Some argue that, while finite mental entities may be inexplicable on a naturalist worldview, they may be explained by theism, thereby furnishing evidence for God’s existence. In this chapter, I shall clarify and defend this argument from consciousness (AC) by describing three issues in scientific theory acceptance relevant to assessing AC’s force, presenting three forms of AC and offering a brief defense of its premises. Among other things, I hope to show that an important factor in theory acceptance—scientific or otherwise—is whether a specific theory has a rival. If not, then certain epistemic activities, e.g. labeling some phenomenon as basic for which only a description and not an explanation is needed, may be quite adequate not to impede the theory in question. But the adequacy of those same activities can change dramatically if a sufficient rival position is present. In chapter one, we saw reasons for a naturalist to deny the existence of emergent mental properties/events that followed solely from naturalism itself. In this chapter, we shall discover additional reasons for naturalists to eschew emergent mental entities that follow because of the presence of AC. The combined force of chapters one and two will place a severe (and increasing) burden of proof on any naturalist who seeks to reconcile the existence of emergent mental entities (from emergence2a to emergence3) with naturalism.
While theism and naturalism are broad worldviews and not scientific theories, three issues that inform the adjudication between rival scientific theories are relevant to AC. While these are neither individually necessary nor jointly sufficient to justify one theory over its rivals, they are important characteristic marks whose presence or absence carries a great deal of epistemic weight in theory adjudication. The first issue involves deciding whether it is appropriate to take some phenomenon as ontologically basic such that only a description and not an explanation for it is required, or whether that phenomenon should be understood as something to be explained in terms of more basic phenomena. For example, attempts to explain uniform inertial motion are disallowed in Newtonian mechanics because such motion is ontologically basic on this view, but an Aristotelian had to explain how or why a particular body exhibited uniform inertial motion. Thus, what is basic to one theory may be derivative in another.
Ontological basicality should be distinguished from pre-theoretical basi-cality. According to the latter, the pre-theoretical description of an entity’s nature is to remain in tact and the theoretician’s aim is to explain the entity’s origin or behavior but not to reduce it. Consciousness is ontologically basic for theism since it characterizes the fundamental being. The appearance of finite consciousness qua finite requires explanation and theism may employ the explanatory resources of its basic ontological inventory (e.g. consciousness in God) for that explanation. But consciousness per se is ontologically basic. Not so for a naturalist though he or she may treat consciousness as pre-theoretically basic. According to naturalism, consciousness is emergent, derivative and supervenient, and both its finitude (Why did the Grand Story lead to and through consciousness as opposed to taking alternative paths?) and intrinsic nature require explanation.
Issue two is the naturalness of an accepted entity in light of the overall theory (or research program) of which it is a part. The types of entities embraced, along with the sorts of properties they possess and the relations they enter should be at home with other entities in the theory, and, in this sense, be natural for the theory. Some entity (particular thing, process, property, or relation) e is natural for a theory T just in case either e is a central, core entity of T or e bears a relevant similarity to central, core entities in e’s category within T. If e is in a category such as substance, force, property, event, relation, or cause, e should bear a relevant similarity to other entities of T in that category. This is a formal definition and the material content given to it will depend on the theory in question. In chapter one, I argued that the basic entities constitutive of the Grand Story provide this material content for naturalism.
Moreover, given rivals R and S, the acceptance of e in R is ad hoc and question-begging against advocates of S if e bears a relevant similarity to the appropriate entities in S, and in this sense is ‘‘at home’’ in S, but fails to bear this relevant similarity to the appropriate entities in R.1 The notion of ‘‘being ad hoc’’ is notoriously difficult to specify precisely. It is usually characterized as an intellectually inappropriate adjustment of a theory whose sole epistemic justification is to save the theory from falsification. Such an adjustment involves adding a new supposition to a theory not already implied by its other features. In the context of evaluating rivals R and S, the principle just mentioned provides a sufficient condition for the postulation of e to be ad hoc and question-begging. Moreover, in the presence of such a dialectical situation, advocates of R are under intellectual pressure to treat e along reductive or eliminativist lines.
The issue of naturalness is relevant to theory assessment between rivals because it provides a criterion for advocates of a theory to claim that their rivals have begged the question against them or adjusted their theory in an inappropriate, ad hoc way. And though this need not be the case, naturalness can be related to basicality in this way: Naturalness can provide a means of deciding the relative merits of accepting theory R, which depicts phenomenon e as basic, vs. embracing S, which takes e to be explainable in terms that are more basic. If e is natural in S but not in R, it will be difficult for advocates of R to justify the bald assertion that e is basic in R and that all proponents of R need to do is describe e and correlate it with other phenomena in R as opposed to explaining e. Such a claim by advocates of R will be even more problematic if S provides an explanation for e.2
By way of application, consider the following argument presented by Evan Fales:
Darwinian evolution implies that human beings emerged through the blind operation of natural forces. It is mysterious how such forces could generate something nonphysical; all known causal laws that govern the physical relate physical states of affairs to other physical states of affairs. Since such processes evidently have produced consciousness, however construed, consciousness is evidently a natural phenomenon, and dependent on natural phenomena.3
Given the presence of theism and AC, it should be clear that this argument is ad hoc and question-begging, especially since it includes an acknowledgement of the unnaturalness of consciousness in light of the Grand Story (its forces and causal laws) and the rest of the naturalist ontology.
Issue three involves epistemic values. Roughly, an epistemic value is a normative property, which, if possessed by a theory, confers some degree of rational justification on that theory. Examples of epistemic values are these: theories should be simple, descriptively accurate, predicatively successful, fruitful for guiding new research, capable of solving their internal and external conceptual problems, and use certain types of explanations or follow certain methodological rules and not others (e.g. ‘‘appeal to efficient and not final causes’’). Studies in scientific theory assessment have made it clear that two rivals may solve a problem differently depending on the way each theory depicts the phenomenon to be solved.
Moreover, it is possible for two rivals to rank the relative merits of epistemic values in different ways or even give the same virtue a different meaning or application. Rivals can differ radically about the nature, application, and relative importance of a particular epistemic value. Thus, given rivals A and B, in arguing against B, it may be inappropriate for advocates of A to cite its superior comportment with an epistemic value when B’s proponents do not weigh that value as heavily as they do a different one they take to be more central to B. For example, given rivals A and B, if A is simpler than B but B is more descriptively accurate than A, then it may be inappropriate—indeed, question-begging—for advocates of A to cite A’s simplicity as grounds for judging it superior to B. I am not suggesting that rivals are incommensurable. In fact, I believe that seldom, if ever, is this the case. Only on an issue-by-issue basis can one appropriately make judgments about the epistemic impact of the conflict of disparate epistemic values.
