. It is not a new kind of property; it is merely a new pattern of subvenient entities. In what follows, I will set aside global supervenience and employ only strong property supervenience with a modality of (de re) metaphysical necessity.7

3. Why Contemporary Naturalists Ought to Be Staunch Naturalists (and Most Are)

Space considerations forbid me from developing the points in this section. Fortunately, I have done that elsewhere (Rickabaugh and Moreland 2023), so here I will provide a precis that I trust has enough details for the reader to get the important ideas and their relations to each other. Among other things, what follows makes explicit and provides justification for why J. L. Mackie advanced his argument from queerness against non-natural moral properties and facts (Mackie 1977). The ontological component of his argument entails the rejection of the exemplification of moral properties and the obtaining of moral facts due to their being utterly different from any other properties and facts (ibid.). This conviction led him to affirm that “Moral properties constitute so odd a cluster of properties and relations that they are most unlikely to have arisen in the ordinary course of events without an all-powerful god to create them” (ibid.).

3.1. The Inner Logic of Scientific/Philosophical Naturalism

Intellectually responsible naturalists cannot merely deny God’s existence. Additionally, they must provide an account of what ideas naturalists ought to hold regarding epistemological commitments, a broad creation story (the Grand Story) about how all entities have come-to-be, and a resulting ontology such that all entities can be located in the Grand Story as certified by naturalist epistemological commitments.

Just exactly what are the central features of contemporary naturalism? (Rosenberg 1996) Though versions vary, there is a specific form that is rightly enjoying widespread hegemony. Further, by clarifying the relationship between a naturalist ontology, epistemology, and creation account, a picture will emerge regarding what ought to constitute that ontology. This picture places a substantial burden of proof for alternative naturalist ontologies bloated beyond what is justifiable within the constraints that follow from a naturalist epistemology and creation account.

Fundamentally, naturalism is the view that the spatio-temporal universe of entities postulated by our best current (or ideal) theories in the physical sciences, particularly physics, is all there is. Scientific naturalism includes: (1) different aspects of a naturalist epistemic attitude; (2) an etiological, microphysical, combinatorial account of how all entities have come-to-be, constituted by a bottom-up, event-causal story (especially the atomic theory of matter and evolutionary biology); and (3) a general ontology with and only with entities bearing a relevant similarity to those in a completed form of physics. Whether this ontology should be expanded to include sui generis emergent properties or abstract objects, will occupy our attention shortly.

The ordering of these three ingredients is crucial. The naturalist epistemology justifies the etiology which together justifies and places constraints on ontological commitments. Moreover, naturalism requires coherence among these three areas of the naturalistic turn.

Thus, in setting up his naturalist project, David Papineau sets philosophy within science—philosophical investigation should be conducted within the framework of our best empirical theories. It follows, says Papineau, that “…the task of the philosophers is to bring coherence and order to the set of assumptions we use to explain the empirical world” (Papineau 1993).

Hence, there should be a coherence among third-person scientific ways of knowing; a physical, evolutionary account of how our sensory and cognitive processes came-to-be; and an ontological analysis of those processes. Any entities that are taken to exist should bear a relevant similarity to entities that characterize our best (or ideal) physical theories, their coming-to-be should be intelligible in light of the Grand Story, and they should be knowable by scientific means.

3.2. The Naturalist Epistemic Attitude

First and foremost, naturalism expresses an epistemic posture, specifically, scientism. Wilfrid Sellars expressed this posture when he said that “in the dimension of describing and explaining the world, science is the measure of all things, of what is that it is, and of what is not that it is not” (Sellars 1963). Steven Wagner and Richard Warner claim that naturalism is “the view that only natural science deserves full and unqualified credence” (Wagner and Warner 1993).

Most contemporary naturalists embrace either weak or strong scientism (Moreland 2018). According to the former, nonscientific fields offer some intellectual results, but they are vastly inferior to science in their epistemic standing. According to the latter, unqualified cognitive value resides in science alone. Either way, naturalists are extremely skeptical of any reality-claims that are not justified by scientific methodology in the hard sciences. That methodology is a third-person one, and the entities justified by it are capable of exhaustive description from a third-person perspective. Entities that require the first-person perspective as their basic mode of epistemic access are met with skepticism.

Naturalists believe this posture is justified by the success of science relative to other fields of inquiry. In addition, some naturalists justify this standpoint by appealing to the unity of science. As John Searle notes, since for these naturalists, science exhausts what we can know, then belief in the unity of science entails a belief in the unity of all knowledge because it is scientific knowledge:

Every fact in the universe is in principle knowable and understandable by human investigators. Because reality is physical, and because science concerns the investigation of physical reality, and because there are no limits on what we can know of physical reality, it follows that all facts are knowable and understandable by us.

(Searle 1992)

For such naturalists, the exhaustive or elevated nature of scientific knowledge entails that either the only explanations that count or the ones with superior, unqualified acceptance are those employed in the hard sciences.

Second, scientific theories that are paradigm cases of epistemic/explanatory success (e.g., the atomic theory of matter, evolutionary biology) employ combinatorial modes of explanation. Thus, any process that constitutes the Grand Story and any entity in the naturalist ontology should exhibit an ontological structure analyzable in terms that are isomorphic with such modes of explanation. Colin McGinn has defended this idea along with what he takes it to entail, viz., the inability of naturalism to explain EPs:

Can we gain any deeper insight into what makes the problem of consciousness run against the grain of our thinking? Are our modes of theorizing about the world of the wrong shape to extend to the nature of mind? I think we can discern a characteristic structure possessed by successful scientific theories, a structure that is unsuitable for explaining consciousness… Is there a “grammar” to science that fits the physical world but becomes shaky when applied to the mental world? Perhaps the most basic aspect of thought is the operation of combination. This is the way in which we think of complex entities as resulting from the arrangement of simpler parts. There are three aspects to this basic idea: the atoms we start with, the laws we use to combine them, and the resulting complexes… I think it is clear that this mode of understanding is central to what we think of as scientific theory; our scientific faculty involves representing the world in this combinatorial style.

(McGinn 1999)

We have seen that many philosophers express different aspects of the naturalist epistemic attitude. Let us turn to an overview of the Grand Story.

3.3. The Naturalist Grand Story

I use “the Grand Story” for the naturalist account of how all things came-to-be.8 Speaking generally, all of reality—space, time, and matter—came from the original “creation” event, and the various stars, galaxies, and other heavenly bodies developed combinatorially as the expanding universe went through various stages. On at least one of those heavenly bodies—earth—some sort of pre-biotic soup scenario explains how living things came into being from non-living chemicals. The combinatorial processes of evolution gave rise to all life forms, including human beings. Thus, all organisms and their parts exist and are what they are because they contributed to (or at least did not hinder) the struggle for reproductive advantage.

Note three important things about the Grand Story. First, at its core are two theories that exhibit combinatorial explanation: the atomic theory of matter and evolutionary theory.

Second, it expresses a scientific version of philosophical monism with two features: (1) An acceptable explanation must come from the hard sciences, and must show why something, e.g., a state of affairs, had to happen, given certain conditions. For example, PV = nRT is a lawlike description of an ideal gas (P, V, T and n are pressure, volume, temperature and the number of moles of gas, respectively; R is a constant.). But it is not a scientific explanation. For that we must employ ideal gas theory that, in turn, explains why the pressure of a gas must increase with an increase in temperature at constant volume. (2) No non-physical entities exist, including EPs.

When naturalists embrace EPs, however, they must depict them as being rooted in, emergent from, dependent upon, and necessitated by their subvenient physical states constituting the Grand Story. Further, their origin and continued existence are brute facts for which, in principle, there are no explanations. Later, we examine whether EPs should be allowed in a naturalist ontology.

