Language,  Mind,  Wittgenstein

Could God Talk to Himself?

Does the private language argument succeed if we apply it to God? Could God have a private language?

A private mental entity (PME) is an entity only accessible by the mind that has it. A sensation, like toothache, is a good example (L. Wittgenstein, Philosophical Investigations). No one apart from the person with toothache has access to that toothache. But a thought might also count as a PME. A human thought is usually about something public, a tree perhaps, but it might be about something private like a sensation. Thoughts like these are privately owned. My thought that the Christmas tree looks good is my thought; I own it, it is had by me, and only I know I am having it (of course, God knows I have this thought and now I have hit “publish” you know that I have had this thought). However, the only way for you to know that I have had the thought is through my expression of it through a sentence in natural language.

Consider now a singular mind. Call it the Divine Mind. Imagine that the Divine Mind is the only thing in the universe. There is no matter; just one mind. Since the Divine Mind has unlimited power, the Divine Mind can think any thought it wants. It seems plain that all thoughts had by the Divine Mind would be private mental entities. There are, in our one-mind universe, no other minds that could have those thoughts or know what they are. There is also nothing the Divine Mind could think about outside the Divine Mind.

What if the Divine Mind creates another mind and intends to inform the other mind of some of his thoughts? If there was another mind and that mind was blank (not preloaded with thoughts) the other mind would not be able to know what the thoughts of the Divine Mind were. If a blank mind was confronted with the divine mind how would there be communication? All the content would be in the Divine Mind. If the Divine Mind spoke of a tree there would be no common, public, sense of tree. There is no tree and no way to make a public sense of tree. The blank mind would have nothing to go on. There could be no communication.

This is some of the problem I have been pondering lately and it depends on propositions being identical to the thoughts of God. If, instead of propositions being identical to his thoughts, God created a realm of propositions, then they would be accessible to another mind as per Platonism. But if propositions are identical to the thoughts of God then they are contained within his mind and possessed by him and him alone.

The problem also depends on Wittgenstein’s private language argument. The mind of God containing all possible thoughts could only communicate to another mind (prior to creation) if he could speak. But if anything that is expressed by God is only a PME then there is good reason to think that the language he uses is not possible (or, at least incomprehensible)

A private language is a language comprised of words that “refer to what can only be known to the person speaking.”[11] Let us assume that God has one mind. (Richard Swinburne argues that the Trinity is composed of three minds. I am going to assume this is false.)


God has comprehensive access to his inner thoughts and before creation there is no one else who thinks anything. Furthermore, the sphere of inner knowledge is necessarily accessed through introspection even if someone else was present. No one can think the thoughts of other minds unless they are communicated somehow. Let’s say God creates another person with a mind. The other person could not, in principle, have access to God’s mind. God’s speech, then, would entirely refer or express what is solely accessible to himself and no one else. Wittgenstein argues that such a language is impossible. A language that only functions to refer to some thought, sensation or idea within the mind of the one speaking can have no public criteria for any other mind to access. And language, according to Wittgenstein, is not possible without public criteria.
But, what might such criteria involve? At the very least it would include publically knowable senses for the terms of a declarative sentence. Frege distinguished between the sense of a term and its reference.[12] According to Frege, a sense of a word can be understood without an object that is picked out. For example, the term “son” has a sense before anyone has pointed out that Ben is the son of Tim. Accordingly, and applied to the private language argument, when a person uses a language to express a thought the person relies on a public sense even when picking out a private reference. Wittgenstein uses the example of a private sensation. A pain is referred to by the speaker, but relies on a public sense known by other people in order to use the language. The sense of a term, once known, can be used to refer to a suitable entity.