For example, in philosophy of mind, property dualists will argue that descriptive accuracy is on their side since their position accurately captures the intrinsic features of mental states and this accuracy justifies viewing those features and not relational ones as what constitutes the essence of the mental states in dispute. Property dualists argue that descriptive accuracy is more important than simplicity considerations as employed by physicalist rivals because epistemic simplicity becomes a factor in selecting between rivals only after it is judged that ‘‘all things are equal’’ between those rivals. And based on descriptive accuracy, this is precisely what property dualists deny.
Ontological simplicity is another matter. Among other things, strong physicalists eschew irreducible, uneliminable mental properties and claim that ontological simplicity is on their side. Property dualists respond that this appeal to ontological simplicity is done at the cost of denying the obvious facts as they are accurately described from a first-person perspective. According to Kim, at this stage of the argument, ‘‘the only positive considerations [for strong physicalism] are broad metaphysical ones that might very well be accused of begging the question.’4
It is clear that an essential part of assessing this debate is an analysis of the different epistemic values in play and their employment by the disputants. Applied to theism, the AC, and the characterization of robust, positive naturalism in chapter one, the central epistemic values of robust naturalism—including ontological simplicity, epistemic preference for the third-person point of view, and so on—place severe intellectual pressure on naturalists to be strong physicalists. Theists have no such pressure, and one aspect of evaluating AC is the asymmetrical pressure to avoid irreducible mental properties or other mental entities. The pressure towards some form of reductionism or eliminativism flows from the very nature of naturalism itself and is exacerbated by the presence of theism and AC as a rival.
32 The argument from consciousness The argument from consciousness
Three forms of the argument
Theists such as Angus Menuge5, Robert Adams6, and Richard Swinburne7 have advanced different forms of an argument from consciousness for the existence of God. The argument may be construed as an inference to the best explanation, a Bayesian-style argument, or a straightforward deductive argument in which its premises are alleged to be more reasonable then their denials.
An inference to the best explanation begins with certain data to be explained (the existence of irreducible mental entities or their regular, lawlike correlation with physical entities), assembles a pool of live options that explain the data, and usually on the basis of certain criteria—e.g. explanatory scope, explanatory power by making the data more epistemically likely than rivals, being less ad hoc—one option is chosen as the best explanation of the data. According to AC, on a theistic metaphysic, one already has an instance of consciousness and other mental entities, e.g. an unembodied mind, in God. Therefore, it is hardly surprising that finite consciousness or other mental entities should exist in the world. However, on a naturalist view, mental entities are so strange and out of place that their existence (or regular correlation with physical entities) defies adequate explanation. There appear to be two realms operating in causal harmony and theism provides the best explanation of this fact.
Richard Swinburne draws a distinction between a C-inductive (one in which the premises add to the probability and, in this sense, confirm the conclusion) and a P-inductive (one in which the premises make the conclusion more probable than not) argument. Understood as an inference to the best explanation, I will try to show in this book that AC is at least a correct C-inductive argument, though as a part of a cumulative case, consciousness contributes to a P-inductive theistic argument.8
Construed as a Bayesian argument, assuming the presence of background knowledge, we have:
Pr(T/C)
Pr(T) x Pr(C/T)
Pr(T) x Pr(C/T) + Pr(: T) x Pr(C/: T)
T and C stand for theism and either the existence of conscious properties or their regular correlation with physical features. We will be assuming that naturalism and theism are the only live options under consideration (see below) and, thus, :T = N (naturalism).9
Relative to our background knowledge, Pr (T) is much higher than many naturalists concede. The problem is that many naturalists are either ignorant of or simply disregard the explosion of literature in the last twenty-five
years or so providing sophisticated and powerful justification for theism. And the face of Anglo-American philosophy has been transformed as a result. In a recent article lamenting ‘‘the desecularization of academia that evolved in philosophy departments since the late 1960s,’’ prominent naturalist philosopher Quentin Smith observes that ‘‘in philosophy, it became, almost overnight, ‘academically respectable’ to argue for theism, making philosophy a favored field of entry for the most intelligent and talented theists entering academia today.’’10 He complains that “Naturalists passively watched as realist versions of theism ... began to sweep through the philosophical community, until today perhaps one-quarter or one-third of philosophy professors are theists, with most being orthodox Christians.’’ He concludes, ‘‘God is not ‘dead’ in academia; he returned to life in the late 1960s and is now alive and well in his last academic stronghold, philosophy departments.’’ This explosion of Christian philosophy includes fresh, highly sophisticated defenses of theism. In chapter nine, we will explore why this massive proliferation of Christian theism in philosophy is largely ignored by naturalist philosophers as seen by, among other things, a nearly complete lack of interaction with sophisticated versions of theism or substance dualism in their writings.
Pr (C/T) is highly probable (> > .5). Richard Swinburne’s version of AC provides several grounds for this ranking of Pr (C/T). Here are two: First, given theism, mental properties are basic characteristics of the fundamental being that constitutes a theistic ontology, so the theist has no pressing issue regarding the existence or exemplification of the mental. Such is basic on theism. As a result, the theist is under no pressure to explain how the mental data of AC could exist in light of the Grand Story. Second, a basic datum of persons is that they are communal beings who love to share in meaningful relationships with others and who desire to bring other persons into being. Thus, theism would predict a proliferation of persons besides God Himself.
According to advocates of AC, Pr (:T) x Pr (C/:T) is highly improbable (< < .5). To see why, first recall that the formula is equivalent to Pr (N) x Pr (C/N). Let us set aside Pr (N) for the moment. Pr (C/N) is so low that it approximates to zero. Why? Recall that in the early days of emergentism in the eighteenth and early nineteenth centuries, emergent properties were characterized epistemically, viz., as those that were unpredictable, even from a God’s-eye perspective, from a complete knowledge of the sub-venient base. That subvenient base provided no explanatory or predictive grounds for emergent properties precisely as emergent entities. Now it makes no difference for the relevance of this point that today we construe emergent properties ontologically and not epistemically. Even on the ontological construal, emergent properties are completely sui generis relative to the entities and processes at the subvenient base. In this regard, the following characterization by Timothy O’Connor may be taken as canonical:
An emergent property of type E will appear only in physical systems achieving some specific threshold of organized complexity. From an empirical point of view, this threshold will be arbitrary, one that would not be anticipated by a theorist whose understanding of the world was derived from theories developed entirely from observations of physical systems below the requisite complexity. In optimal circumstances, such a theorist would come to recognize the locally determinative interactive dispositions of basic physical entities. Hidden from his view, however, would be the tendency ... to generate an emergent state.11
As we saw in chapter one, applied to mental phenomena, it is almost impossible for advocates of a naturalist worldview to avoid admitting that these phenomena are explanatorily recalcitrant for them and must be
admitted as brute facts. The sort of separable-part/whole framework
and type of structural change at the core of the Grand Story is simply inadequate in principle for explaining consciousness. And this is to admit that Pr (C/N) is very, very low indeed. In this case, the denominator in Bayes’ Theorem approaches the numerator and Pr (T/C) approaches 1. This is the claim of a Bayesian form of AC.
