Responding to James White’s Anti-Molinist Critiques: Abstract Objects

By Jonathan Thompson

|

July 3, 2018

Summary: In The Dividing Line episode Behold the Secular Woman & WLC on Molinism (Once Again) James White argues that if Molinists are anti-realists about abstract objects, then they cannot consistently affirm that middle knowledge demarcates the range of feasible worlds available for God to create. This is because, according to White, counterfactuals of creaturely freedom (hereafter, CCFs) must first exist in order for them to be able to demarcate the range of those feasible worlds available to God. Here, I argue that White’s response begs the question against the anti-realist Molinist. I then entertain and refute a stronger version of White’s objection which turns out to be a variant of The Indispensability Argument for Platonism.

Articulating & Responding to White’s Objection
White begins his critique of Molinism by responding to a segment of the Reasonable Faith podcast in which William Lane Craig states the following:

On Molinism, God does not determine the truth value of. . . counterfactuals of creaturely freedom. . . [F]or some persons that gives the appearance of making God dependent upon creatures in some way . . . [b]ut none of these [CCFs] actually exist on an anti-realist view. These are just ways of talking; it is a façon de parler – a way of speaking or a heuristic device for talking about modal notions.1

White responds to Craig’s statement with the following:

Did you catch that? None of these things actually exist! They do not have any real existence so as to not threaten the aseity of God! . . . Then how can they circumscribe and determine the range of God’s possible actions if they don’t exist?2. . . If they limit what God can do as to what world He can actualize then they have to have some existence, not a physical existence, but a sufficient existence to delimit God’s activities and actions.3

Notice that White presents no justification for the idea that CCFs must exist in order for it to be true that they demarcate the range of feasible worlds available for God to create. White instead merely assumes that Molinists ought to accept this assumption. He therefore begs the question against the anti-realist Molinist. Now, at this point, we’ve sufficiently responded to White’s objection and so may call it a day. However, for the sake of a more substantive discussion, in the proceeding section I shall attempt to formulate a criterion for White which, in the context of an argument I shall give on behalf of him, will attempt to conclude that Molinists who are anti-realists about CCFs cannot consistently affirm that CCFs demarcate the range of worlds God is able to create.

Formulating a Criterion of Ontological Commitment for White
For the sake of argument let’s stipulate that White’s justification for his unargued assertion is that he thinks that the only way we can utter a true sentence of a certain kind is by also being committed to the existence of the object(s) to which we refer to in or by that sentence. This is what is known in academia as a criterion of ontological commitment. To be clear, such a criterion does not tell us what exists. Rather it tells us what must exist if a sentence of a certain kind is to be true. Moreover, by ‘exist’ it’s important to clarify that I am using this term in a heavyweight platonic sense. In order to understand what is meant by ‘heavyweight platonic sense’, it may be helpful to distinguish lightweight from heavyweight platonism. Craig explains:

Lightweight Platonism treats abstract objects merely as the semantic referents of certain singular terms like proper names and definite descriptions. On lightweight Platonism abstract objects are individuals merely in the sense that Wednesdays and the hole in your shirt are individuals, namely, as referents of the terms “Wednesday” and “the hole in your shirt,” but not in a sense which would require God to create such things in order for us to speak meaningfully of their existence.4

In contrast, according to heavyweight platonism “. . . abstract objects exist just as robustly as the fundamental particles which make up the physical world.”5 So, to be clear, when the adherent of the relevant criterion of ontological commitment attempts to commit us to the reality of abstract objects, the sense in which they attempt to commit ourselves to them is in this heavyweight sense. Of course, it’s worth mentioning that abstract objects qua abstract objects still essentially lack causal powers, even if they exist in a heavyweight sense, but nonetheless, given the aforementioned criterion of ontological commitment, they still must exist. What we seek to answer, then, is if we have good reason to accept White’s criterion of ontological commitment and whether or not such a criterion is plausible.6

That said, so far I’ve stipulated that White is committed to the idea that the only way we can utter a true sentence of a certain kind is by also being committed to the existence of the object(s) to which we refer to in or by that sentence.7 But this still leaves unanswered the question as to the exact type of sentence White thinks entails the relevant criterion. It also fails to specify the sentential devices we deploy in order to engage in the act of reference. For the sake of argument, let’s stipulate the following criterion:

(1) If a simple sentence (i.e., a sentence of the form ‘a is F’, or ‘a is R-related to b’, or . . .) is literally true, then the objects which its singular terms denote exist.  Likewise, if an existential sentence is literally true, then there exist objects of the relevant kinds; e.g., if ‘There is an F’ is true, then there exist some Fs.8