Third, the Grand Story is constituted by bottom-up, event causality, eschewing irreducible teleology and agent causation in which the first relatum of the causal relation is a substance and not an event. Further, the Grand Story is deterministic in two senses: diachronically such that the state of the universe at any time t, coupled with the laws of nature, determine or fix the chances of the state of the universe at subsequent times; and synchronically such that the features of and changes regarding macro-wholes are dependent on and determined by micro-phenomena.

3.4. The Naturalist Ontology

(a)

Faint-Hearted vs. Staunch Naturalism One More Time

Characterizing a naturalist ontology requires distinguishing staunch (SN) and faint-hearted naturalism (FN). I describe that distinction above, but since it is so important for what follows, I provide a precis here to get this distinction before the reader. Staunch naturalists (e.g., David Papineau) accept a strict/strong version of physicalism (all individuals, events, states of affairs, properties, relations, and laws are entirely physical) for the natural world, while faint-hearted naturalists (e.g., John Searle) embrace various EPs.

(b)

The Location Problem

Let’s focus on criteria for naturalist ontological commitments. The canonical place to start is with what Frank Jackson calls “the location problem” (Jackson 1998). According to Jackson, given scientism, naturalists are committed to a widely accepted physical story about how things came-to-be (the Grand Story) and what they are. This generates the location problem: the task of locating some entity in that story.

For Jackson, the naturalist must either locate a problematic entity in the basic story or eliminate it. Roughly, an entity is located in the basic story just in case it is entailed by that story. Entailment is accomplished by some sort of reduction. Jackson provides examples of location. First, just as density is a different property from mass and volume, it is not an additional feature of reality over and above mass and volume because an account of things in terms of mass and volume implicitly contains, i.e., entails, the account in terms of density.

More importantly, Jackson focuses on the location of macro-solidity. He acknowledges that prior to modern science there was a widely accepted commonsense notion of macro-solidity, viz., being everywhere dense. However, due to modern science, this notion has been replaced with being impenetrable. So understood, macro-solidity may be located in the basic micro-story: given a description of two macro-objects in terms of their atomic parts, lattice structures, and sub-atomic forces of repulsion, this description entails that one macro-object is impenetrable with respect to the other. Here, SE and EP are reduced to weak supervenience and structural properties, respectively. This move relieves the naturalist from embarrassing—inexplicable—brute entities and preserves the correct ontology developed within the naturalist epistemology and Grand Story. In turn, this (allegedly) retains the naturalist claim to explanatory and epistemic superiority.

Jackson says that the naturalist must locate (reduce) or eliminate four troublesome entities: mental properties/events, first-person indexical facts, secondary qualities, and moral properties.

3.5. The Logic of the Standard Mereological Hierarchy

What has come to be known as the Standard Mereological Hierarchy has become the ubiquitously accepted world picture allegedly bequeathed to us by the hard sciences. Speaking of the hierarchy, Jaegwon Kim asserts what he takes to be obvious and without need of defense:

Today, any proposed general ontology of the world…is defined by its relation to materialism, the position that the world consists exclusively of bits of matter and structures made up of bits of matter, all behaving in accordance with physical law. Everything is an arrangement of matter and living organisms and minded creatures are no exceptions.

(Kim 2018)

To elaborate, the Standard Mereological Hierarchy is the picture within which philosophers must work. Construed ontologically, it is an ascending, hierarchically arranged taxonomy of individuals, properties, and relations. The fundamental level is microphysical entities. Call this level n. Focusing on living organisms, in the category of individual, the hierarchy ascends through level n+1, n+2, etc.—atoms, molecules, cells, living organs/systems, living organisms, psychologically endowed living organisms, ecosystems. The relation between a level and the one above it is the standard part/whole relation of extensionalist mereology. Moreover, level n is fundamental and not ontologically dependent on any lower level since there is none. Each level above n lists individuals that are dependent on the level below it in two ways (Inman 2019):

Essentially₁ Grounded: (IG) x is essentially₁ grounded in y = _def. There is a two-place predicate “F” such that it is part of the identity of x that x is related by F to y.

Rigid Existentially Grounded: (RG) x is rigidly grounded in y = _def. @ (Ex→Ey). (NB. The @ should be replaced with the box).

To clarify (IG), let x be some aggregate of parts, y be one of y’s parts, and F be the part-whole relation or, perhaps, its converse. In (RG), “E” is “exists” and the arrow is strict entailment. (RG) means that y’s existence is necessary for x’s existence. Each principle in its own way implies mereological essentialism for individual wholes at level n+1 relative to n as they are depicted in the hierarchy.

In the category of property, all properties above level n are structural properties composed of the parts at the lower level standing in a new structural arrangement constituted by numerous instances of external relations between and among those parts (e.g., being H₂O, being a homo sapiens). In the category of relation, all relations are external.

It is important to keep in mind that IG and RG are expressions of ontological grounding or (non-causal) metaphysical priority or dependency. For example, RG captures the idea that it sometimes happens that one thing (x) may depend on another specific entity (y) for its existence. It is metaphysically impossible for x to exist without that very entity y existing. The word “rigid” is meant to capture the idea that x cannot exist without that specific y such that if there is a replacement of or variation within y, x ceases to exist. The classic example of RG is mereological essentialism applied to mereological aggregates.

Jackson grasps the connection between accepting the epistemic superiority of naturalism and deciding between SN and FN and the importance of working within the limits set by the Standard Mereological Hierarchy. For Jackson, epistemic superiority entails SN.

This inner logic of naturalism implies at least three ontological constraints:

• Entities should conform to the naturalist epistemology.
• Entities should conform to the naturalist Grand Story.
• Entities should bear a relevant similarity to those found in chemistry and physics or be capable of one-to-one or one-to-many correlation with entities in chemistry/physics or depend necessarily on entities in chemistry/physics. This is an expression of microphysical priority.

Regarding the naturalist epistemology, all entities should be subject to combinatorial modes of explanation, be entirely publicly accessible, be knowable entirely by third-person scientific means, and be knowable by faculties formed by the evolutionary struggle for reproductive advantage. Regarding the Grand Story, one should be able to show how any entity had to appear, given combinatorial, microphysical, event-causal processes. At higher levels above the fundamental one (if there is such), mereological aggregates and structural properties fit nicely into the Grand Story and satisfy the epistemological criterion of simplicity compared to substantial selves (or homeodynamic systems) and EPs. Thus, the first disjunct of (3) follows nicely from (1) and (2), but the second and third disjuncts do not.

Unfortunately, when construed in terms of EPs, the second disjunct of (3) “solves” the so-called hard problem of consciousness or the emergence of moral properties by simply naming the problem (they are “emergent”) or by employing weak supervenience and dismissing the need for further explanatory work. Further, the second disjunct fails to satisfy criteria (1) and (2). The third disjunct of (3) suffers from this latter problem and from difficulties with justifying the claim that emergent entities are “necessitated” by their subvenient microphysical bases.

4. Problems with Naturalistic Emergent Properties

4.1. The View

For those who accept the existence of genuine, irreducible, non-physical conscious states and, at the same time, affirm naturalism, the most popular current strategy is to adopt faint-hearted naturalism and weak (usually, nonreductive) physicalism, and appeal to some sort on natural emergence. According to the staunch-naturalist account, matter at the most basic level—the microphysical level—is or will be completely describable within an ideal physics. Staunch naturalist David Papineau’s view on this matter is canonical. While acknowledging that progress towards an ideal science may require categorial supplementation of some sort, he maintains the following: “What I do not believe is that they [the current categories of physics plus future supplementation] will need supplementation by psychological categories” (Papineau 1993, 2002) (brackets mine).

That ideal description will be able to capture the complete nature of basic matter, and it will entail that basic matter is bereft of any mental entities, including mental potentialities. So far, a faint-hearted naturalist would agree. However, they go on to affirm that when matter reaches a certain, “appropriate” level of complexity, a completely new range of sui generis emergent properties, especially conscious and moral properties, appears from a subvenient base wholly bereft of, say, consciousness/intrinsic rightness/goodness or potentialities for them.