In response, a naturalist could argue as follows: It seems like the conclusion to draw from the fact that the explanatory connection between the natural world and consciousness is opaque is not that P(C/N) is low but that it is inscrutable. Suppose you think it is a brute, inexplicable fact that consciousness sometimes “pops into existence’ when matter-energy is arranged a certain way. You learn then of a parallel universe with very different physical laws. Somehow, you come to have exhaustive knowledge of the distribution of matter/energy in that world, though you know nothing so far about whether there is consciousness in that world. So you wonder: is there? One could easily think the right attitude here is agnosticism: If it is a brute, inexplicable fact that consciousness emerges on certain configurations of matter/energy, then one simply would not know whether consciousness emerges from the configurations of matter/energy in this parallel universe. But if so, then the right thing to think about P(C/N) isn’t that it’s low, but that it’s inscrutable.12
There are at least four things to say in response to this argument, and we are now in a position to understand the first rejoinder: The presence of theism and AC provide intellectual grounds for rejecting this move. Given the presence of AC as a rival to naturalism, the postulation of the appearance of consciousness as a brute, inexplicable fact is clearly ad hoc and question-begging. AC provides a clear and powerful explanation for finite consciousness. There is no good reason to postulate it as a brute fact, especially when it does not bear a relevant similarity to the rest of the naturalist ontology. Moreover, the Grand Story cannot explain it, and it is not fundamentally known from the third-person point of view as are the rest of the entities over which the naturalist quantifies.
This response is an example of a broader dialectic that theists often encounter in debates with atheists. The theist argues that the existence of God is the best explanation for P (the Big Bang, fine-tuning, the instantiation of normative properties, consciousness) and provides grounds for why this is so and for why atheism cannot adequately explain P. The atheist responds by suggesting that we hold the actual world constant including the reality of P, i.e., consider a duplicate world containing P, and just leave God out. Well then, he or she concludes, it looks like the existence of God is irrelevant to P. In light of what I have argued above in the context of scientific theory acceptance, this is argumentation by theft and not honest toil.
Second, naturalism itself provides intellectual pressure against brute, non-physical facts. Our knowledge of this world would give us positive reasons for not believing that irreducible consciousness would appear in it, e.g. the geometrical rearrangement of inert physical entities into different spatial structures hardly seems sufficient to explain the appearance of consciousness. Thus, naturalism itself provides positive reasons for rejecting irreducible consciousness and, thus, for rejecting the claim that its appearance is a brute, natural fact. Coupled with various epistemic constraints, e.g. ontological and epistemic simplicity, a commitment to serious metaphysics, the centrality of combinatorial modes of explanation, and an ontology exhaustively describable from the third-person point of view, naturalism entails an inherent drive towards some form of reductionism.
In a way, this entire book is an attempt to argue this point in the context of AC, and I shall offer detailed criticisms of the major non-theistic attempts to explain the appearance of consciousness or simply to label it as a brute fact. For now, I merely note that most naturalists who work in philosophy of mind hold to some sort of strong physicalism and rightly cast a suspicious eye towards those who allow for emergent mental properties because they correctly understand that the Grand Story nicely explains structural particulars and properties but not emergent ones. In addition, they realize that the postulation of consciousness as a brute fact leaves them vulnerable to alternative worldviews that offer an explanation for what they admit cannot be explained within naturalist constraints.
To cite one representative example, David Papineau warns that the naturalist ought to deny that mental properties are not identical to physical ones because if they do not, they will have to face the question
why does consciousness emerge in just those cases [where physical properties correlate with them]. And to this question, [weak physical-ism] seem[s] to provide no answer.
I suspect that many philosophers regard the inability to answer this question as the fatal flaw in the physicalist approach to consciousness.
Surely, they feel, any satisfactory philosophical view of consciousness ought to tell us why consciousness emerges in some physical systems but not others.
I think that physicalists should simply reject the question. ... [T]he physicalist should simply deny that there are two properties here.13
Third, given that ‘‘popping into existence’’ is instantaneous and not a process, it is not something that can be governed by anything, e.g. natural constraints. Natural constraints such as the laws of nature govern processes of alteration for entities that already exist. If an entity does not exist, there is nothing on which the constraints can operate. Thus, there could be no reason in principle as to why consciousness as opposed to an angel or a Toyota Camry would appear. Thus, the regular ‘‘popping into existence’’ of consciousness when matter-energy is arranged in a certain way would be sheer magic, indeed, magic without a Magician. In fact, the very idea of ‘‘matter-energy being arranged in a certain way’’ would be utterly vacuous as an expression of constraints for consciousness and could only be characterized in a circular way.
The naturalist may respond that there are certain cases in which such constraints can be conceived, e.g. there would have to be space before a spatial entity could pop into existence, so in this case it is plausible to think there could be constrains even if coming-to-be is not a process. But for two reasons, this response fails. First, it is not analogous with the origin of consciousness. The reason that a spatial entity could not pop into existence without there being such a thing as space is the same as why a square circle could not pop into existence. In both cases, one is dealing with a logically contradictory state of affairs—a spatial entity existing without space and a square circle—so it is not a constraint that resists something popping into existence that is in play; rather, it is the inconsistency in the nature of the relevant states of affairs themselves. But there is no parallel contradictory nature to there being consciousness in a physical world. In my view, the reason a naturalist should not accept the reality of irreducible consciousness is not that such a state of affairs is logically contradictory, but that matter is bereft of what is needed to ground its origin. If one claims that consciousness does not pop into existence, but rather, is actualized from mental potentialities in matter, then as we shall see in chapters four through seven, this move amounts to an abandonment of naturalism in favor of something else, e.g. panpsychism.
Additionally, consider a non-spatial world. With no adjustments to that world, it is true that a table could not pop into existence for the reasons just mentioned. But there is no reason why a table and the necessary spatial conditions could not jointly pop into existence. So even in cases where there are prima facie constraints on what can pop into existence, when a broader state of affairs in taken into consideration, the constraints may be otiose.
