The Contingency of Language Causes Problems for King

Edward Perez, Santa Fe College

“But on the present view, prior to the existence of language there were no propositions, which are the objects of attitudes like intending and believing.”

Jeffrey King, The Nature and Structure of Content

Introduction

According to Jeffrey King, language is temporally prior to thought. More specifically, language must exist in order for human beings to believe, entertain, and doubt propositions. His reasons for thinking this stem from his account of propositions.¹ According to this account, propositions are structured entities which are composed of both non-linguistic (e.g., properties and relations) and linguistic elements (e.g., lexical items, sentential relations, and logical form representations).² Without language, these linguistic elements cease to exist. Thus, without the existence of these linguistic elements, propositions cannot exist. Hence, on King’s view, if language does not exist, propositions do not exist. King’s position also entails that the existence of language itself is contingent, and that, thus, the existence of propositions is contingent. Obviously, language exists now, so at some moment in the past language was born and along with it came its younger twin, the proposition.

A commitment to the contingent existence of language presents at least two problems for King’s account of propositions. The first problem concerns the time in the distant past when language and propositions did not exist. On traditional accounts of propositions, and on King’s account, propositions are taken to be the ultimate truth-bearers. Therefore, if there was a time when propositions did not exist, then there was a time when nothing was true or false. Hence, on King’s account, in the past there were no truths or falsehoods. I call this “The No Past Truths Problem.”

The second problem concerns the moment when language was born.³ Consider the first language user, call him “Linguaneer,” and call his language “Languish.” King would have us believe that prior to the existence of Languish, Linguaneer had no beliefs, doubts, or thoughts. This is because beliefs, doubts, and thoughts have propositional content, and propositions did not exist prior to the existence of Languish. I call this “The Birth of Language Problem.”

In what follows I will argue that the No Past Truths Problem is more severe than King acknowledges, and I will argue that King’s proposed solution to the Birth of Language Problem fails.

Before moving any further, let me clarify what is meant by “language.” Arriving at a satisfactory definition is thorny and complicated work. Fortunately, for our purposes, all that matters is that language requires the existence of the following elements: lexical items, semantic values, syntactic relations, sentential relations, and logical form representations. So, if none of these linguistic elements exist, then King’s understanding of language fails to convince.⁴

The No Past Truths Problem

Consider a moment just after the big bang, the moment when physical matter burst onto the scene. At that time, matter existed. At that time, language users did not exist and so, according to King, neither did language. Therefore, at that time it was neither true nor false that matter existed. Two clarifications are in order. First, it is consistent with King’s view that matter existed at that time. What his view entails is that it is not the case that it was then true that matter existed, nor was it then false that matter existed. Second, King thinks it is true now that matter existed jus after the big bang. Propositions exist now, so there is now something which fulfills the role of truth-bearer. Nevertheless, a consequence of his view is that at that time, the moment just after the big bang, it was neither true nor false that matter existed.

King entertains the idea that one might object to this consequence of his view. King argues that “If the present view is committed to there being no propositions, and hence no truths, in the remote past, it is just wrong, an objector might say.”⁵ It is unfortunate that before providing his initial response, King does not say what it is that he thinks is motivating the objector to claim that King’s view is “just wrong.” What, according to King’s interlocutor, is the problem with there being nothing true in the distant past? This is difficult to say, but based on King’s response I suspect that he has in mind that his interlocutor is responding based on an intuition about the claim that there is nothing true in the distant past. Let me explain.⁶

King responds to his interlocutor by pointing out that just because there was nothing true in the remote past, it does not follow that the past was not a certain way. As mentioned above, it is perfectly consistent with King’s view that matter existed in the remote past. So, King can consistently hold that in the remote past reality was a certain way, namely, that matter existed in a specific state. All that his view entails is that it is not the case that it was then true that matter existed, nor was it then false that matter existed. King seems to be concerned that his interlocutor might be worried, perhaps because of an intuition, that if nothing is true in the past, then reality was not a certain way in the past; hence, King’s response to the No Past Truths Problem.⁷

I suppose that one might be initially worried that if there was nothing true in the past, then the world wasn’t a certain way. Surely, this is not a problem since there were no false propositions in the past either. In the past, there were no truths and there were no falsehoods. One might be initially worried that propositions cannot change their truth-value. Setting aside the reasons for such a view, we can simply add that that is not a problem here since there is no change in truth-value on King’s view. His view entails that propositions begin to exist when language begins to exist, not that they change truth-value when language begins to exist. King, I believe, has available to him good responses to such initial worries.