4.2. Seven Criticisms of the View

For one thing, an appeal to emergent properties has always seemed suspect to us if it is offered as an explanation: “emergence” is not a solution but a name for the problem to be solved. It is a label and that is all. This is especially problematic in the current setting in which scientists and philosophers of science are increasingly elevating the importance of seeking mechanisms that explain how and why emergent properties appear under a specific set of base conditions (Tabaczek 2019).

In a closely analogous context, Jaegwon Kim has convincingly made the same charge against the widely employed, but illicit, inclusion of dependency in strong property supervenience (Kim 1998). He notes that such supervenience is merely a principle of property covariance and, as such, it is consistent with a number of disparate views on the mind/body problem, including some forms of property or substance dualism. According to Kim, supervenience is a name that provides no explanation of the problem to be solved, viz., the dependency and determination of the mental on or by the physical.

Kim’s solution to this problem is to adopt a version of physicalist functional realizationism, which entails that mental properties supervene on physical properties because physical properties realize mental properties. The point here is not to evaluate Kim’s solution; rather, it is to acknowledge his charge that slapping the label “supervenience” on the problem of mental dependency on the physical is no solution at all. Some further theoretical work is needed to satisfy that problem-solving need.

I am making the same claim about the naturalist employment of “emergence.” I recognize that supervenience tout court is not precisely analogous to genuine emergence, but I think that Kim’s treatment of strong property supervenience is appropriately analogous to our claim about “emergence.” In fact, employment of “emergence” is more egregious than using “supervenience” in the Kim discussion. Why? As Kim shows, at least the “supervenience” case allows room for the development of an explanatory theory that grounds and explains why supervenient dependency is true.

But this is not possible for emergent properties. Recall that until the middle of the twentieth century, emergent properties were characterized epistemically, e.g., as being unpredictable from exhaustive knowledge of base entities. Genuine emergence is the Platonic Form of being a brute fact, and as such, there is in principle no deeper explanation of emergent properties on offer. Consequently, the charge that emergence is just a name for the problem and not a solution seems justified.

In the present context, while I do not agree with his overall position, it is instructive to notice an important aspect of Colin McGinn’s Mysterian Naturalism (McGinn 1999).9 According to McGinn, given that there is no knowable, current naturalistic solution to the origin of consciousness or its regular correlation with matter, due to our cognitive limitations, he sets out two conditions for any solution to these issues: (i) It must be a naturalistic solution. (ii) It must depict the emergence of consciousness and its regular correlation with matter as necessary and not contingent facts.

More specifically, there must be three kinds of unknowable natural properties that solve the problem. First, there must be some general properties of matter that enter into the production of consciousness when assembled into a brain. Thus, all matter has the potentiality to underlie consciousness. Second, there must be some natural property of the brain he calls C* that unleashes these general properties under the right conditions. Third, just as the brain must have a hidden unknowable structure that allows consciousness to emerge from it, so consciousness must have a hidden unknowable essence that allows it to be embedded in the brain.

The important point about McGinn’s project is this: He recognized that a naturalist employment of “emergence” to solve the problems of the origin of consciousness and its regular correlation with matter is no explanation at all. It is what needs to be explained, and on this basis, McGinn offers a solution for the label. The simple fact that he does this demonstrates his recognition of the explanatory vacuity of the label “emergence”.

Second, as a label, it leaves open other empirically equivalent views, viz., that the new kind of property may reside in a new substance (see below) or be due to God’s regular intention that the property appear under the same set of circumstances. As Timothy O’Connor has noted, if the lawlike link between occurrent subvenient properties and their associated occurrent emergent properties is contingent, then the only adequate explanation for the link and the appearance of an emergent property is God’s direct activity and stable intention that things be so (O’Connor 2000). Now it seems to many, perhaps most, philosophers who do not start with a prior commitment to strong physicalism, that the link is indeed contingent, as seen in several well-known thought experiments.

Third, emergence seems to be a case of getting something from nothing, a case of magic without a Magician. If matter is relevantly similar to what current physics and chemistry tell us, and it is bereft of any sort of mental entities, then matter does not have mental potentialities. Given this, it is hard to see how the mere spatial rearrangement of, say, atomic simples to form a more complicated spatial structure would be an adequate cause for bringing into existence ex nihilo a completely new sort of entity.

Fourth, this “ex nihilo” problem generates another difficulty with emergent properties. To see this, consider the following tensed definition of coming-to-be (Chisholm 1989):

B_t “X comes into existence” = Df. There is a property P which is such that x has P, and there is no property Q which is such that x had Q.

Stated without tense, we have:

B_nt “X comes into existence at t” = Df. There is some property P which is such that Px at t, and for all times t’ prior to t there is no property Q such that Qx at t’.

Now, consider two contrasting views of the composition of time:

Instantism: Temporal intervals are derived and not fundamental entities. Intervals are wholly grounded in dimensionless instants.

Intervalism: There are extended temporal intervals that are fundamental in that they are not wholly grounded in the existence of other entities.

B_t and B_nt are meant to capture the idea that coming-to-be takes place at an instant of time and is not an extended temporal process or interval. For example, it is not a case of alteration or elanguescence (when something gradually transitions from existence to non-existence.) Further, these definitions, while consistent with Instantism, do not entail it. If, as seems likely, intervals have unextended boundaries at their beginning, then coming-to-be would occur at that boundary.

The important point about B_t and B_nt is the fact that coming into existence is not a process that could be governed by something, e.g., a (Non-Humean) law of nature. Thus, there is no naturalistic or scientific explanation whatsoever as to why one entity comes into existence at t as opposed to any other entity. Under the view being considered, emergence is a case of coming-to-be and, as such, it leaves as a bizarre brute fact that whenever matter is at such and such configuration, the same emergent property—e.g., being painful—always emerges. Besides the fact that the naturalist ontology is now chock full of emergent brute facts—an embarrassing state of affairs, given the naturalist’s favorite trump card of ontological parsimony—there is also the improbable coincidence, bordering on a miracle, that the same emergent property always seems to show up when the same physical conditions obtain, and in principle there is no explanation for this.

Galen Strawson has given considerable reflection to our last two points (Strawson 2006). He notes that when it comes to liquidity or heat (the objective physical phenomena and not our experience of them), we have a good story to tell as to how and why these (structural) properties regularly appear under certain circumstances. But when it comes to genuinely emergent properties—he focuses on experiential phenomenal qualia—due to the nature of emergence, there is in principle no story to tell.10 He argues that if it is actually the case that Y is emergent from X, then in the relevant sense, it must be the case that Y is wholly dependent on and only on X such that all the features of Y can be metaphysically intelligibly traced, in principle, back to X. If this is not true, then Y did not emerge solely from X.

Based on this, Strawson rightly asserts that emergence cannot be brute since, given the nature of brute emergence, there is absolutely nothing about the intrinsic or relational nature of X in virtue of which Y emerges. That is simply what it means for emergence to be brute. Brutality rules out nothing. Thus, Strawson concludes, emergence is a genuine miracle, not only because it is the appearance of something from nothing in virtue of which it appears, but given that whenever the “appropriate” physical conditions obtain, there is also the bizarre fact that the very same emergent property appears.

However, this is utterly ridiculous and, indeed, incoherent. In principle, there is no reason why, says Strawson, a negative number could not emerge from the addition of certain positive numbers. I could not agree more with his summary warning: “If someone says he chooses to use the word “emergence” in such a way that the notion of brute emergence is not incoherent, I will know that he is a member of the Humpty Dumpty army and be very careful with him” (Strawson 2006).