Now let us define our terms:

  • Simple sentence: a sentence with a single clause, with a single subject, and predicate.
  • Singular term: a word or phrase that serves to denote some particular. Singular terms consist of proper names (e.g., ‘Jesus’, ‘The Eiffel Tower’,  ‘Star Wars: Attack of The Clones’, etc.), definite descriptions (e.g., ‘the man with the golden voice’, ‘my best friend’, ‘the first female President’, etc.), and demonstrative terms (e.g., ‘this essay’, ‘that desk’, etc.).  With respect to the sentence “CCFs are grounded in counterfactual states of affairs.” the singular terms that appear in that sentence (i.e., ‘CCFs’ and ‘counterfactual states of affairs’) both qualify as definite descriptions.
  • a: a symbol which we may substitute in for a singular term that appears in the domain of quantification in order to form a simple sentence. So, for example, we can substitute the relevant symbol in the sentence ‘a’s are f’  so that it now states ‘CCFs are f’.
  • f: a symbol for which we may substitute any property which we might predicate of the individual picked out by the singular term. So, for example, the simple sentence “CCFs are f” could be restated as “CCFs are grounded in counterfactual states of affairs”.
  • R: any relation in which the object specified by the singular term may be said to stand.

Those who are already familiar with the contemporary philosophical scene concerning abstract objects will recognize (1) to be the so-called neo-Quinean criterion of ontological commitment. I’ve chosen to address this particular criterion because I suspect that White, if pressed, would espouse something roughly similar. Moreover, so far as I can tell, this particular criterion appears to be the most popular among philosophers and so an evaluation of this criterion will be of interest to many.

Now let us now add further clarity to (1). According to this criterion we make ontological commitments by utilizing singular terms and/or existential quantification in literally true simple sentences. The reason why this criterion applies itself only to literally true sentences is to preserve the meaning of non-literally true sentences which we employ in our ordinary, non-formal discourse. For example, when James White asserts a statement like “Kira is dead in sins and trespasses”, White surely does not mean to communicate that Kira is literally dead—soul separated from body in the afterlife—and is such in these curious objects known as “sins” and “trespasses”. Rather, being faithful to his Calvinism, White simply means to convey that Kira is affected by his own sin nature such that he is totally depraved (in the Calvinist sense), has total inability (in the Calvinist sense), or something roughly similar. So although it is true that Kira is dead in trespasses and sins, nevertheless, faithfulness to the intended meaning of White’s statement about Kira requires that we construe it as being metaphorically true rather than as literally true. Thus, as this example illustrates, if one wants to preserve the intended meaning of certain sentences, then one’s criterion of ontological commitment will have to exclude non-literally true sentences as (1) attempts to do.

Next, notice that the relevant criterion only applies itself to simple sentences. This is in order that it applies only to extensional contexts as opposed to non-extensional (aka intensional) one’s. To elaborate, extensional contexts are sentence phrases that (i.) possess singular terms that can be substituted in for other co-referring singular terms without affecting the sentence’s truth value and (ii.) allow you to quantify into their contexts from outside the contexts.9 To give an example of (i.), the statement “The Batman defeated The Joker” is an extensional context because one can substitute in the co-referring term “Bruce Wayne” for “The Batman” without affecting the sentence’s truth value. In contrast, the statement “Commissioner Gordon believes that The Batman is investigating the crime scene” is an intensional one since it cannot maintain its truth value if we substitute “Bruce Wayne” in for “The Batman”. This is because Commissioner Gordon, we shall stipulate, is not aware that Bruce Wayne is The Batman. So, with respect to this example, the point is that even though “The Batman” and “Bruce Wayne” are co-referring terms, nevertheless, one’s thought that Batman will investigate the crime scene may not be identical to the thought that Bruce Wayne is doing such. Thus, if the anti-realist wants to preserve the content of what we mean to convey by our thoughts (such as the one stated above about The Batman), then their criterion of ontological commitment will have to be narrowed by avoiding intensional contexts as does (1). Now, to give an example of (ii.), one can validly infer that ‘there is someone who is the child of Bruce Wayne’ from the statement ‘Damien Wayne is the child of Bruce Wayne’. Notwithstanding, one cannot validly infer that ‘Damien Wayne is the child of Bruce Wayne’ from the statement ‘I hope Damien Wayne is the child of Bruce Wayne’. In this case one’s intentional attitude (i.e., hoping) supplies the intentional context. Aside from intentional attitudes, however, it’s also worth mentioning that Craig also notes that modal operators (like ‘necessarily, . . . ’ or ‘possibly, . . . ’), and temporal operators (like ‘it was the case that . . . ’ or ‘it will be the case that . . . ’) supply intensional contexts.10

In sum, (I) expresses a criterion of ontological commitment which attempts to not overreach in its scope in so far as it does not impose itself upon non-literally true sentences as well as intensional contexts. (I), therefore, will be inapplicable to  perhaps the majority of human discourse.