But there are other worries that make the No Past Truths Problem worse for King. Here are further considerations for initially thinking that any view which entails there are no truths in the distant past is “just wrong.” Consider future tense propositions such as the proposition “In the future, humans will begin to exist” (anywhere in the universe for the first time).⁸ According to King, this proposition has never been true. In fact, this proposition has been false from the first moment of its existence. The reason is that humans existed before language; therefore, when propositions began to exist, humans already existed. So, at the first moment of the existence of the proposition “In the future, humans will begin to exist,” humans already existed, so the proposition was false. In sum, the claim that there were no propositions in the remote past entails that the proposition “In the future, humans will begin to exist” was never true. But, his interlocutor might say, surely there was a time when it was true that in the future, humans will begin to exist. So, King needs to say more to successfully respond to his interlocutor.

Consider the proposition “In the future, language users will begin to exist.” King’s view entails that this proposition has never been true. But, his view also seems to entail that this proposition is not even a metaphysically possible truth. In other words, there is no possible world in which the proposition is true because every world that contains that proposition is a world that contains language users, and, given King’s view about language, language users must exist prior to the existence of a language. Hence, in no possible world does the proposition exist without language users existing. So, the proposition that in the future, language users will begin to exist is not possibly true. In a similar line, consider the proposition that language began to exist. This proposition, on King’s view, cannot be false because if it exists, language exists and, thus, must have come into existence at some determinant starting point. These considerations make the “just wrong” charge more plausible than King gives it credit for and requires further response.

The final negative consequence for King’s view concerns the existence of language itself. As mentioned, King takes language to be contingent. Even though language actually exists, things could have been different. It could have been the case that language did not exist. It is not that way, but it could have been. Consider the proposition “language does not exist.” On King’s view, this proposition could not have been true. Language must exist before the proposition does, thus, whenever the proposition exists, language must exist. But, whenever this proposition exists, it is necessarily false. The existence of language is contingent, yet the proposition that language does not exist is not possibly true.

In short, King’s view of propositions entails the following: (1) it has never been true that in the future, humans will begin to exist, (2) the proposition “In the future, language users will begin to exist” is not possibly true, and (3) the proposition “language does not exist” is necessarily false. Such considerations, I have argued, make the No Past Truths Problem more severe than King admits.

The Birth of Language Problem

Language appears to be a language users’ endeavor. What I mean is that language seems to require the intentions, beliefs, and thoughts of language users. Never does this seem more true than when language first arrived. On the occasion of the use of the first linguistic sign, say, a bodily movement, it seems that the language user had to have intended, believed, and thought something, however primitive the intention, belief or thought would have been. It seems highly implausible that a being such as Linguaneer⁹, the first language user, could succeed at creating the first elements of the first primitive language without believing or intending anything. Suppose Linguaneer’s first use of language involved making a noise. At the very least, it seems that he would have intended that noise to communicate something, and he would have believed that making that noise would do the trick. On King’s view, Linguaneer had no beliefs or intentions until the first primitive language had already appeared.

King provides a three-fold response to the Birth of Language Problem. First, he claims that if the language of thought hypothesis is true, then there is a clear solution to this problem. King, however, says that he does not want the plausibility of his account to depend on the truth of the language of thought hypothesis.¹⁰ So, he provides a second response, one independent of the truth of the language of thought hypothesis. This independent response will be my target. King’s independent response is an appeal to what he calls “proto-intentional states.” King’s response is to claim that a version of the Birth of Language Problem haunts all accounts of propositions, so his account is no worse off because of it. I will now argue, first, that King’s appeal to proto-intentional states falls short of providing a satisfying response to the problem, and second, that propositional accounts, according to which propositions exist eternally, are much better off answering the Birth of Language Problem than King’s account.