Since emergence provides no explanation for why either emergence obtains or there is a regular covariance between a physical base property and an emergent property, this leads to troublesome metaethical implications. Surely it is strongly conceivable that there are physical duplicate worlds of the actual world that resemble zombie worlds in philosophy of mind in that they have no moral norms or value properties. Moral/axiological inverted worlds seem similarly conceivable, worlds like ours in all physical respects but in which torturing children is right and caring for the poor is wrong. Further, there could be worlds in which moral properties, etc., emerge in a random, fluctuating basis such that kindness and cruelty alternate between being right or wrong. Surely, any metaethical theory that entails such possible worlds is deeply flawed.11

Fifth, sorites problems are lurking in the neighborhood. Could this emergent property, say being painful or moral rightness, be instantiated with one less atomic part in the subvenient base when that base is the “right” level of complexity? Surely the answer is yes. How about two fewer atomic parts? And so on. Sans ontological vagueness, at some point, the emergentist must say that the subtraction of one small atomic part has a huge, disproportionate metaphysical effect on the disappearance of the emergent property or properties, e.g., all the irreducibly mental properties constituting consciousness. Grant that humans are physical objects of some sort that exemplify intrinsic goodness. Given the sorites problem, at some point the removal of one tiny atom out of myriads causes that intrinsic goodness to cease being exemplified. Surely, such a dramatic ontological effect cannot result from such a tiny, seemingly insignificant cause. Keep in mind that the sorites problem here is ontological and not epistemological or conceptual.

But how can such a significant metaphysical effect be due to such an insignificant cause? And why was the “right” level of complexity one atomic part greater than the actual level? And the very notion of “the right level of complexity” is completely vacuous. Why? What is the right level? Answer: Whatever the level is when the emergent property appears. When does the emergent property appear? When the right level of complexity is reached. There are no independent criteria for identifying the “right level of complexity” besides the appearance of the so-called emergent property.

Sixth, David Chalmers has raised an additional difficulty with SE, given some form of panpsychism or (we add) strict Microphysicalism—the Quality Combination Problem (viz., plausibly explaining how microqualities combine to yield macroqualities, e.g., phenomenal and moral ones?) of which one especially difficult version is the Palette Problem (viz., given the highly likely fact that there is a limited number of microqualities but a large number of macroqualities, how do so many get yielded from so few resources?) (Chalmers 2017).

Chalmers considers unfavorably two attempts to resolve the Palette Problem: small-palette solutions (viz., somehow, all macroqualities can be generated from just a few microqualities, if we find the right ones with sufficient flexibility and generality) and large-palette solutions (viz., the full range of macroqualities are included among the microqualities). Against the former, Chalmers argues that it is pure speculation and still faces the Quality Combination Problem (viz., how do microqualities combine to yield simple macroqualities?).

Against the latter, Chalmers argues that there are, in fact, only a small number of causally efficacious microqualities; the rest will be epiphenomenal. To deny this, one would have to deny the simplicity of physics with a small number of properties and laws. Says Chalmers, “Large-palette solutions seem once again to be stuck with a form of epiphenomenalism or radical revision of the fundamental dynamics of the physical world.” (Ibid, p. 206.)

Seventh, Ben Page has advanced a Bayesian argument to the effect that the obtaining of the correct conditions for the emergence of consciousness (and, presumably, moral properties) in the actual world is substantially more likely, given theism compared to atheism (Page 2020). Thus, these emergent properties (if there are such) confirm theism over atheism.

The prior probability of conscious (and intrinsically valuable) beings is higher on theism than on atheism. Why think this? Page argues that given theism, God would have several reasons for creating conscious (and intrinsically valuable) beings and bring about whatever conditions that would do the job: It seems likely that God would want to have love relationships with creatures, to share His knowledge and what it is like to experience various things with them; being creative, God desires to create all kinds of beings, those like human persons being one kind, and so on. But on atheistic hypotheses about the actual world, there is no such being who wants to bring about creatures like us. There is no teleological driving force behind the laws of nature that would be directed towards bringing about the conditions required for our coming-to-be.12 Rather, on atheist worlds with laws like ours, it is extremely unlikely that those conditions would obtain.

5. Ontological Difficulties with Shafer-Landau’s Ethical Non-Naturalism

In this section I will present three features of Shafer-Landau’s ontology relevant to his ethical theory and provide criticisms of each after they are presented. I will close this section by offering one general problem with his entire approach.

5.1. Three Features of Shafer-Landau’s Ontology13

5.1.1. Non-Natural Moral Properties and Facts

Schafer-Landau (Shafer-Landau 2003) holds to the existence of non-natural moral and value properties, facts, truths, moral principles/rules/standards that are objective and non-perspectival. Moral standards are general conditionals linking moral facts with moral evaluations (e.g., taking someone’s possessions without their permission is wrong.). A particular moral fact obtains as a result of the conjunction of moral standards and specific non-moral facts (e.g., Jane’s action last night of taking Bill’s property without his permission was wrong). Worlds without agents have no moral facts but moral standards still exist in those worlds. Additionally, he advocates for non-reductive moral non-naturalism according to which moral properties exist and are not identical to natural scientific properties. Finally, moral properties and facts are epiphenomenal.

There are several problems with this aspect of Shafer-Landau’s ontology, but since some appear below, I will postpone my responses to them until later. Here I make one criticism. I agree with Christian Miller who notes that by restricting moral realism to only being a view about the objectivity of moral standards, Shafer-Landau’s moral realism is both too broad and too narrow (Miller 2006). It is too narrow because it leaves out views that are versions of moral realism. For example, a moral realist could deny the existence of moral standards and accept the reality of objective moral facts. It is too broad because it includes views that are not versions of moral realism. Shafer-Landau’s position would wrongly seem to count versions of ethical expressivism14 as moral realist positions.

5.1.2. Universal Moral Laws and Absolutism

According to Shafer-Landau (2004), there are universal moral laws (standards, rules) in the sense that they apply to everyone who can understand them whether they do or do not endorse them. These laws exist in all possible worlds, including those without agents. Ethical objectivism entails this view of “universal” and adds that moral laws are claims about what people ought to do. The content of moral rules is fixed independently of our opinions. Ethical objectivism is about the status of moral laws; absolutism is about the stringency of moral laws (there are moral laws that should never be broken). Shafer-Landau accepts ethical objectivism and remains neutral about absolutism.

As I see it, the fundamental problem with this aspect of Shafer-Landau’s position is that it is a paradigm case of begging the question against alternative views, especially theistically-grounded metaethical views as well as versions of ethical naturalism. He could respond that his criticism of moral skepticism justifies some version of objectivism and that it is a self-evident Moorean fact that values and moral standards and properties are normative. I will examine this latter claim below when I assess Shafer-Landau’s epistemology. For now, I want to address his critique of moral skepticism in defense of objectivism.

The main problem here is that Shafer-Landau seems to know only about moral skepticism as a skeptical-threat argument, and he addresses all of his criticisms to this interpretation. Applied to moral claims, if successful, a skeptical-threat argument demonstrates that there is no such thing as an objectively true moral proposition. An undercutting defeater of such an argument shows that there is no good reason to believe moral skepticism. A rebutting defeater seeks to show that there are sufficient reasons to reject moral skepticism.

But there is a second and, in my view, better interpretation of the point of moral skepticism in this context, especially as employed by theistically grounded metaethical theorists. This is the Inference-to-the-Best-Explanation (IBE) approach according to which one grants the existence of objective moral rules and value properties and goes on to seek among a pool of explanatory options the best explanation for their existence.