Reformulating White’s Objection
So far we’ve formulated a criterion of ontological commitment according to which we make ontological commitments by means of employing singular terms or existential quantifiers in literally true simple sentences. Now how might this premise appear in an objection that is faithful to the essence of White’s concern? Recall that White attempts to argue against Craig’s claim via reductio ad absurdum. In specific, White attempts to demonstrate that an anti-realist Molinist cannot consistently endorse the Molinist premise that CCFs demarcate the worlds God is able to create because CCFs do not exist. In accordance with logical validity, as well as White’s attempt to run a reductio argument against the Molinist, we may formulate White’s argument as follows:

(1) If a simple sentence (i.e., a sentence of the form ‘a is F’, or ‘a is R-related to b’, or . . .) is literally true, then the objects which its singular terms denote exist.  Likewise, if an existential sentence is literally true, then there exist objects of the relevant kinds; e.g., if ‘There is an F’ is true, then there exist some Fs.

(2) The sentence “CCFs demarcate the range of feasible worlds God is able to create” is a literally true simple sentence containing singular terms (i.e., ‘CCFs’ and ‘feasible worlds’) that refer to things that could only be abstract objects. Likewise, the existential statement “There are CCFs which demarcate the range of feasible worlds God is able to create” contains an existential quantifier (i.e., ‘there are’) which ranges over things that could only be abstract objects (i.e., ‘CCFs’ and ‘feasible worlds’). [anti-realist Molinist assumption]

(3) Therefore, abstract objects exist.

The above argument shouldn’t be alien to those already familiar with contemporary philosophical debates concerning abstract objects. For the argument just is a modified version of the so-called Indispensability Argument for Platonism. The only difference here is that I’ve singled out particular sentences in premise (2) rather than generalizing the premise.

Now, prior to evaluating the relevant argument we should first understand its implications so that we can grasp what exactly lies at stake for the Molinist. Most obviously, notice that even if the relevant objection is successful, it does not conclude that Molinism is false. Rather, the conclusion entails that anti-realism is false, thereby implying that Molinists who are anti-realists about abstract objects ought to be committed to the reality of CCFs and feasible worlds. Therefore, the argument’s conclusion is irrelevant to Molinists who are already realists concerning abstract objects. With this in mind, in the proceeding section, we move to evaluating premise (1).

Responding to Premise (1) – Relational vs Non-Relational (aka Deflationary) Theories of Reference
Discussing premise (1), how might the anti-realist object to it? First it should be noted that no argument has been given in favor of this premise. Its truth has instead been assumed from the outset. For White to assume (1)’s truth against the Molinist without any sort argument will therefore beg the question against the anti-realist Molinist.11 Second, notice that the premise stipulates that the only way certain types of simple sentences can be true is if we are also committed (in a heavyweight sense) to the existence of the entity of which its singular term(s) refer. But why suppose that singular terms refer to anything? As Craig has argued, while there is this widespread (and sometimes question-begging)12 assumption that reference should be understood as a relation between words and objects in the world, such an assumption appears to be incoherent. Craig explains why:

It is an experiential datum that referring is a speech act carried out by an intentional agent. Words in and of themselves engage in no such activity. Lifeless and inert, words are just ink marks on paper or sounds heard by a percipient. Absent an agent, shapes or noises do not refer to anything at all. If, for example, an earthquake were to send several pebbles rolling down a hillside which randomly came to rest in the configuration JOHN LOVES SUSIE, the names—if we would even call them names—would not refer to anybody. . . An interpreting agent uses his words as a means of referring to something. Referring is thus an intentional activity of persons, and words are mere instruments. 13

Even further, as we previously noted (1) assumes that reference is to be understood as a relation. But notice that no argument has been given for why we ought to accept this presupposition. What needs to be appreciated here is that there are multiple theories of reference and of these theories are ones that construe reference not as a relation between word and object, or even as a relation between speaker and object, but instead as a property of persons that is relation-like. So, for example, on such a view my engaging in the activity of referring to X (where X stands for any given singular termjust is identical to my having the mental-relational property of thinking-about-X, being-directed-toward-X, or something roughly similar, whether or not X exists.