King’s second attempt at solving the Birth of Language Problem appeals to what he calls “proto-beliefs” and “proto-intentions.” He says:

Strictly speaking our proto-linguistic ancestors did not have propositional attitudes, because propositions didn’t exist then. Hence they did not have the sorts of beliefs and intentions we enjoy today. However, we can say that they had some sorts of “proto-intentional states”: proto-beliefs and proto-intentions.¹¹

The prefix “proto” suggests that these beliefs and intentions are the first beliefs and first intentions, but this is not what King means here. The reason is that beliefs and intentions, due to their having propositional content, require propositions and, according to King, proto-beliefs and proto- intentions “don’t have propositional content.”¹² Thus, proto-beliefs and proto-intentions are not beliefs and intentions in the propositionally structured sense. What King seems to have in mind is that proto- beliefs and proto-intensions are belief-like and intentional-like entities that exist prior to regular run of the mill beliefs and intentions.

How does appealing to such entities help solve the problem at hand? King continues:

The idea would be that these proto-intentional states were enough to begin to attach lexical items to semantic values and more generally to do what had to be done to bring propositions into existence. The view, then, is that propositions and real intentional states with propositional content came into existence together.¹³

The picture seems to be that the proto-intentional states are what get primitive language off the ground, and once language is airborne, propositions and “real” intentional states follow. Therefore, beliefs with propositional content are not needed to bring propositions into existence. In sum, proto-intentional states are not full-blown intentional states, that is, they have no propositional content, yet they get primitive language off the ground.

My Response to King

King’s appeal to proto-intentionality falls short of providing an adequate solution to the Birth of Language Problem. To see this, notice the tremendous burden that is placed on proto- beliefs and proto-intentions. They not only enable beings to develop lexical items (with semantic referents), but they must also enable beings to develop the precursors (whatever those may be) to syntactic structure.¹⁴ It seems highly unlikely that such lexical items and precursors could develop in beings whose “thoughts” lack any propositional content whatsoever. What makes this all the more problematic is that King says very little about what he takes the proto-intentional states to be.

All King offers is an admission that he is “not sure exactly how to describe such states that play a role in ‘proto-intentional’ action,” and a suggestion that proto-intentional states can be found in some animals today.¹⁵ He thinks that it is “likely” that many animals today have only proto-beliefs and proto-intentions, that is, belief-like states that have no propositional content. Which animals does King have in mind? I’m not sure, but the key thing to remember here is that for this example to carry any force, the animals must have proto-beliefs that could enable them to develop lexical items (with semantic referents) without there being any propositional content associated with those proto-beliefs (or with any other proto-beliefs animals may have).

Surely, there are animals that are capable of using a primitive set of lexical items, gorillas for example. But it seems dubious to deny that gorillas are capable of having some beliefs with propositional content. Of course, the propositional content involved in their beliefs is not nearly as complicated as that involved in a human’s beliefs. Something as simple as a desire for a banana, or an intention to obtain a banana, will do. Furthermore, these simple desires and intentions must be connected to any lexical item that a gorilla might use. Therefore, it seems highly plausible that gorillas have some very basic propositional content to their beliefs. Perhaps King has some other animals in mind, but which ones? For a particular example to work, it must be of an animal that uses a primitive language involving lexical items without the aid of propositional content. I am not as confident as King that there is such an example.

But, what if I am wrong about this, and gorillas do have real, albeit crude, intentional states? Remember that the proto-intentional states must also allow for the development of the precursors to syntactic structure. Throwing this into the mix would, I think, rule out gorillas as having the kind of proto-intentional states King needs for his example to work. At the very least, we need an argument for the claim that the primitive use of lexical items gorillas engage in involves a type of syntactic structure that can usher in the type of syntactic structure needed for propositions to exist, and King has said nothing to help here.

A big problem with this response to the Birth of Language Problem is that King just does not give any details about what these proto-intentional states are like. I’ve called these states “belief- like states” but they are really pre-belief states that (1) lack propositional content, (2) enable the development of lexical items (with semantic referents), (3) enable the development of syntactic structure, and (4) enable whatever else has “to be done to bring propositions into existence.” Hence, according to King, these states are enough to get primitive language off the ground. Without anything more than gesturing toward the animal kingdom, King falls short of providing a satisfying response.