I acknowledge that Shafer-Landau criticizes theistic ethics. But for two reasons, I don’t think this carries much weight in this discussion. First, as far as I know Shafer-Landau has not provided a serious interaction with what is widely acknowledged as an explosion of sophisticated literature in natural theology defending the existence of God (Craig and Moreland 2009). Thus, if the moral argument for God is the sole focus of critique as it seems to be with Shafer-Landau, that critique may be plausible within that limited scope, but it could still be the case that belief in the existence of God is epistemically justified, all things being considered. If so, then God’s existence is still within the pool of options for an IBE regarding the existence of objective moral rules and values. An IBE approach to moral skepticism seeks to argue that moral facts/principles and value properties are much more to be expected based on theism that on naturalism; thus, theism has a better response to moral skepticism. Moreover, some of the arguments for God’s existence (e.g., the argument from reason) provide defeaters for Shafer-Landau’s critique of theistic ethics. In light of the contemporary situation, it is question-begging for Landau to assert his ethical non-naturalism and the success of his critique of theistic ethics. That assertion is premature.

Second, the lion’s share of his defense of ethical objectivism relies on his critique of moral skepticism, and as we have seen, he fails to address the IBE interpretation of this view.

5.1.3. Where Do Moral Standards Come From?

Shafer-Landau breaks this question into two different issues (Shafer-Landau 2004). First, to the simple question of where such standards come from, he holds that no one made them up since not every law requires an author. Second, to the question of whether moral standards are eternal or just popped into existence, he acknowledges that they certainly seem to be eternal and to have existed before humans (agents) came to be. But, he says, this seems implausible.

To address this implausibility, he argues that there are two strategies one can employ. First, one can deny that an authorless morality must be eternal laws. After all, laws governing DNA did not exist before DNA existed in organisms, and we can in principle date the origin of such laws as the time when DNA first existed. The same goes with moral laws. Second, one can deny the absurdity of eternal moral laws by recalling the distinction between moral principles, conditionals such that if certain conditions obtain, then a certain moral verdict follows (e.g., if x kills, then x did something wrong) and moral facts.15 By contrast, moral facts are the realization of moral principles, the instantiation of goodness/rightness/intrinsic virtue. Moreover, he adopts a Platonist view of principles according to which some are never exemplified. Thus, moral principles are eternal but moral facts are not.

What should we make of these claims? It will come as no surprise that I do not think they are plausible. First, consider the assertion that there are authorless moral laws. Moral laws are prescriptive and, as such, we discover them to have imperatival force and impose duties on us. However, imperatives and duty-impositions come from persons with wills. I suppose one could respond that they are grounded in emergent normative properties. I have already presented several serious problems with natural emergence, so I do not think this is a plausible response for a naturalist. Regarding an objective non-natural moral order, I think evolutionary naturalist Michael Ruse got it right when he noted.

Morality is a biological adaptation no less than are hands and feet and teeth. Considered as a rationally justifiable set of claims about an objective something, ethics is illusory. I appreciate that when somebody says ‘Love thy neighbor as thyself,’ they think they are referring above and beyond themselves. Nevertheless, such reference is truly without foundation. Morality is just an aid to survival and reproduction… and any deeper meaning is illusory (Ruse 1989).

As far as I can tell, our uniform experience of the origin of laws with imperatival force and duty-imposition come from persons, and Shafer-Landau’s authorless moral laws run against this uniform experience. To be sure, moral laws differ from laws of government, custom, or other social groups. What they all have in common, however, is imperatival force and duty-imposition. Surely this problem places a significant burden of proof on Shafer-Landau’s authorless moral laws. Moral norms exhibit a high degree of incumbency when compared to norms in law or custom; accordingly, they transmit a kind of authority and accountability that is hard to conceive as being generated by impersonal sources.

Second, consider Shafer-Landau’s commitment to the existence of intrinsic value properties—especially objectively good virtues—in worlds without agents. Now I have defended a hybrid of Platonism and Aristotelianism about universals elsewhere (Moreland 2013; Kim 2011). Here I want to call attention to a relevant issue. I grant that there are Platonic properties in that they can exist without being exemplified. Examples would be being a triangle, being such and such color. So far, I agree with Shafer-Landau.

However, there are other properties that are Aristotelian in that they require at least one particular to exemplify them in order to exist. Chief among these are properties that require conscious being to have them. They conform to the old adage that necessarily, thoughts have thinkers. Among these properties are being a thought, desire, sensation, and being loving, kind, just, truthful. Take the virtue of being loving. It is hard to conceive of how such a property could simply exist as an impersonal entity. It is hard to know what that would even mean, but it is not hard to see how being loving could exist if and only if it is exemplified by a subject of consciousness.

Since being loving does not seem to depend for its existence on any or all finite, contingent beings, there would seem to be a transcendent Subject of consciousness who exemplified being loving before humans appeared or in agentless, personless worlds. It seems far more reasonable to take the existence and exemplification of being loving as a property of a person and a mode of personal reality as opposed to being an impersonal stand-alone property whose exemplification is a mode of impersonal reality.

Third, Shafer-Landau is guilty of a fundamental ontological conflation of realization and exemplification. He explicitly identifies the two, but anyone familiar with general ontology knows to keep them distinct. As Jaegwon Kim canonically observed, concepts are realized by items that play a role specified by the concept. By contrast, properties are exemplified by things and characterize their being (Kim 2011). Consider NFL quarterback Patrick Mahommes. He has a number of properties that characterize him and, arguably, are constituents within his being: being a certain height, mass, shape, color, and so on. Does Mahommes also have the property of being a quarterback? No. Being a quarterback is a concept, a functional role he realizes, plays, satisfies. It is not a property he exemplifies. If an ideal observer inventoried Mahommes’ properties, being a quarterback would not be included. Schafer-Landau seems unfamiliar with this widely acknowledged distinction, and by equating them, it makes one wonder just how much work he has done in basic ontology.

Finally, consider his assertion that authorless moral laws need not be eternal by citing cases in which a law (e.g., governing DNA in organisms) comes into existence when the item it governs first comes to be. Surely this is another confusion, especially given his Platonic views that some principles (e.g., of geometry) and some properties (e.g., triangularity) could exist without ever being exemplified. It seems obvious that within Shafer-Landau’s own ontology, it is more consistent and avoids laws popping into existence from nothing if one holds that in such cases, a law is first exemplified rather than first coming to be. If I am right about this, then the ontological status of such laws (and properties) would seem to be described as abstract entities, entities that are timeless and spaceless. Abstract entities are eternal in that there never was a time when they did not exist.

It is important to keep in mind an important distinction between a traditional realist—e.g., Platonist—construal of properties and property-instances. The former are universals taken as abstract necessary beings in that they exist in all possible worlds. Viewed ontologically, the latter are particulars; viewed diachronically, they are instances of emergent properties. Thus, realist understandings of properties are fully consistent with taking them as emergent properties. While abstract, when the “right” subvenient conditions obtain, the property is instantiated and when viewed from the subvenient/supervenient perspective, the property and its instance are emergent.

5.2. The Problematic Nature of Contingent Brute Facts

In Section 3, I provided a detailed critique of emergent properties given naturalism. Here, I want to revisit that topic and focus on a different difficulty that emergent properties present for Shafer-Landau’s ontology, one that targets a fundamental aspect of his view. That aspect involves the problematic nature of brute facts or entities. Alexander Pruss has provided a thorough critique of contingent brute facts which he takes to be true propositions (Pruss 2009). I have no problem with this but in what follows, I will take a brute fact to be a state of affairs, e.g., Joe’s being in pain, Arthur’s action being wrong. Shafer-Landau shares this view of moral facts.

Contingent brute facts have two important characteristics: (1) Qua brute, there is in principle no explanation for their obtaining. (2) Qua contingent, they could have not obtained, and others could have obtained instead. Given (1), the obtaining of a contingent brute fact is utterly inexplicable. Moreover, if such a fact could appear brutely then it becomes inexplicable why just any and every fact does not appear brutely. Finally, the regular obtaining of the same type of brute fact when certain conditions obtain (e.g., people being in pain given that their C-fibers are firing) is so improbable that it seems to be almost incoherent and miraculous.