Fourth, armed with the distinction between a state of affairs seeming to be impossible to a subject S vs. a state of affairs failing to seem possible to S, one could argue that he has strong defeasible modal intuitions that it is impossible for consciousness to arise from matter by way of purely natural, physical processes. This strategy could be undercut by a counter-claim according to which one either raises skeptical problems with modal intuitions in general or simply denies the relevant intuitions. Regarding modal skepticism, I think that modal intuitions of the sort just mentioned are ubiquitous in philosophy and, in fact, employed by physicalists in supporting their own views (e.g. it seems impossible—in some modal sense or other—to most physicalists that such different states as mental and physical ones could causally interact with each other). Regarding a failure to have the relevant intuitions, one could respond that the dualist intuition is the pervasive, commonsense one and the physicalist fails to share that intuition only because of a question-begging prior commitment to physicalism.
Finally, there is a third form of AC. Taken as a straightforward deductive argument, AC becomes the following:
(1) Mental events are genuine non-physical mental entities that exist.
(2) Specific mental event types are regularly correlated with specific physical event types.
(3) There is an explanation for these correlations.
(4) Personal explanation is different from natural scientific explanation.
(5) The explanation for these correlations is either a personal or natural scientific explanation.
(6) The explanation is not a natural scientific one.
(7) Therefore, the explanation is a personal one.
(8) If the explanation is personal, then it is theistic.
(9) Therefore, the explanation is theistic.
Overview of deductive premises
The relationship among these three forms of argumentation is controversial, e.g. is inference to the best explanation (IBE) reducible to other, perhaps Bayesian forms of argumentation. I do not wish to enter that controversy because I think that a defense of AC can be formulated based on one of the three argument forms I am presenting or based on three independent arguments. I do believe, however, that by stating the argument deductively we gain clarity on the precise considerations that most likely provide the basis for an IBE argument or for assignment of probabilities to key factors in the Bayesian approach. Therefore, I shall develop and defend the deductive form of the argument in some detail.
In my view, premises (3) and (6) are the most crucial ones for the success of AC since they are the premises most likely to come under naturalist attack. Let us set them aside for the moment.
We are assuming the truth of premise (1), since all the naturalist rivals of AC we are considering agree with it. There have been a number of variants on or alternatives to (1) that have been cited as problems which science cannot explain but which can be given a theistic personal explanation:
(a) the existence of mental properties themselves;14
(b) the fact that mental properties have come to be exemplified in the spatio-temporal world;15
(c) the nature of the relation, e.g., causal or supervenient, between mental and physical entities since it is as inexplicable from a naturalist perspective as is Cartesian causal interaction because the problem consists in the nature of the relation itself, along with the desperate nature of its relata, and is not a function of the category of the relata;16
(d) the fact that certain particular mental events are correlated with certain particular physical events;17
(e) the fact that the correlations mentioned in d are regular;18
(f) the existence of libertarian freedom and the type of agent necessary for it;19
(g) the aptness of our noetic equipment to serve as truth gatherers in our noetic environment;20
(h) the evolutionary advantage of having mental states as opposed to the evolution of organisms with direct stimulus-response mechanisms that have no mental intermediaries.21
Even though we are assuming the truth of (1), I want to make a few observations about the dialectical status of property dualism. My main argument in this book is that naturalists should be strict physicalists and, given property dualism, we have evidence against naturalism and for theism. A naturalist may well respond with a hearty “So what!’’ because strict physicalism is clearly the case. Of course, I reject strict physicalism, but more importantly, I believe certain issues are conspicuous by their absence in defenses of strict physicalism or criticisms of property dualism. In addition, I want to get these issues before the reader. So if you wish, consider the next several pages to be an excursus, though a relevant one.
Property dualists argue that mental states are in no sense physical since they (or at least some of them) possess six features that do not characterize physical states:
(a) There is a raw qualitative feel or a ‘‘what it is like’’ to a mental state such as a pain.
(b) At least many mental states have intentionality—ofness or aboutness— directed towards an object.
(c) Mental states exhibit certain epistemic features (direct access, private access, first-person epistemic authority, are expressed in intentional contexts, self-reflexivity associated with ‘‘I’’) that could not be the case if they were physical.
(d) They require a subjective ontology—namely, mental states are necessarily owned by the first-person, unified, sentient subjects who have them.
(e) Mental states fail to have crucial features (e.g. spatial extension, location) that characterize physical states and, in general, cannot be described using physical language.
(f) Libertarian free acts exemplify active power and not passive liability.
A few observations about (a)-(c) are important. Regarding (a), I believe the so-called Knowledge Argument has been misrepresented in two ways: what Mary comes to know is usually understated and the argument really is not an argument at all. Let us consider these in order.
A standard presentation of the argument has it that Mary, a brilliant scientist blind from birth, knows all the physical facts relevant to acts of perception. When she suddenly gains the ability to see, she gains knowledge of new facts. Since she knew all the physical facts before recovery of sight, and since she gains knowledge of new facts, these facts must not be physical facts and given Mary’s situation, they must be mental facts.
To understand the richness of what Mary comes to know, we need to grasp the nature of self-presenting properties. Insights about self-presenting properties go back at least as far as Augustine, but, more than any other contemporary philosopher, Roderick Chisholm has done the best job of analyzing them.22 While Chisholm proffered slightly different definitions of a self-presenting property, the following is representative of his views:
P is self-presenting = Df Every property that P entails includes the property of thinking.23
To understand this definition, we need to get clear on Chisholm’s definitions of property entailment and inclusion.24 Properties may sustain these different relations to each other:
Inclusion: Property P includes property Q = Df P is necessarily such that whatever exemplifies it exemplifies Q.
Entailment: Property P entails property Q = Df P is necessarily such that, for every x and every y, if y attributes P to x, then y attributes Q to x or, alternatively, believing something to be P includes believing something to be Q.
Inclusion requires it to be the case that the very same entity that exemplifies a property, P, must also exemplify the property P includes. Determinates (being red) include their determinables (being colorful). The notion of entailment contains the concept of attribution. In an attribution, one ascribes a property to something, e.g. a judgment or belief that x is F is an attribution of being F to x. For example, the property of believing there to be red circles entails the property of believing there to be things that are red.