What is meant by the term “relation-like”? An illustration may help elucidate the concept: Suppose that you were to induce a hallucination in me such that it appears to me that there is a red balloon in front of me. I then form a thought about the balloon so that I bare the property of thinking-about-the-red-balloon-in-front-of-me. In such a scenario, this mental-relational property of my thinking-about-the-red-balloon-in-front-of-me obtains, notwithstanding, the balloon itself doesn’t actually exist. Hence, in this case I possess a mental-relational property about a particular that does not exist in a heavyweight sense. Similarly, for those who take reference to be an intentional property of persons, we can bare mental-relational properties about abstracta such as being-directed-to-X or referring-to-X, even though the abstract object of our intention does not exist in a heavyweight sense.

Now, at this point, one might argue that since anti-realists believe that abstract objects don’t exist, then it follows that when anti-realists claim to think about abstract objects they are really thinking about nothing, and hence, not thinking at all, since there isn’t any existent object to which the intention of our thought corresponds. But, the objector further argues, we obviously are engaged in the act of thought when thinking about abstract objects. Therefore, abstract objects must exists in a heavyweight sense.

This objection fails to appreciate the point we just made about competing views of reference. For the objection stipulates that in order for us to think about something (and hence, refer to it) there must be this relation in so far as there is some existent which corresponds to the intention of our thought. Hence, the objector is assuming without argument that reference should be construed as a relation rather than as a property. The objector has therefore begged the question against non-relational views of reference.

Responding to Premise (1) – Are Singular Terms Devices of Ontological Commitment?
According to Craig, there are also problems with taking singular terms as being devices of ontological commitment.14 In arguing this point, he highlights the fact that standard logic (aka classical logic) takes singular terms as being devices of ontological commitment. Accordingly, an identity statement such as T = T ontologically commits one to the existence of whatever object is denoted by T. But, Craig opines, it seems bizarre to think that one can infer the existence of some object based on something as trivial as an identity statement. This is because, as Karel Lambert notes, it is at odds with the “intuition that logic is a tool of the philosopher and should ideally be neutral with respect to philosophical truth. . . So if there are preconditions to logic that have the effect of settling what exists and what does not exist, they ought to be eliminated because they corrupt the ideal of logic as a philosophical tool.”15

Craig then highlights that since standard logicians take singular terms as being devices of ontological commitment, their way of avoiding making untoward ontological commitments is to treat identity statements involving vacuous singular terms (i.e., singular terms with no corresponding existent(s) denoted by the term(s)) as false. But, this too, Craig argues, is intuitively wrongheaded. For this would imply that a statement such as Vulcan = Vulcan isn’t true, even though the statement is an obviously true tautology irrespective of if Vulcan exists. Craig thus concludes that standard logic can’t distinguish whether statements like Zeus = Zeus or Allah = Zeus are true or not even though the first identity statement seems to be necessarily true while the other necessarily false.

Furthermore, Craig thinks that, given the A-theory of time, standard logic will be unable to affirm the truth of intuitively obvious tautologies like Colossus of Rhodes = Colossus of Rhodes or The Library of Alexandria = The Library of Alexandria. This is because the objects denoted by those statements no longer exist implying that there are no existents which correspond to the objects denoted.16 Having pointed this out, Craig notes that one might attempt to argue that standard logic can affirm the truth of the aforementioned identity statements by appealing to a B-theory of time to ground them so that all objects throughout time—including the Colossus of Rhodes and The Library of Alexandria—exist on ontological par with the moment in which we exist.17 Craig, however, argues that such an inference is unwarranted since it would only further run contrary to the intuition articulated previously by Lambert. This is so because inferring the B-theory of time from an identity statement has the “effect of settling what exists and what does not exist” by asserting that non-present entities exist.18

Now, because standard logic does not permit us to infer the truth of vacuous statement like Zeus = Zeus, standard logic obviously becomes limited in the inferences it permits us to make. Not only that, neither will standard logic have the resources for differentiating between (a) inferences where whether or not the singular term(s) refers to something that exists is essential to the validity of an argument, and (b) inferences where it is irrelevant whether or not the singular term(s) refers to something that exists. To demonstrate this point, Craig invites us to consider the following argument:

(1) Lincoln was the Great Emancipator
(2) Lincoln brooded
(3) Therefore, the Great Emancipator brooded