In this regard, as Timothy O’Connor has noted, if the lawlike link between occurrent subvenient properties and their associated occurrent emergent properties is contingent, then the only adequate explanation for the link and the appearance of an emergent property is God’s direct activity and stable intention that things be so (O’Connor 2000). If O’Connor is right, as I think he is, this would be bad news for Shafer-Landau’s appropriation of contingent brute facts.

Regarding (2), the obtaining of a contingent fact cries out for an explanation. Indeed, throughout history, the search for an explanation for the obtaining of contingent facts is what has constituted part of the very notion of what it is to search for knowledge and explanations. When someone announces that the contingent fact is brute, that is tantamount to giving up on the explanatory enterprise.

Necessary facts and entities do not need an external explanation for their existence. Indeed, the search for one is a category fallacy. The existence of such entities flows from their nature as being necessary.

By way of application to Shafer-Landau’s ontology, let us grant that when certain states of consciousness obtain (e.g., being in pain, intentionally murdering an innocent person), they de re metaphysically necessitate the relevant moral/value properties (being right, being wrong, being good, being bad). So far so good. The problem is that the conscious states of personal agents are utterly contingent. As far as I can tell, the obtaining of moral properties (which is a moral fact) for Shafer-Landau amounts to moral properties being second-order ones that are exemplified if and only if a relevant first-order property of consciousness is exemplified. This is why he holds that in agentless worlds there are no moral facts, namely, there are no morally relevant conscious facts in those worlds (e.g., evil intentions).

The fact that the obtaining of the properties of consciousness is contingent is supported by two lines of reasoning. First, Shafer-Landau seems to agree with me that they are emergent properties and, as such, since their obtaining is brute, one cannot simply announce that they are somehow necessitated by subvenient conditions.

This line of reasoning needs supplementation by my second point: The contingency claim is based on the widespread and, in my view, successful employment of a host of thought experiments (e.g., the knowledge argument, the possibility of absent or inverted qualia, zombie worlds, and disembodied existence) that highlight the contingency in question. In regard to the employment of these thought experiments to support the reality and contingency of qualia, Kim observes.

The [property-dualist] case against qualia supervenience therefore is not conclusive, though quite substantial. Are there, then, considerations in favor of qualia supervenience? It would seem that the only positive considerations are broad metaphysical ones that might very well be accused of begging the question (Kim 2011).16

I conclude that Shafer-Landau’s ontological support of his version of ethical non-naturalism rests on the reality of a relevant range of contingent brute facts or entities. I have argued that this is a serious problem for his overall position.

6. Difficulties with Shafer-Landau’s Moral Epistemology

Earlier, I referred to the ontological aspect of J. L. Mackie’s argument from queerness. This aspect is widely known and debated. However, the argument also contained a lesser-known epistemological component, viz., knowing such queer entities would require “some special faculty” of moral perception or intuition, utterly different from our ordinary ways of knowing everything else (Mackie 1977). This “special-faculty” problem has caused several philosophers to adopt an externalist, usually causal or reliabilist accounts of modal knowledge and justification. In this last section, I want to expand on this problem and add further considerations about the epistemological difficulties with Shafer-Landau’s epistemology. To accomplish this, I will provide a precis of Shafer-Landau’s view, followed by an evolutionary argument against his and similar epistemologies. Next, I will consider and provide defeaters for the main ethical non-naturalist rejoinder to evolutionary arguments.

6.1. A Precis of Shafer-Landau’s Moral Epistemology

Shafer-Landau adopts a combination of internalist foundationalism (he is tentative about this and limits its range) and reliabilism. Roughly, we can know moral truths because some of them are self-evident, and we can employ moral reliabilism to extend our knowledge to non-self-evident moral truths (Shafer-Landau 2003, 2004). Here is a precis of his position.

Step 1: Provide defeaters for constructivism and non-cognitivism.

Step 2: Moral knowledge begins with non-empirical, non-natural, perhaps, a priori access to moral facts and true moral propositions. Thus, we begin by assuming there are such self-evident moral facts and truths.

Step 3: Mere intuition fails due to the widespread reality of moral disagreement.

Step 4: Thus, a version of moral reliabilism is the proper way to move from knowledge of self-evident moral facts/truths to those that are not self-evident. This account of reliabilism is based on current empirical theories of moral psychology, and we arrive at an account of moral reliabilism without causation. A moral process is reliable if it triggers more true than false moral beliefs.

Before we proceed, a few words of evaluation are in order. First, in step 1, constructivism and non-cognitivism are far from exhaustive. This leaves several naturalistic alternatives as live options without any criticism of them.

Second, his employment of “self-evident” is ambiguous and confused. He claims that we must assume the reality of moral facts, yet he says that foundational ones are self-evident. He explicitly rejects the view that moral beliefs about act and trait tokens are inferred from self-evidently known moral principles. Along with this, he denies a role for intuition in moral knowledge. However, the ubiquitous way of understanding a self-evident truth p is to claim that p achieves this status as follows: p is justifiably believed or known if either p is adequately grasped and understood or p is appropriately attended to and is clearly and directly known by acquaintance.

Moreover, Shafer-Landau holds that either basis just mentioned is sufficient for justification/knowledge, although these are clearly internalist notions that Shafer-Landau explicitly rejects. Yet he fails to provide an alternative account of the nature and knowability of the self-evident. Finally, though he goes back and forth between de re knowledge of moral facts and de dicto knowledge of moral truths, it is clear from his rejection of intuition that de re knowledge by acquaintance is off the table. However, without such de re knowledge, it becomes opaque as to how one is to know the truth of associated de dicto propositions or the reliability of the faculties that trigger them. I will say more on faculty reliability below.

6.2. An Evolutionary Argument against Shafer-Landau’s Epistemology

An evolutionary argument against naturalist epistemology—including Shafer-Landau’s—seeks to undercut the assumption of the cognitive adequacy of the cognitive faculties we have. As we will now see, this assumption is a very problematic one for naturalists who are Darwinists.

As Mark Linville notes, “[T]he naturalist’s commitment to a Darwinian explanation of certain salient features of human psychology presents an undercutting defeater for our moral beliefs as a whole” (Linville 2009). According to Michael Ruse and E. O. Wilson, “Ethical premises are the peculiar products of genetic history, and they can be understood solely as mechanisms that are adaptive for the species which possess them….No abstract principles exist outside the particular nature of individual species” (Ruse 1989; Ruse and Wilson 1986).

This statement may be taken as canonical for evolutionary naturalists such as Shafer-Landau, and, in my view, Thomas Nagel has presented the most persuasive case for what follows from it (Nagel 1997). The central problem is that, given Evolutionary Naturalism, the fact that people have certain moral faculties or processes that bring about moral dispositions/beliefs and the reasons they have them are quite independent from whatever would make the dispositions proper and the beliefs true. Our moral dispositions/beliefs and their development are survival conducive, not truth conducive. If Darwinian processes had turned out such that we had evolved with different survival faculties/processes that produced other survival enhancing dispositions, our beliefs would have tracked those.

The fact that those beliefs do or do not correspond to a non-natural, non-causal, non-empirical, objective moral realm has nothing to do with the faculty/process-generated moral beliefs/dispositions we develop. Thus, Evolutionary Naturalism generates an argument for moral skepticism and, lacking any warrant whatsoever for belief in moral facts (values, duties), for agnosticism towards or the denial of irreducible moral facts. Given Evolutionary Naturalism, we have no reason at all to believe the world has anything other than natural facts and subjective reactions to them.

Thus, Evolutionary Naturalism undercuts practical reason, according to Nagel. Our moral dispositions/beliefs are not counterfactually sensitive. We would form our beliefs irrespective of their veridicality. They are mere expressions of contingent moral faculties/processes that bring about a contingent set of dispositions that evolved with no regard to their correspondence to an independently existing moral order; they are not genuine insights into a non-contingent realm of objective value and duty. Wilson chides “ethical philosophers who wish to intuit the standards of good and evil, but who do not realize that their moral sentiments really spring from the hypothalamus and the limbic system, not from veridical intuition into an independently existing, objective moral realm” (Rachels 1990).