I mention Chisholm’s unique formulation of a self-presenting property to get before us a widely, though not universally accepted, characterization of it. Obviously, the notion has deep Cartesian roots in two aspects of Descartes’ thought. First, when Descartes characterized the essential attributes of mind and body as thinking (i.e. consciousness) and extension, respectively, the former satisfies the definition of being self-presenting and the latter does not. Second, Descartes’ approach to the mind was an expression of his view of epistemology as a distinctively first-person enterprise that, nowadays, would be described as a version of internalist foundationalism. On this view, a self-presenting property is that by way of which a first-person knowing subject can take other entities as objects of intentionality, including as objects of propositional knowledge. In different ways, self-presenting properties present their intentional objects and themselves to first-person knowing subjects. Thus, the twofold Cartesian roots of being self-presenting are such that, even though shared in his own way by Chisholm, they provide grounds for alternative formulations of being self-presenting other than those presented by Chisholm. For those who reject Chisholm’s formulation, the notion of a self-presenting property could still be cashed out in a way relevant to the Knowledge Argument.
Self-presenting properties characterize thoughts, sensations, and other states of consciousness. For example, the property of being-appeared-to-redly entails the property of being-appeared-to and the latter property includes the property of thinking. By ‘‘thinking,’’ Chisholm means the same thing as being conscious, and he claims that self-presenting properties are psychological or ‘‘Cartesian’’ properties. According to Chisholm, a thing is conscious if and only if it has a self-presenting property. From the fact that a person has a self-presenting property, it follows logically that the person is conscious but it does not follow that the person has any property that does not include consciousness.
Some and, perhaps, all mental states are constituted by self-presenting properties. I can be aware of the external, physical world only by means of my mental states, but I need not be aware of my mental states by means of anything else. In different ways, a self-presenting property presents to a subject the intentional object of that property and the self-presenting property itself. Such properties present other things to a subject intermediately by means of them, and they present themselves to a subject directly simply in virtue of the fact that he has them. In each case, in normal circumstances, it is by virtue of exemplifying the relevant self-presenting property that a subject comes to have knowledge by acquaintance with the intentional object of the property and with the property itself. Moreover, it is in virtue of exemplifying the relevant self-presenting property that a subject is prima facie justified in believing that the intentional object has a certain feature.
For example, it is by way of a sensation of red that one is aware of the surface color of an apple, but one is not aware of the sensation of red by way of another sensation. The red sensation makes the apple’s surface present to one by virtue of one having the sensation; but the sensation also presents itself directly to one without another intermediary. A person sees red by means of the sensation of red and is made directly aware of (but does not actually see in the same way he sees the red on the apple’s surface) the sensation itself by having that sensation. The person gains knowledge by acquaintance with the property of being red on the apple’s surface and the property of being-an-appearing-of-red. Further, it is on the basis of the latter, viz., that the apple’s surface seems red to one, that the person is prima facie justified in believing that the apple’s surface is red.
When a person exemplifies a self-presenting property, he is modified in some way. We may put this by saying that when a person has a red sensation, the person is in the state of being-appeared-to-redly. Suppose the light is such that an orange jar looks red to Smith. If Smith says the object is red, his statement is about the jar is false. If Smith says, ‘‘I seem to see something red’’ or ‘‘the jar appears red to me’’, what he says is true because he is reporting a description of his own sensation. He is not talking about the jar. Smith’s statements employ what is called a phenomenological use of ‘‘seems’’ or ‘‘appears.’’ When people use ‘‘seems’’ or ‘‘appears’’ phenomenologically, they use them to report their own description of their self-presenting properties, i.e. to report the private, directly accessed mental states going on inside them.
By way of application, if one is appeared-to-redly while looking at a red apple, then we may say that this self-presenting property has two intentional objects. First, there is the primary intentional object, the red surface, which is the object of a focused awareness, in this case, a sensory awareness. Such an object requires a distinct mental state to be purposefully directed upon it. However, the self-presenting property also presents itself to the subject as a secondary, peripheral object. The self-presenting property being-appeared-to-redly is one that the subject is aware of but not in the sense that it is the primary intentional object of a distinct mental act. As a secondary, peripheral object, the self-presenting property presents itself to a subject by simply being instantiated by that subject, and it may, though it need not be the object of a distinct mental act of attending to it. So understood, no infinite regress is present.
In order to apply adequately the notion of a self-presenting property to the Knowledge Argument, let us review three different forms of knowledge. In keeping with the epistemic role of self-presenting properties in the Knowledge Argument, dualists will, or at least should argue that these three forms of knowledge are irreducible to each other. The three kinds of knowledge are: 1) Knowledge by acquaintance: This happens when we are directly aware of something, e.g. when I see an apple directly before me, I know it by acquaintance. One does not need a concept of an apple or a knowledge of how to use the word ‘‘apple’’ in English to have knowledge by acquaintance of an apple. A baby can see an apple without having the relevant concept or linguistic skills. 2) Propositional knowledge: This is knowledge that an entire proposition is true. For example, knowledge that ‘‘the object there is an apple’’ requires having a concept of an apple and knowing that the object under consideration satisfies the concept. 3) Know-how: This is the ability to do certain things, e.g. to use apples for certain purposes.
We may distinguish mere know-how from genuine know-how or skill. The latter is know-how based on knowledge and insight and is characteristic of skilled practitioners in some field. Mere know-how is the ability to engage in the correct behavioral movements, say by following the steps in a manual, with little or no knowledge of why one is performing these movements. In Searle’s Chinese Room thought experiment, the person in the Chinese Room who does not know Chinese has mere know-how. A person in the room with a mastery of Chinese would have skill.
Moreover, since the Knowledge Argument captures a first-person internalist perspective regarding Mary as a knowing subject, advocates of the argument will claim that, in general, knowledge by acquaintance provides the epis-temic basis for propositional knowledge which, in turn, provides the epis-temic basis for genuine know-how, i.e. skill. It is because one sees the redness on the apple’s surface that one knows that the apple’s surface is red, and it is in virtue of one’s knowledge of the apple’s surface color that one has the skill to do things to or with that surface color.
We are now in a position to describe six different forms of knowledge that dualists should claim are central to the Knowledge Argument. To repeat the thought experiment, suppose Mary is a neuroscientist who lives thousands of years in the future. Mary knows all there is to know about the physics and neurophysiology of seeing. She can describe in complete detail what happens when light reflects off an object, interacts with the eye, optic nerve, and brain, and so on. However, suppose that Mary was born blind and, suddenly, gains sight for the first time and sees a red object. There will be some new facts Mary learns that were left out of her exhaustive knowledge of all the relevant physical facts prior to gaining vision. Since Mary knew all the physical facts before gaining sight, and since she now has knowledge of new facts, these facts are not physical facts; at least some are mental facts, facts involved in what it is like to see.
To expand the argument a bit, Mary comes to exemplify the self-presenting mental property of being-appeared-to-redly. In this way, Mary gains six new kinds of knowledge - she gains knowledge by acquaintance, and on that basis, propositional knowledge, and on that basis, skill regarding the color red. She also gains these three types of knowledge, along with a similar epistemic order among them, about the phenomenological aspects of her own red sensation. Moreover, it is in virtue of exemplifying the property of being-appeared-to-redly that Mary’s knowledge by acquaintance with redness itself is both possible and justified.