It might be suggested that God is sensation free. He cannot, for example, have toothache. There are two responses to this concern. First, God would presumably have known, before creation, what it would be like for a human being to experience embodiment. God would have known both that human beings would experience pain and what pain would be like. The latter claim does not entail that God could feel pain without a human body. Of course, Jesus felt very real, physical pain during the crucifixion, but God, before creation, knew what pain would be like for human beings. God’s knowing what it is like to be in pain means God knowing all the information about human pain much like a blind man might know information about an object he cannot see that he could gain if he could see.[13] Likewise, if God knows what green is before he makes anything that is green he can communicate information about something green to another speaker only if the other speaker is able to understand what God has to say about green. In the same way a blind man is only able to get information about an object if he is able to understand the language in which it is given.

Second, God’s thought contains propositions that can presumably be expressed in declarative sentences that use terms with both a sense and a reference. God, when describing his plan for creation, had the thought: “a tree will be in the garden.” If merely describing his “vision” for his creation, God would utter the sentence, “there is a tree in the garden.” In order for this sentence to be intelligible to a hearer without the publically accessible tree, garden, and what it is for something to be “in” something else, the sentence would not communicate anything. The sentence might express a true proposition, one that, by day six of the creation week, would have a publically available reference. But it only succeeds in conveying anything if there is a public sense to the terms within the sentence and before creation, there is only a private sense, one within the divine mind.

One might wonder why God could not have a private language to speak about his own thoughts. A common objection to the private language argument is that it seems we can imagine a Robinson Crusoe figure that is placed on an island and develops a language only known to him.[14] Perhaps, but the point is not about an island, a public environment accessible to anyone who might be living on it; instead it is about entities only accessible to the speaker. In respect to divine thoughts, all thoughts are within the divine mind. They are inaccessible to any other mind. In order to be known by another mind the thoughts of God must be expressed. And if God wishes to do this before he creates what is in his mind he must do it with a language or some other communicative medium.[15]

The crux of the private language argument is that language only works if there are publically known rules according to which a given language conveys that information. Wittgenstein argued that this would be impossible for a private language. Crucially for this discussion, Wittgenstein argued that language obeys a set of rules and that those rules are publically known. Wittgenstein argues that if meaning is primarily seen in terms of use then its comprehensibility relies on an implicit convention between speakers about the correct and incorrect use of terms and syntax. It is not that Wittgenstein is holding up a set of rules. Rather, Wittgenstein is suggesting that the necessary condition for the possibility of a correct or incorrect judgment in language use is found in some fact about language users. The fact, it seems, is that users of language enter into a game with a plurality of speakers who all agree to a set of conventions.[16] Roger Harris suggests that we consider the convention to be a normative agreement between speakers not merely what happens to be the way language users do in fact agree on rules.[17] In other words, there is a criterion for use that presupposes a correct and incorrect use of terms. Wittgenstein argues that such criteria are not possible if what one is expressing is only internally known. Whatever term is used seems right and there is no way to tell the difference between what seems right and what is actually the correct way to use a term.

Let us assume that God’s thoughts amount to what we call propositions. From all his propositions, let us consider those that are about creation.[18] Prior to creation, God’s thoughts would have been about creation, but without an actual creation to point to. When he created humans, God could speak to them and tell them about his creation. His propositions about creation, once only known to him, are made public and attach to the world due to the world being exactly as he had determined it to be.[19] Whether or not God taught the first humans to speak or created them with an innate capacity for speech does not matter. What matters is whether or not it is possible for God to have a language prior to creation. Since all the entities that are truth-bearers are only known to the mind of God there is no public criterion for communication, no way another mind can access those propositions, without access to the mind of God. But since only God exists, there is no possibility of such a criteria and therefore no possibility of language.

The solution to this problem lies in the doctrine of the Trinity. The triune nature of God allows God the capacity for speech. God, in other words, could be involved in an eternal conversation. So, how does this work? How can the members of the Trinity have a conversation?