Shafer-Landau takes himself to have objectively reliable moral processes and dispositions, and warranted, true moral beliefs that correspond to an objective moral realm. However, those dispositions and beliefs are there only because of their survival value. If there were no objective moral realm, or if the objective moral realm were constituted by different values and duties, the ones Shafer-Landau adopts would all be false and inappropriate. Nevertheless, he would hold them anyway due to their survival value. He would not exhibit counterfactual sensitivity towards the content of the objective moral realm he posits. Thus, even if it is there in the way he adopts, he would have no reason at all to believe it is there or that he has genuine insight into it. That is, indeed, a big problem.

The final defeater involves our apparent ability to know things about exemplification itself and abstract objects. As I have argued elsewhere, we have knowledge of abstract properties that is quite different from our knowledge of concrete property-instances. However, such knowledge seems to be ruled out by naturalism if, for no other reason than it involves the employment of a “spooky” non-sensory intuitive faculty that is impossible to locate in the naturalist ontology.17

Moreover, Laurence BonJour noted that it is virtually impossible to accommodate traditional realist properties (especially Platonic ones) with naturalist depictions of human cognizers as physical entities and the physical causal interactions that constitute cognitive processes such as perception (BonJour 1998). For the naturalist, knowledge of some entity e entails a standard physical, causal interaction between e and the knower, but such a causal interaction is impossible when e is a causally effete abstract object.

Shafer-Landau could respond by appealing to some cognitive faculty with the power of rational intuition capable of grasping abstract objects. Once again, this would be a case of helping himself to an entity that just cannot be located in the most reasonable naturalist epistemology, Grand Story, and resulting ontology, bereft of queer, brute, non-natural entities. As Thomas Nagel correctly observed, adopting the existence of the ability to exercise rational intuition or grasping abstract objects is one that is secularly uncomfortable; it should make naturalists nervous, and is suspiciously religious or quasi-religious. Thus, it brings up the “fear of God,” precisely because such abilities are much more at home in a theistic than an atheistic worldview (Nagel 1997). A naturalist appropriation of them is borrowed capital on the cheap.

Alternatively, Schafer-Landau could appropriate his moral reliabilism about knowledge of abstract objects. However, this move would face the undercutting defeater provided by the Nagelian evolutionary argument.

6.3. The Main Ethical Non-Naturalist Rejoinder to the Evolutionary Argument

In brief, the main rejoinder to the evolutionary argument goes like this: Our moral intuitions about a host of moral facts are so strong, that they provide defeater defeaters for the less epistemically weighty evolutionary argument. Maybe we just got lucky, but whatever the case, fundamental moral intuitions are self-evident and reliable moral processes clearly exist, and that is all one needs to appropriate to defeat the evolutionary argument.

Unfortunately, this argument is unsuccessful. The objection conflates and confuses two ways of presenting the evolutionary argument: a skeptical threat argument or an Inference to the Best Explanation (IBE). As a skeptical threat argument, evolutionary argument seeks to show that we should be skeptical that we actually have moral knowledge. The objection under consideration is the claim that, yes, the evolutionary argument should make us skeptical except for one thing: We do have moral knowledge and our warrant for knowing this is stronger that than for the evolutionary argument for skepticism. It seems that we just got lucky. This may or may not be a successful response to the skeptical-threat employment of the evolutionary argument, but I, and many theists, employ the IBE use of the argument. Here, one grants that we have moral knowledge and goes on to seek the best explanation for why this is true. The theist will argue that theism is a much better explanation of this fact than is the “we just got lucky” response. As Christian theist Dolores Morris astutely notes:

Yes! Absolutely! Objective moral values are clearly a part of reality… The question of course, is how this all came to be… These atheistic theories tell us that we ought to act morally, but they do so with a conception of the universe that makes it difficult to see why and how we should believe that these shared moral values have anything like true normative objectivity. Objective morality may not be impossible on an atheistic worldview, but it feels like an ad hoc addition.

(Morris 2021)

Morris goes on to say that objective morality is very much at home and natural in a theistic worldview.

6.4. Responding to Leibowitz’s Defeaters for Evolutionary Debunking Arguments

In an important article, Uri D. Leibowitz seeks to place advocates of evolutionary debunking arguments on the horns of a dilemma, and to provide defeaters for each horn (Leibowitz 2021). I cannot address everything in Leibowitz’s article, but fortunately, there is a subset of his article that is particularly relevant to defending my evolutionary argument against the possibility of objectively justified beliefs about true moral realist claims. It is important to note when observing the two arguments that constitute his dilemma’s horns, that while they both make reference to an inference to the best explanation, these arguments are skeptical-threat arguments that employ IBE considerations within certain premises to establish the skeptical threat. They are not IBE arguments as depicted above. In what follows, I will present Liebowitz’ two arguments followed by my responses to them.

Here is the first argument offered, to which Leibowitz’s responses are given as rebutting defeaters for this type evolutionary debunking argument. Let us call this debunking argument EDA #1:

• Evolution provides the best explanation of moral phenomena.
• Evolutionary explanations of moral phenomena do not require that our moral judgements are true.
• If (2), then the chance of any one of our moral judgements being true is slim.
• If the chance of any one of our moral judgments being true is slim, then even after rational reflection the chance of any one of our moral judgments being true remains slim.
• If even after rational reflection the chance of any one of our moral judgments being true remains slim, then each one of our moral judgments is unjustified.
• Therefore, each one of our moral judgments is unjustified.

Here is the second argument, to which Leibowitz’s responses are offered as undercutting defeaters for this type of evolutionary debunking argument. Let us call this debunking argument EDA #2:

• Evolution provides the best explanation of moral phenomena.
• Evolutionary explanations of moral phenomena do not require that our moral judgments are true.
• If (2), then we have no reason to believe of any one of our moral judgments that it is true.
• If we have no reason to believe of any of our moral judgments that it is true, then even after rational reflection we have no reason to believe of any one of our moral judgments that it is true.
• If even after rational reflection we have no reason to believe of any one of our moral judgments that it is true, then each of our moral judgments is unjustified.
• Therefore, each one of our moral judgments is unjustified.

What does Leibowitz say in response to these arguments. Regarding EDA #1, his basic rejoinder goes like this. Even if we grant that the argument succeeds in showing that some moral belief targeted by the argument is unlikely to be true, there is still another moral belief in the vicinity that escapes EDA #1. Suppose we judge that A is morally permissible, but considering EDA #1, we would be justified in believing the negation of our initial judgment, i.e., we would judge that it is highly likely that A is not morally permissible. Given a standard, commonsense understanding of moral terms of appraisal, this last judgement would amount to the judgement that it is highly likely that A is wrong. So, while EDA #1 may succeed in showing that some of our moral judgments are unlikely to be true, the argument supports that other related moral judgments are highly likely to be true. Thus, premise (6) is false.

Regarding EDA #2, Leibowitz offers two defeaters. First, he argues that it commits the well-known genetic fallacy. A belief’s origin and justification can come apart and evolutionary debunking arguments do not affect theories in moral epistemology that do not aim to show that moral beliefs are justified by tracking their causal origins.

Second, building on his first defeater, Leibowitz claims that even if EDA #2 succeeds in showing that we have no evolutionary grounds for believing our moral judgments to be true, there are other theories in moral epistemology that do show this. So, EDA #2 fails to establish the general claim made in (6).

I offer two general responses to Leibgowitz’s critique of these two debunking arguments. First, it may be that the specific people who support these two debunking arguments use them to undercut specific moral judgements and generalize to undercut each and every moral judgment. Call this the bottom-up approach according to which both arguments begin with defeating any particular moral judgment and, by generalization, it concludes by defeating each of our judgments.