Mary now knows by acquaintance what redness is. Upon further reflection and experience, based on this knowledge by acquaintance, she can now know things like: 1) Necessarily, red is a color. 2) Necessarily, something cannot be red and green all over at the same time. 3) Necessarily, red is darker than yellow.25 Finally, based on this propositional knowledge, she gains skill about comparing or sorting objects based on their color, of how to arrange color patterns that are most beautiful or natural to the eyes, etc. Assuming a realist and not a representative dualist construal of secondary qualities, the three kinds of knowledge just listed are not themselves knowledge of mental facts, but the dualist can argue that they are forms of knowledge that can be gained only by way of mental states that exemplify the relevant self-presenting property.
She also gains knowledge about her sensation of red. She is now aware of having a sensation of red for the first time and can be aware of a specific sensation of red being pleasurable, vague, etc. For example, at the eye doctor, when someone reports a letter on the eye chart as appearing vague, he is accurately describing his sensation of the letter, not the letter itself. Indeed, the doctor can see that the letter on the chart has clear borders; however, he needs the person to tell him how it appears to him since the doctor has no access to the patient’s inner mental sensation. Mary could now report things like this about her red sensations.
Based on the knowledge by acquaintance just mentioned, she also has propositional knowledge about her sensations. She could know that a sensation of red is more like a sensation of green than it is like a sour taste. She can know that the way the apple appears to her now is vivid, pleasant, or like the way the orange appeared to her (namely, redly) yesterday in bad lighting. Finally, based on this propositional knowledge, she has skill about her sensations. She can recall them to memory, re-image things in her mind, adjust her glasses until her sensations of color are vivid, etc. Mary had none of this knowledge prior to gaining color vision. These are all examples of knowledge of mental facts.
In addition to the richness of what Mary comes to know, there is the debate about precisely how the dualist employs the Knowledge Argument. From what we have seen about self-presenting properties, I believe the argument is not an ‘‘argument’’ at all. As I am using “argument,’’ an argument begins with premises that are better known than the conclusion they support. However, the property dualist will, or at least should say that first-person awareness of phenomenal consciousness is epistemically primitive and provides the ultimate ground for the relevant properly basic beliefs. When one tries to establish the fact that an alleged primitive is, in fact, a primitive, there are two things she can do. She can show that a denial of the primitive leads to conclusions that are unacceptable or she can use examples that invite interlocutors to attend to the primitive carefully by becoming aware of it and the various relations it sustains to other things. I suggest the Knowledge Argument be understood in this way and not as an “argument.’’26
My purpose is not to defend the Knowledge Argument in detail, e.g. to defend against the claim that Mary already knew these things and merely gained a new way to access or talk about them. Rather, I am simply correcting what I believe to be two mistakes widely made in discussing the Knowledge Argument, viz., what Mary (allegedly) comes to know and the way dualists should present the ‘‘argument.’’
So much for (a). What about (b) and the issue of intentionality? The topic is too vast to cover in detail. However, it needs to be said that people regularly experience sequences of mental states that exhibit what Husserl called a fulfillment structure. In such cases, one forms a concept of something (e.g. that a specific book is waiting to be picked up in the university bookstore), and goes through a series of experiences (walking to the store instead of going swimming in the Pacific Ocean, entering the store, walking closer and closer to the relevant section of the store, seeing the book from a distance to being right next to it) that terminate in the experience of comparing one’s initial concept of the thing to the thing itself to see if there is a match.
We all experience fulfillment structures regularly. What is important about them is that from a first-person perspective, one is able simply to grasp one’s initial (vague) concept (Husserl called it an empty intention), understand (perhaps somewhat vaguely) what its intentional object is, infer a series of experiential steps that would help to verify or disconfirm the initial concept, and eventually compare the concept with the object. Fulfillment structures and the cognitive success to which they (often) lead require that whatever account of intentionality one gives, it must be such that (1) the essence of conceptual meanings can be grasped from the first-person perspective; (2) the intentionality of conceptual meanings and the nature of their objects can be grasped from a first-person perspective; and (3) the enduring I that remains identical through the series of mental events can experience the initial somewhat empty concept become fuller and fuller until the thing itself is present in knowledge by acquaintance.
I fear that most strong physicalist accounts of intentionality run amuck of these abilities. For example, certain functionalist reductions of inten-tionality entail the holism of the mental in that a specific mental state is individuated by its relation to one’s entire psychology. How anyone could know what concept one was entertaining at a given moment or grasp its intentional features becomes quite opaque for such a view. I think similar difficulties confront attempts to identify beliefs (or thoughts) with indicators (a.k.a. ‘‘indicator meanings’’ or ‘‘representations’’) that satisfy some sort of functional or causal criterion. I suggest that the nature of fulfillment structures provides a test that any view of intentionality must meet on pain of being disregarded as inadequate. While I cannot argue the point here, I believe that (at least) property dualism for the relevant mental states is the only adequate solution that meets this criterion.
Finally, we come to (c), the fact that mental states exhibit various epis-temic features. It is fairly typical for physicalists to attempt to undercut arguments associated with (c) because (1) epistemic modality should not be confused with metaphysical modality nor is the former a good guide to the latter; and (2) these arguments just show that there are two ways of knowing the same (physical) things, not that there are two things known. I think these responses are too dismissive and fail to grasp the fact that arguments associated with the epistemic phenomena captured by (c) are really ontological arguments and not epistemic ones at all.27 The arguments are not simply pointing out that there are certain epistemic avenues to mental states (private access, direct access, first-person authoritative access) and there are other avenues to physical states. Rather, the arguments begin by surfacing these epistemic features of mental states and they go on to say that if strong physicalism were true, none of these features would obtain in the world. The arguments, then, move quickly to the sorts of properties, particulars and relations that characterize physical entities (e.g. brain states) and entail that if mental states were physical, they would exhibit and only exhibit these sorts of features. Because they do not, and because they would if they were physical, they are not physical after all. Successful or not, this sort of argument is not making a simple move from epistemic observations straightaway to a conclusion about dualism, and rebuttals like the ones mentioned above seem to me to be entirely irrelevant.