The best answer that I can find is from John Feinberg. Feinberg argues that it is plausible that God’s comprehensive knowledge does not entail that members of the Trinity are equally conscious of all that he knows. God, the Son, may be consciously aware of only some of his knowledge. This does not imply any lack of knowledge in God since what God is not consciously aware of at any given moment he retains as part of what he knows. Any member of the Trinity could, if he wanted, be consciously aware of all his knowledge. If knowledge does not entail conscious awareness there is no reason why God could not have one mind with three centers of consciousness. One member of the Trinity is able to talk about what he is thinking and draw others’ attention to it: “if the members of the Godhead are not always consciously thinking of everything they know, then conversation, drawing of attention to one particular truth they know, and fellowship are possibilities.”[20] This does not entail any lack in God any more than it would in human beings. We would not accuse anyone of not knowing what 6×6 is just because they are not thinking about it at a given time. The same is true of God. It does not appear to demonstrate any deficiency in God’s omniscience just because no member of the Trinity is attentive to the fact that a particular raindrop landed in the place we call Seattle at 5:45am on Jan 1, 1250.

This view successfully answers each condition for both objectivity and linguistic capability. First, the view entails a one-mind view of God. If God has one mind then all the thoughts are his thoughts, they are owned by one mind and logically cohere with one another. Second, the three persons of the Trinity can be three centers of consciousness without having three minds. Third, the language community condition is met. Each person in the Trinity can be aware of thoughts that other members are not (though they could be) and are able to “draw attention” to those thoughts. The thoughts are public in that they are knowable by each center of consciousness and yet are not independent of God’s mind since there is only one mind in which every possible thought inheres. Since each center of consciousness has access to the full array of possible thoughts the condition for criteria for terms is met. Words can have meaning since they can express propositions and can be used in multiple ways obeying criteria for sameness of use. There is, in other words, a genuine community of language users. Even on the strictest views of the impossibility of a private language this view meets every demand. 

There is a cost to such a view. Feinberg notes that this view commits one to the view that God is a temporal being. In other words, for God to be consciously aware of p at t1 and not at t2 God cannot be atemporal. To be atemporal, God would have to be consciously aware of all his knowledge. If this is not an unpalatable cost, then the suggestion is plausible. 
[11] Ludwig Wittgenstein, Philosophical Investigations (Upper Saddle River, NJ: Prentice House, 1958) §243.
[12] Gottlob Frege, (1948). “Sense and reference” Philosophical Review 57 (3):209-230.
[13] John Feinberg, No One Like Him (Wheaton: Crossway, 2001), 307.
[14] Simon Blackburn, Spreading the Word (Oxford: Oxford University Press 1984).
[15] Recall, I am assuming the one-mind view at this point.
[16] This is an argument extended by Kripke whose version of the private language argument concludes that there are no fact about what a speaker means by an expression. Saul Kripke, Wittgenstein on Rules and Prvate Language (New York: Blackwell Publishers, 1992). Cf. Bob Hale, “Rule-following, Objectivity and Meaning,” in A Companion to the Philosophy of Language eds. Bob Hale and Crispin Wright (Oxford: Blackwell Publishers, 1997), 370-374.
[17] Roger Harris, “The Private Language Argument Isn’t As Difficult, Nor As Dubious As Some Make Out.” Sorites: An International Electronic Magazine Of Analytical Philosophy 18 (2007), 98-108.
[18] There are possibly many thoughts God has that no one outside God will ever know. For those thoughts there need be no language. However, when God “speaks” creation into existence we can be sure that God knew how to communicate what was in his mind so that other persons would understand it.
[19] His propositions also include how the world could have been had he chosen to make it that way. Perhaps possible worlds include worlds that are only conceivable by God, but unintelligible to us. Perhaps all possible worlds are this-world relative in God’s mind. Either way, the once private set of propositions becomes, in part, public after God has created human beings. 

[20] John Feinberg, No One Like Him (Wheaton: Crossway, 2001), 319.

Assistant Professor of Philosophy and History of Ideas at Southeastern Baptist Theological Seminary and The College at Southeastern.