The above argument appears to be formally valid, exhibiting a modus ponens structure. Moreover, in order for the conclusion to be validly derived, one of the premises will have to be identical or else logically equivalent to the premise ‘Lincoln brooded’. Prima facie, this appears to be the case with respect to the above argument. However, if for the sake of argument, we take it to be the case that Lincoln no longer exist implying that ‘Lincoln’ is a vacuous term, then the argument will turn out to be logically invalid. This is because standard logic will prohibit us from inferring whether or not Lincoln = Lincoln.  Moreover, neither can the argument’s formal validity be redeemed by substituting the term ‘Lincoln’ in premise (2) for a co-referring term such as ‘The 16th President of the United States’. For if ‘The 16th President of the United States’ is vacuous term, then standard logic will prohibit us from inferring whether The 16th President of the United States = The 16th President of the United States or if The 16th President of the United States = something non-identical or something that doesn’t co-refer to Lincoln.

Because Craig finds standard logic’s treatment of singular terms untenable, he finds himself sympathizing with and defending neutralism, a view which maintains that neither singular terms or existential quantifers in first-order logic are devices of ontological commitment. Such a view, if correct, will obviously force revisions in classical logic since classical logic affirms that singular terms are devices of ontological commitment. Notwithstanding, Craig does think that a thoroughgoing logic that is neutral with respect of ontological commitment does free us from these unwanted ontological commitments and does make for a less bloated ontology.

Responding to Premise (1) – Is Existential Quantification a Device of Ontological Commitment?
Proceeding ahead, the second horn of the neo-Quinean criterion asserts that we make ontological commitments by means of employing existential quantification. In other words, according to this criterion, we make these commitments by employing existentially quantifying phrases or symbols such as “For all”, “For some”, “There exists”, “∃”, etc. Craig has given a number of reasons why he thinks existential quantification shouldn’t be taken as being ontological committing.19 Let us discuss three of these reasons.

First, Craig argues that taking existential quantification as being a device of ontological commitment is ontologically inflationary. That is to say that it commits us to the existence of entities needlessly. To demonstrate the point, consider the notion of ‘months’. Presumably most of us will want to affirm that months are nothing more than a created social convention. Not so, given the neo-Quinean criterion of ontological commitment. For the sentence “There are 12 months in a year” would commit us to the existence of these abstract objects called ‘months’ simply in virtue of our existentially quantifying over them. Similarly, consider the statement “There is a lack of love in the world”. According to the neo-Quinean criterion, since I have asserted that there is a ‘lack’ (thereby quantifying over the term ‘lack), I am thereby ontologically committed to there being ‘lacks’, which, on Craig’s view, is surely wrongheaded.

Second, Craig thinks that taking existential quantification as being ontologically committing prohibits us from accepting the truth of tensed statements given the truth of a certain version of presentism. For example consider the past-tense statement “There have been 6 ice cubes which have melted in this cup”. We commonly accept the truth of such innocuous sounding statements. Furthermore, let us suppose that some version of presentism which asserts that there exist no non-present entities is also true; I shall take it for granted that most of my readers, whether they are Molinists or otherwise, will be sympathetic to such a view. Now, if such a version of presentism is true, and if we take it to be the case that existential quantification is a device of ontological commitment, then that statement involving the 6 ice cubes will turn out to be false. This is because there will be no existent ice cubes which correspond to the objects denoted in that statement. Of course, one could attempt to ground the statement by appealing to the B-theory of time so that the ice cubes bare a tenseless existence, but I take it to be the case that there are good independent reasons for thinking that the B-theory of time is false.20 Therefore, given the truth of these reasons, ontological frameworks which require B-theory should be avoided unless we have some good independent reason(s) for preferring it.

Third, Craig argues that taking existential quantification as being ontologically committing will negatively impact ones mereology if one is committed to mereological universalism.21 To elaborate, one of the issues which mereologists are concerned are the conditions which have to be met in order for a plurality to compose an object. The predominant view on this matter, according to Craig, is what is known as mereological universalism22 which assumes the following principle of unrestricted mereological composition (UMC):

UMC: Necessarily, whenever there are some things, then there are a fusion of those things.