So much for (1). What about premise (2)? Physicalist treatments of the mental, multiple realization, and the existence/irreducibility of laws in the special sciences are irrelevant here because we are granting the existence of genuine mental events constituted by mental properties. Thus, such phy-sicalist attempts to avoid the reduction of psychological to physical laws by denying such laws in the first place do not count against (2).28 For example, both the functionalist account of the mental offered by Fodor and the anomalous monism of Davidson deny the existence of general exceptionless psychological or psycho-physical laws. But both positions depict the mental as being realized by the physical. Moreover, most are naturally associated with token physicalism when it comes to an ontological analysis of individual mental events.29 Yet, if mental and physical events are what the argument from consciousness takes them to be, then it seems reasonable for individual events of both kinds to be instances of general types of events that could in principle be correlated.
Premise (2) would be accepted by an advocate of emergent physicalism since there are two desiderata for this position: non-reductive physicalism plus the dependency of supervenient entities on the physical. If one accepts premise (1) but denies (2), then the mental becomes too autonomous for naturalism. An example of such a view is weak dualism according to which the mind is a Humean bundle of mental states that neither belong to nor depend on a specific body but which at best are more or less generally associated with specific physical states.
The main justification for premise (4) is the difference between libertarian and event causal theories of agency. J. L. Mackie rejected (4), claiming that personal explanation is simply a sub-class of event causal explanation. Moreover, divine action, as it figures into Swinburne’s account of personal explanation, involves the direct fulfillment of an intention on the part of God. But, argued Mackie, since human action is a type of efficient event causality between the relevant prior mental state (e.g. an intending, and a fulfillment which runs through and depends on a number of intermediate events which are part of a complex physical mechanism), there is a dis-analogy between human intentional acts in which intentions are fulfilled indirectly and those of a god in which, supposedly, intentions are directly fulfilled. On Mackie’s view, this disanalogy makes alleged divine action and the relevant sort of personal explanation mysterious and antecedently improbable. Thus, (4) is false and, even if it is true, it makes theistic personal explanation less, not more probable.
Is Mackie’s argument successful against (4)? I don’t think so. For one thing, pace Mackie, it is not at all clear that libertarian agency and the associated form of personal explanation are not to be preferred as accounts of human action to event causal accounts. Obviously, we cannot delve into this issue here, but if libertarian agency is correct, then Mackie is wrong in his claim that (4) is false.
Secondly, a defense of (4) may only require a concept of libertarian agency and personal explanation, even if we grant an event-causal theory of action for human acts. If we have such a clear conception, then even if human acts do not fall under it, under the right circumstances, it could be argued that a form of explanation clearly available to us is now to be employed. What those circumstances are and whether they obtain are more centrally related to premises (3) and (6) of AC and not (4). But since Mackie criticized (4) because if true it would make theistic explanation antecedently improbable, I want briefly to say something about what could justify the claim that a personal explanation of the libertarian sort should actually be used.
There have been a number of attempts to state necessary and sufficient conditions for personal action in event causal terms with John Bishop’s account being the most sophisticated to date. But Bishop admits that our common sense concept of agency is different from and irreducible to event causality and is, in fact, libertarian.30 For Bishop, the pervasiveness and power of the libertarian conception of agency places the burden of proof on the defender of a causal theory of action. Bishop claims that his own causal theory works only for worlds relevantly similar to ours in being naturalistic (strictly physical) worlds. He does not offer an analysis of action true across all possible worlds because he admits that our concept of action is libertarian and there are worlds in which it is satisfied. His justification of this minimal task is a prior assumption of naturalism, but such an assumption is clearly question-begging against AC. Therefore, if we have a clear, powerful and, prima facie justified libertarian conception of agency, Mackie’s point about the mysteriousness and antecedent improbability of anything answering to this concept is seriously overstated.
Now, if we grant the non-physicality of mental states, then a causal theory of action for human acts will boil down to the claim that person P does some act e (raising one’s hand to vote) if and only if some event b (the hand going up) which instantiates the type of state intrinsic to e-ing is caused by the appropriate mental state in the appropriate way. Note carefully that, regardless of the details of such an account, it will amount to nothing more than a causal correlation between certain physical states and the relevant mental events. According to premises (2) and (3) of AC, these correlations need and have an explanation. A causal theory of action will not do for the origin, regularity, and precise nature of these correlations, since these are what constitute a causal theory of action in the first place. If a causal theory of action presupposes mental states, then it will be important to explain the existence, regularity, and precise nature of those mental states themselves unless, of course, a divine causal theory of action is used. If this is so, and if we possess a clear concept of libertarian agency and personal explanation, then there is no good reason why a theist cannot use this type of explanation in this case.
However, when it comes to defending AC, I think one could deny a libertarian view of agency and personal explanation altogether and still defend (4). After all, some Christian theists, e.g. certain Calvinists, employ a causal theory for divine action. One could argue that there is some difference between normal physical event causality in physics and a causal theory of personal action. At the very least, the latter utilizes appropriately related mental states as parts of causal chains. Since (4) simply notes that there is a distinguishable difference between personal and natural scientific explanation, the alternative we are now considering may be all that AC needs to rebut Mackie. Bishop claims that for a naturalist causal theory of action must be combined with a strong physicalist theory of mental states.31 I agree. I also reject a causal theory of action.32 But setting this aside, since we are assuming the reality of mental states, Bishop’s physicalist rendition of the causal theory of action simply does not apply here and a suitable statement of the nature and role of mental states in a causal theory could be all that is needed to distinguish personal from natural scientific explanation according to (4).