Whether or not the bottom-up approach defeats evolutionary debunking arguments such as EDA #1 and EDA #2, I do not think that Leibgowitz’s responses defeat what I call top-down debunking arguments such as the one presented above by Nagel and which I advocate. According to the top-down approach, all of our moral judgments and beliefs are produced by unreliable faculties and processes. This is true for two reasons: (i) Our moral faculties/processes are counterfactually insensitive to truth or rationality conditions; thus, the latter are irrelevant to our moral judgments/beliefs. (ii) Evolutionary processes select traits solely because they aided (or did not hinder) the survival-enhancing 4 Fs: feeding, fighting, fleeing, reproducing. Truth and rationality are irrelevant.

Moreover, as Plantinga, Reppert, and many others have argued, the evolutionary accounts of our rational faculties/process—e.g., sensations, “intuitions,” reasoning and rational deliberation—are victims of the top-down evolutionary debunking argument just as much as our moral faculties/processes are (Plantinga 1993; Reppert 2009).18 For the purposes of assessing Leibgowitz’s responses, let us assume the point made by Plantinga and others.

Let us apply these general points to some of Liebgowitz’s specific arguments. Consider his counterargument to EDA #1 according to which the evolutionary debunking argument may defeat our justification for thinking A is permissible but in so doing, it justifies another belief—that A is wrong. However, if our moral faculties/processes themselves are unreliable, then once we judge that A is wrong, the evolutionary argument now makes it the case that it is highly unlikely that this judgement is true. By inference, it is now highly likely to be true that A is morally permissible. Thus, we have entered a skeptical dialectical loop from which there is no escape, or so it seems to me.

Next, consider the second counterargument to EDA #2 according to which even if some moral judgments are defeated by evolutionary debunking arguments, others which are the result of alternative moral epistemologies and not tracing a moral judgment to its origins would still be justified. The problem here is that if Plantinga et. al. are right, then this counterargument uses reason in formulating and judging to be reasonable some alternative moral epistemologies. However, given that all of our faculties/rational processes are unreliable, then the formulation of justified moral epistemologies falls prey to the general evolutionary debunking argument. Further, a major way of developing a moral epistemology is, at least in part, by employing reflective equilibrium. An important aspect of reflective equilibrium is an attempt to establish coherence among general moral theories, specific moral principles or more specific moral rules, and particular test cases involving an independent, prima facie justified moral judgement. This means that moral judgements and the moral faculties/processes that produce them are an uneliminable aspect of developing a moral epistemology. As such, this developmental task will be undercut by the evolutionary debunking argument, though, perhaps, in a different way than moral epistemologies that focus on tracing a moral judgment back to its origin are undercut.

The final counterargument against EDA #2 is the charge that it commits the genetic fallacy. This is, indeed, a fallacy. However, there are clear cases in which discounting the justification for some claim based on its origin is a proper way of reasoning and not a genetic fallacy. One place this happens is when the origin of a claim directly impacts it credibility. Consider this example. Suppose you are riding a train from London to Wales. While you are looking out the window, you see etched into the hillside an indented set of shapes: WALES IN 10 KM. Now suppose you knew that trains in the UK sometimes used signs and sometimes used digging out letters in hillsides to inform riders of the next place the train stops and how far away it is. In this case, you would be justified in believing that Wales was 10 km ahead.

But suppose that you found out from highly reliable sources that the upcoming sign about Wales had been randomly shaped by erosion and flooding. It could still be the case that Wales is in 10 km, but given the causal story about the origin of the hillside scratchings, you would no longer be justified in judging Wales to be coming up in 10 km. In a court case, if you learn that a witness has a conflict of interest regarding his testimony or that due to childhood abuse, he was a chronic liar, then his testimony might be true but given your information about the causal origins of the witness’s reliability, you would no longer be justified in believing his testimony.

I think that the top-down evolutionary threat argument is just like these. If a solid causal story about the processes that lead up to the formation of our rational and moral faculties/processes provides a significant defeater for the reliability of the beliefs or judgments they produce, then this counts against the justification of those judgments or beliefs. Arguing this is not to commit a genetic fallacy.

There is a final point I want to make that seems to me to be the most important one. Liebgowitz is arguing against evolutionary debunking arguments construed as skeptical threat arguments. His appeal to best explanation claims occurs with premises of the two arguments he seeks to defeat, but those claims are used to justify the idea that an evolutionary account of our moral judgments and beliefs is the best one we have and, thus, this account raises a skeptical treat against those judgments and beliefs which he seeks to rebut or undercut. For these two EDA arguments to count as examples of IBE, their structure needs to be different. An IBE would grant that our moral and rational faculties/processes are reliable and, thus, our moral judgments are generally true and justified. These are taken as givens. The issue becomes this: From within a pool of the most important positions that attempt to explain these givens, an inference is made to the best explanation for them. The two key members of that pool are naturalism—staunch and faint-hearted, sometimes supplemented with Platonism—and theism. I cannot undertake an assessment of those like me who claim that theism is the best explanation for these givens, but I offer three brief reasons for the theistic inference.

First, a number of facts must be true: people have consciousness and intentionality, rational powers, including the ability to be directly aware through rational intuition of important moral properties (e.g., rightness, goodness), they must have libertarian freedom to perform genuinely intentional actions and be responsible moral judgers and agents, and they must exemplify immanent teleology since reasoning—moral or otherwise—involves a deliberative process for the sake of obtaining true or justified beliefs.

Staunch naturalists deny all of these, and faint-hearted naturalists can accept them only by appealing to the contingent, brute fact of their emergence. However, such facts are in principle incapable of having reasons offered for their emergence since that is what it means for them to be brute. Their contingency cries out for an explanation for why they obtain rather than alternative states of affairs. For natural emergence advocates, since natural entities are characterized by the ontology of the (especially) hard sciences, the physical subvenient base from which these properties emerge are examples of coming-to-be ex nihilo. Moreover, their regular obtaining when the “right” physical conditions obtain is nothing short of a miracle. The theist is in no such pickle. The theist’s fundamental being is not strings, particles, mere stuff and the like, but the theistic God who exemplifies all of the features that constitute the relevant facts listed above that inform an IBE. Thus, in a theistic universe, it is easy to explain the existence of these facts, but naturalists cannot offer an explanation. They must simply point to a range of brute facts.

A Platonic naturalist may have an explanation for the existence of value properties as abstract objects. However, since the time of Plato, non-theistic Platonists have acknowledged that they have no explanation, e.g., no efficient cause, to explain why these abstract objects were ever exemplified. Plato saw this and appealed to the demiurge to solve the problem, a minimal theistic solution. Further, Platonic moral realism cannot explain these features of the facts mentioned above: libertarian freedom and responsible intentional action, immanent teleology. There is no explanation for what grounds equal human rights or why our various faculties are reliable.

Third, as noted earlier in this article, Ben Page has argued (successfully in my opinion) that on theism, the prior probability of the universe containing conscious—and I would add, moral—persons is much, much higher than on naturalism. I will not repeat his argument here.

In any case, when I provided a critique of Shafer-Landau’s moral epistemology, I explicitly claimed that I was using an IBE and not a skeptical-threat argument. Whether successful or not, this fact means that Liebgowitz’s article is irrelevant to my IBE argument.

I have provided a rational context for assessing Shafer-Landau’s ethical non-naturalism by presenting the inner logic of contemporary naturalism and highlighting the distinction between staunch and faint-hearted naturalism, arguing that the former is substantially preferable to the latter. Since Shafer-Landau is a faint-hearted naturalist, before we examine the specifics of his views, he faces a severe burden of proof, or so I argued. I sought to raise that burden of proof by providing a critique of emergent properties, given naturalism. Next, I presented Shafer-Landau’s ontology and epistemology and argued that neither is adequate in supporting his ethical non-naturalism. The reader may not be convinced by my case, but I hope it is powerful enough to give readers something to think about when assessing Shafer-Landau’s ethical non-naturalism.