That in mind, it is obviously true that there exists sharks and there exists tornadoes. Thus, given the principle of UMC, there exists a fusion of those objects which we shall call a Sharknado, where by ‘Sharknado’ I am not referring to a tornado with sharks in it, but am instead referring to an object that is composed of, say, a localized tornado in Kansas and a shark in the Pacific Ocean. Moreover, since, I have existentially quantified over that fusion, then, given the truth of the neo-Quinean criterion of ontological commitment, I am thereby ontologically committed to there being Sharknadoes. Craig takes this untoward conclusion to be counter-intuitive and ontologically inflationary, especially in light of the fact that are a potentially infinite number of fusions. Given the truth of mereological universalism and given the falsity of the Neo-Quinean Criterion of ontological commitment, Craig thinks that the obvious solution to this problem will be to adopt neutralism which permits us to freely speak about fusions without being ontologically committed to their existence.

Finally, at this point, we shall briefly discuss a common reason philosophers take existential quantification as being a device of ontological commitment. Quite simply, many philosophers take this view on ontological commitment because they think that the meaning of these quantificational phrases analytically requires it. So, for example, the defender of this view would affirm that the sentence “There exists true CCFs” ontologically commits the Molinist to the existence of CCFs in virtue of what the word “exists” means. Unfortunately, such an interpretive modus operandi betrays hermeneutical nescience. For every good exegete knows that one cannot sufficiently interpret a speakers statement by only considering the meaning of the words they employ. For one must also consider the speaker’s intentions. This point is illustrated well when one considers Jesus’ parable of the mustard seed. Virtually no good theologian thinks that when Jesus says that the mustard seed is the smallest of all seeds (Mark 4:31), that he is trying to convey some fact about botany simply because of the meaning of the words he uses. Rather, they recognize that Jesus is attempting to teach a lesson about the Kingdom of God, and the point about the size of mustard seed is simply incidental to his lesson. Similarly, although it is true that we sometimes make ontological commitments by using existentially quantifying phrases, these phrases, in and of themselves, carry no such commitments in virtue of their meaning. Rather, we must consider what a speakers intentions are in utilizing these phrases. The invaluable insight Craig provides on this matter is that our making ontological commitments is a varying and person relative matter; we use quantifying phrases in different contexts with different commitments in mind. Furthermore, there are contextual hints and rhetorical devices we can use such as tone of voice, emphasis on words, addition of words like ‘really’, or contrastive uses of the word ‘exist’ with the phrase ‘there is/are’ in order to make our intentions clear. All this is to say that our making ontological commitments is a varying, person relative matter as opposed to something necessitated by a monolithic criterion, applicable to all persons. What, then, on Craig’s view, is the role of first-order existential quantifiers? Simply put, it is to serve the function of facilitating logical inferences.23

Responding to Premise (2) – Craig’s Combinatorial Anti-Realism
It is at this point we move to evaluating the second premise of the reformulation of White’s argument. Recall that premise (2) asserts:

(2) The sentence “CCFs demarcate the range of feasible worlds God is able to create” is a literally true simple sentence containing singular terms (i.e., ‘CCFs’ and ‘feasible worlds’) that refer to things that could only be abstract objects. Likewise, the existential statement “There are CCFs which demarcate the range of feasible worlds God is able to create” contains an existential quantifier (i.e., ‘there are’) which ranges over things that could only be abstract objects (i.e., ‘CCFs’ and ‘feasible worlds’). [anti-realist Molinist assumption]

While there are a number of ways in which the anti-realist can respond to this premise (e.g. Fictionalism, Figuralism, Pretence Theory, etc.), we shall restrict ourselves to Craig’s view since it is his view of which White is concerned. Craig’s view is unique in that he takes a combinatorial view which adopts certain elements from both Figuralism and Pretence theory without accepting every aspect of these respective theories. Prior explaining how Craig does this, we should first have an understanding of what each of these respective theories involve.

According to Figuralism, abstract object talk is merely figurative language. Therefore, premise (2) of the above variant of the Indispensability Argument, may be taken by the Figuralist-Molinist to be a misrepresentation of their view. This is because the Figuralist affirms that the abstract object talk involved in (2) ought to be construed as figurative rather than literal. That being said, Craig differs from accepting Figuralism, in its fullness, in so far as he thinks that Figuralism tacitly accepts the Neo-Quinean criterion of ontological commitment.24

According to Pretence theory, abstract object talk is merely a fictional discourse that is prescribed to be imagined (or make-believed) as true, similar to how we imagine Santa Claus to live in the North Pole. To be clear, Pretence theory is not identical to the anti-realist view known as Fictionalism. For Fictionalism claims that sentences involving quantification over or reference to abstract objects are untrue while Pretence theory is simply neutral on the matter. Neither should Pretence theory be confused with Figuralism. For Figuralism affirms that, at least, sometimes abstract discourse is figuratively true while Pretence theory, again, is simply neutral on the matter. Simply put, Pretence theory is neutral with respect to whether or not our abstract discourse is true. To avoid further confusion, it’s also worth noting that certain Pretence theorists such as Kendall Walton do not take truth and fictionality to be mutually exclusive.25 For consider the propositions prescribed to be imagined in the historical fiction War and Peace. While the story is itself fictional, nevertheless, some of the propositions in this book are, in fact true. Thus, for Pretence theorists like Walton (and Craig), what is crucial to understanding the notion of fictionality, is not falsehood, but the prescription to be imagined.