The presence of personal explanation as a unique argument form means that when it comes to explaining emergent properties such as those constitutive of consciousness, one does not need to acquiesce with Samuel Alexander’s dictum that such properties are ‘‘to be accepted with the ‘natural piety’ of the investigator.’’33 Thus, it is more than curious to find naturalists jump straightaway from the recognition that mental properties are genuinely emergent and incapable of naturalist explanation to the conclusion that we must take then as brute facts. To cite one example, speaking of the law-like correlations between emergent mental and physical properties, Kim says that
[t]he emergent approach, therefore, asks us to accept these dangling laws as brute, unexplainable laws—that is, we are asked to count them among our fundamental laws, laws that are basic in the sense that no further explanation is possible for them. But this proposal is highly implausible, for we expect fundamental laws of nature to be reasonably simple, but these psychophysical correlations involve, on the physical side, tens of thousands of cells, millions and billions of molecules and basic particles.34
There are two sides to (5): Is personal explanation different from natural scientific explanation and are there other explanations for the facts mentioned in (1) and (2) besides these two? We have already dealt with the first question in conjunction with (4). Regarding question two, I think it is safe to say that, given the current intellectual climate, a personal theistic or a naturalistic explanation would exhaust at least the live, if not the logical, options. It is true that Thomas Nagel suggested that panpsychism might be necessary to explain the mental.35 But it is widely recognized that panpsychism has serious problems in its own right, e.g. explaining what an incipient or proto-mental entity is or how the type of unity that appears to characterize the self could emerge from a mere system of parts standing together in various causal and spatio-temporal relations.36 Moreover, panpsychism is arguably less reasonable than theism on other grounds, though I cannot pursue this point here. Further, it is not clear that panpsychism is an explanation of the phenomena in question. As Geoffrey Madell notes, ‘‘the sense that the mental and the physical are just inexplicably and gratuitously slapped together is hardly allayed by adopting ... a panpsychist ... view of the mind, for [it does not] have an explanation to offer as to why or how mental properties cohere with physical.’’37 Interestingly, Nagel’s own argument suggestive of panpsychism turns on a failure to consider a theistic explanation of the mental, coupled with an admission of the inadequacy of a natural scientific explanation:
One unsettling consequence of such a theory [of mental/physical duality] is that it appears to lead to a form of panpsychism - since the mental properties of the complex organism must result from some properties of its basic components, suitably combined: and these cannot be merely physical properties or else in combination they will yield nothing but other physical properties. If any two hundred pound chunk of the universe contains the material needed to construct a person, and if we deny both psychophysical reductionism and a radical form of emergence, then everything, reduced to its elements, must have proto-mental properties.38
Actually, Nagel’s statement is a near precis of AC. He accepts (1) and (2) in his denial of reductionism, he accepts (3) in his rejection of radical emergence which, I take it, would amount to the claim that the emergence of the mental from the physical is a brute case of something coming from nothing without explanation, and his whole argument rests on the acceptance of (6) as an implicit premise. Elsewhere, Nagel expresses a view about freedom and personal explanation according to which libertarian freedom is what we take ourselves to have, yet we cannot have it, given naturalism and the external, third-person scientific point of view.39 Apparently, Nagel would accept some version of (4). That leaves (5) and, as far as I know, Nagel does not argue for the relative merits of theism vs. panpsychism. At the very least, we may be able to say this: If the other premises of AC are accepted, then scientific naturalism is false and there is an intramural debate left between theists and panpsychists. We will explore this dispute in chapter six. In chapter seven, we will examine Philip Clayton’s pluralistic emergentist monism. There I will show that Clayton’s position is an alternative to and not a plausible version of positive naturalism.
(7) follows from previous steps in the argument and asserts the adequacy of a personal explanation for the facts expressed in (1) and (2). One may reject (7) (or (5)) because personal explanation, theistic or otherwise, does not give us any real understanding of an explanandum, especially one like (1) and (2). Sometimes this objection assumes that an explanation must cite a mechanism before it can count as adequate. My response to this problem centers on the difference between libertarian and event causality and their associated forms of explanation.
Advocates of libertarian agency employ a form of personal explanation that stands in contrast to a covering law model. To understand this form of explanation, we need to look first at the difference between a basic and nonbasic action. Often more than one thing is accomplished in a single exercise of agency. Some actions are done by doing others, e.g. I perform the act of going to the store to get bread by getting into my car and by driving to the store. Basic actions are fundamental to the performance of all others but are not done by doing something else. In general, S’s φ-ing is basic iff there is no other non-equivalent action description “S’s Ψ-ing’’ such that it is true that S φ-ed by Ψ-ing. My endeavoring to move my arm to get my keys is a basic action. A non-basic action contains basic actions as parts, which serve as means for realizing the ultimate intention of that non-basic action. To fulfill a non-basic intention, I must form an action plan certain ordered set of basic actions that I take to be an effective means of accomplishing my non-basic intention. The action plan that constitutes going to the store to get bread includes the acts of getting my keys and walking to my car.40
In my view, an action is something contained wholly within the boundaries of the agent. Thus, strictly speaking, in standard cases the results of an action are not proper parts of that action. A basic result of an action is an intended effect brought about immediately by the action. If I successfully endeavor to move my finger, the basic result is the moving of the finger. Non-basic results are more remote intended effects caused by basic results or chains of basic results plus more remote intended effects. The firing of the gun or the killing of Lincoln are respective illustrations of these types of non-basic results.
With this in mind, a personal explanation (divine or otherwise) of some basic result R brought about intentionally by person P where this bringing about of R is a basic action A will cite the intention I of P that R occur and the basic power B that P exercised to bring about R. P, I, and B provide a personal explanation of R: agent P brought about R by exercising power B in order to realize intention I as an irreducibly teleological goal. To illustrate, suppose we are trying to explain why Wesson simply moved his finger (R). We could explain this by saying that Wesson (P) performed an act of endeavoring to move his finger (A) in that he exercised his ability to move (or will to move) his finger (B) intending to move the finger (I). If Wesson’s moving his finger was an expression of an intent to move a finger to fire a gun to kill Smith, then we can explain the non-basic results (the firing of the gun and the killing of Smith) by saying that Wesson (P) performed an act of killing Smith (I3) by endeavoring to move his finger (A) intentionally (I1) by exercising his power to do so (B), intending thereby to fire the gun (I2) in order to kill Smith. An explanation of the results of a non-basic action (like going to the store to get bread) will include a description of the action plan.41
By way of application, the adequacy of a personal explanation does not consist in offering a mechanism, but rather, in correctly citing the relevant person, his intentions, the basic power exercised, and in some cases, offering a description of the relevant action plan. Thus, if we have some model of God and His intentions for creating a world suitable for human persons (from revelation or otherwise), we can make reference to God, His intentions for creating a world with persons with mental states regularly correlated with their environment, and the adequacy of His power to bring about the basic results captured in (1) and (2).
Premise (8) seems fairly uncontroversial. To be sure, Humean-style arguments about the type, size, and number of deities involved could be raised at this point, but again, these issues would be intramural theistic problems of small comfort to someone like Searle committed to nat-uralism.42 Moreover, if we take live options only, then it seems fair to limit our alternatives in (5) to theistic or naturalistic. If that is acceptable, at least for the purposes of arguing against Searle and other naturalists like him, then (8) should not be objectionable.
In the terms of epistemic appraisal proffered by Chisholm, it seems that, given AC and what we have seen about the naturalist ontology from chapter one, :(N & Emergence2a) is at least beyond reasonable doubt where a proposition is beyond reasonable doubt for a subject S means that S is more justified in believing that proposition than in withholding it.
Alternatively, given AC, (N & Emergence2a) is at least reasonable to disbelieve (S is more justified in disbelieving that proposition than in withholding it).43 However, it would be premature to conclude that this is the correct epistemic appraisal of (N & Emergence2a). We still need to look at premises (3) and (6). Rather than doing so directly, I shall examine them in chapters three through seven in the context of naturalist attempts that, if successful, would defeat (3) and (6).