So, with respect to Craig’s combinatorial approach, he adopts and defends a thesis according to which abstract discourse sometimes involves our using figurative language, while, at other times, it involves our engaging in pretence.26 That said, since premise (2) supposes that an anti-realist like Craig believes that abstract discourse is to be construed as literally true, the premise therefore fails to accurately characterize Craig’s position. As a result, the Indispensability Argument we’ve formulated on behalf of White not only bares two challengeable premises, but also fails to constitute a successful reductio argument against the anti-realist Molinist position.

Conclusion
In conclusion, White’s initial objection to the anti-realist Molinist position was weak in so far as he presented no justification for why the Molinist ought to accept the assumption that CCFs must exist in order to delimit the range of feasible worlds God is able to create. In light of this shortcoming, I strengthened White’s argument so that it paralleled the Indispensability Argument for Platonism. We saw that both premises of this reformulated argument were challengeable in that the first premise stipulates an unargued and implausible criterion of ontological commitment, while the second premise fails to accurately characterize certain anti-realist perspectives concerning abstract discourse. Of course, it goes without saying that a variety of other anti-realist theories are available to the Molinist. So is realism. Thus, so far as I can tell, anti-realism presents no threat to the Molinist.


Notes

Special thanks to Tela Antkowiak for creating the thumbnail image for this article. Visit her website at https://telaa.org/

1https://www.reasonablefaith.org/media/reasonable-faith-podcast/molinism-and-infallibility/

234:34-35:03

335:31-35-52

4https://www.reasonablefaith.org/writings/scholarly-writings/divine-aseity/god-and-the-platonic-host/

5Ibid.

6To be clear, I agree that there are at least some contexts where, in order for a sentence of a certain kind to be true, there must exist, in a heavyweight sense, whichever object(s) I refer to in or by that sentence. My gripe instead lies with the notion that this principle universally applies to all contexts.

7Here, it’s important to understand that, throughout this essay, I am not using the word ‘refer’ in a colloquial sense so as to denote the subject or topic of discourse. Rather, I am using the term in a philosophical context. As we shall later see, philosophers understand relevant term in different ways and there are competing views concerning the nature of reference.

8See Mark Balaguer, Platonism in Metaphysics for the original formulation of the Indispensability Argument for Platonism from which I’ve derived my variant of his argument.

9Definition is adapted from William Lane Craig, God Over All: Divine Aseity and the Challenge of Platonism p. 47.

10For an explanation as to how modal and temporal operators supply intentional contexts see Ibid., p. 48 n.6.

11For further discussion on competing theories of reference see Ibid., pp. 132-139.

12On Ibid., pp. 133-134 Craig supports the notion that philosophers sometimes beg the question in favor of relational views of reference by appealing to a standard encyclopedia entry by Timothy Williamson where Williamson defines reference as being relational.

13Ibid., pp. 134-135

14For a more thorough critique of the view that singular terms are devices of ontological commitment see Ibid., pp. 140-143.

15Karel Lambert, Meinong and the Principle of Independence (Cambridge: CUP, 1983), 98–9.

16I’ve modified Craig’s example. The example Craig originally gives involves the singular terms ‘Lincoln’ and ‘Aristotle’. I’ve withheld using these singular terms because of the following issue which might be raised: Suppose that by ‘Lincoln’ and ‘Aristotle’ Craig means to refer to them as essences. If such is the case, then Craig’s claim that Lincoln and Aristotle no longer exist will be false because the essences of Lincoln and Aristotle never existed (in a metaphysically heavyweight sense) to begin with since, on Craig’s view, essences don’t exist in a heavyweight sense. But let us now suppose that by ‘Lincoln’ and ‘Aristotle’ Craig means to refer to them as concrete particulars. If such is the case, then it will still be false that Lincoln and Aristotle no longer exist since, given Christian theology, they exist as spirits in the afterlife.

Additionally, on June 10, 2018 I had the opportunity to speak personally with Craig about this following his Defenders Class. According to Craig, this criticism is technically correct, notwithstanding, he was speaking in a non-theological context.

17In order to grasp the B-theory of time (also known as the tenseless or static view of time), it may be helpful to first contrast it with the A-theory of time (also known as the tensed or dynamic view of time). As a cursory definition, the A-theory of time asserts only time that exist is the present. In contrast, the B-theory of time asserts that past, present, and future exist on ontological par; that is to say that they are all equally real—they occupy a spatial location, possess causally potent entities, inter alia. On the B-theory of time, one may think of time on the analogy of a yardstick. On a yardstick, the 10th inch exists prior to the 11th inch. Moreover, the 12th inch exists posterior to the 11th inch. Notwithstanding, the 10th and 12th inch aren’t any less real than the 11th inch. Similarly, according to the B-theory of time, although the past and future are respectively prior and to the moment which appears to be present to us, nevertheless, the past and future aren’t any less ontic. So, on the B-theory, for the people who live in 1947, that year will appear to be present to them. Similarly, for the people who live in the year 2300, that year will appear to be present to them. On the B-theory of time, temporal becoming—that is the notion that things really do come into and then go out of existence—is an illusion of human consciousness. For on the B-theory of time all moments in time exist tenselessly.

18I’m assuming the truth of the A-theory of time because I take it to be the case that there are good independent arguments on behalf of this theory. It should be clear that I am not committing the same error that the B-theorist might commit by inferring the truth of some theory of time solely on the basis of some identity statement being true.

19For a fuller analysis of the issues with taking existential quantification as being a device of ontological commitment see William Lane Craig, God Over All: Divine Aseity and the Challenge of Platonism p. 113-124.

20The brevity I seek to preserve throughout this essay will not permit me to defend an argument against the B-theory of time. For this reason I will merely delineate the issues that I think this theory faces without doing anything by way of elaboration. That said, perhaps the most glaring issue which the B-theory faces its approach to the problem of intrinsic change. The problem of intrinsic change asks how something can remain identical to itself if it has different properties at different times. The A-theorist’s solution to this question is to deny that any object is presently such that it possesses different properties at different times. Since, on the A-theory, only the present exist, an object only possesses those properties it presently has. On the A-theory, objects exist wholly at a time and endure through time to later times. This doctrine is known as endurantism.

In contrast, the typical solution offered by B-theorists is what is known as perdurantism. According to perdurantism, no object exists wholly at a time. Instead, the three-dimensional objects that appear to us are really these four-dimensional objects extended in space-time so that the three-dimensional object we see just is a segment of this greater four-dimensional object. It may be helpful to think of this view on the analogy of a flip book animation involving Superman. When one quickly flips through the pages of the flip book there is this illusion of a single, continuous Superman, but, in reality, there are actually these individual drawings—or we can call them segments—of Superman which, when flipped through quickly, create the illusion of a single Superman which endures throughout the animation.

There are at least four issues with the doctrine of perdurantism: (1) Perdurantism’s account of intrinsic change is implausible (2) Perdurantism is incompatible with the phenomenology of personal consciousness (3) Perdurantism is incompatible with moral responsibility, praise, and blame and (4) Perdurantism implies an implausible view of essential properties. Last, it’s worth mentioning that the B-theory of time is itself incompatible with a robust doctrine of creatio ex nihilo. For an in depth discussion of the reasons that I reject perdurantism and the B-theory of time see William Lane Craig, Time and Eternity: Exploring God’s Relationship to Time pp. 200-215.

21Mereology is the study of parts and wholes.

22See William Lane Craig, God Over All: Divine Aseity and the Challenge of Platonism p .122.

23See Ibid. p. 207.

24See Ibid. p. 180.

25See Kendall L. Walton, Mimesis as Make-Believe: On the Foundations of the Representational Arts (Cambridge, MA: Harvard University Press, 1990).

26Craig’s view involves more nuance than what I’ve stated. For example, Craig acknowledges that our recursion to abstract discourse is often due to its convenience. He even acknowledges that it’s epistemically possible that in some cases our recursion to abstract discourse is unavoidable due to the exigencies of our language so that there may be no other way for us to express certain claims. Moreover, Craig views our recursion to abstract discourse as proceeded by our adopting a certain linguistic framework according to which we can speak without making metaphysically heavyweight ontological commitments.

Share:

About the Author

By Jonathan Thompson