My final remarks concern King’s claim that with respect to the Birth of Language problem, his account is no worse off than other accounts. According to King, even propositional accounts (according to which propositions are eternal entities) face a version of this problem. Such accounts must answer how it is that the first beings to have real beliefs and intentions by grasping propositions “managed to get in cognitive contact with propositions.”¹⁶ He thinks that an appeal to proto-intentional states is the most promising solution for such accounts. Are the problem such accounts face just as bad as the problem King faces? No. Let me explain.

The problem facing propositional accounts that take propositions to be eternal entities is one of explaining how it is that the first real believer, say, moved from non-belief to belief by grasping a proposition for the first time.¹⁷ According to these accounts, propositions existed prior to the first belief, so what must be explained is how it is that a being came in contact with an already existing entity for the very first time. There is no problem concerning how the propositions originated because on these accounts, propositions are necessary existents. Of course, there are other things that such accounts need to explain, but necessary entities have no origin. It is King’s account that has the special problem of explaining the origin of these contingent entities.

Notice that, according to the other (non-Kingian) propositional accounts, the way a being encounters a proposition for the first time does not require proto-intentional states rich enough to get a primitive language off the ground. In fact, the first propositions grasped do not need any connection to language at all. If something like a proto-thought¹⁸ is appealed to here, at least there is no requirement that the proto-thought must enable the development of lexical items and the like. But, as we’ve seen, King’s account demands more on this score. The proto-intentional states that he appeals to must be rich enough to allow language to get off the ground. Not only is this an extra requirement of such states, as we saw above, this is also a tough requirement for them to fill.

Note that King, like the other propositional theorists, must provide an explanation of how it is that the first believer encounters an already existing proposition. Once language exists, propositions exist. In addition, King still owes us an explanation of how the first believer “managed to get in cognitive contact with propositions” once propositions, as he understands them, exist. As I see it, King’s account must do what other accounts must do plus explain the origin of proposition and make sense of proto-intentional states that are able to facilitate language. Therefore, I conclude that it is King’s view that is worse off with respect to the Birth of Language Problem.

In conclusion, King’s second and third responses to the Birth of Language Problem are not successful. He will have to do more if he wants to avoid resting on the language of thought hypothesis as the way to solve the problem.¹⁹

Bibliography

King, Jeffrey. The Nature and Structure of Content. Oxford UP, 2007.

The most thorough presentation of King’s account is found in his book The Nature and Structure of Content (Oxford: Oxford UP, 2007).
His account of what propositions are is quite complicated. To see this, consider what he says is the proposition that Rebecca swims. It is the following fact: “there is a context c and there are lexical items a and b of some language L such that a has as its semantic value in c Rebecca and occurs at the left terminal node of the sentential relation R that in L encodes the instantiation function and b occurs at R’s right terminal node and has as its semantic value in c the property of swimming” (King, The Nature and Structure of Content, 58, emphasis added). The italicized phrases reveal that the proposition that Rebecca swims cannot exist without language existing.
King credits Scott Soames for first stating this as a problem for his view (King, The Nature and Structure of Content, 65).
Of course, language is much more than this, especially natural language, but what more will not concern us here.
King, The Nature and Structure of Content, 68, emphasis added.
Thanks to an anonymous referee for insightful comments that helped me improve the points made in this paragraph and the next.
To be more accurate, the issue is King’s response to the problem when it’s combined with his commitment to presentism.
Since there is no time specified, perhaps this way of putting it is too quick. Instead, I should consider all the propositions that could be expressed by utterances of the English sentence “In the future, humans will begin to exist” at different times. No matter what time this utterance is indexed to, there will be a problem for King’s view (or so I argue). For ease of expression, I will continue to speak of the proposition that, in the future, language will begin to exist.
Again, we are assuming, with King, that the language of thought hypothesis is false.
King claims, “I am not happy resting my response here on the claim that there is a language of thought and our proto-linguistic ancestors thought in it. So let us suppose this claim is false” (King, The Nature and Structure of Content, 66.).
King, The Nature and Structure of Content, 66.
King, The Nature and Structure of Content, 66.
King, The Nature and Structure of Content, 67.
I am assuming that developing precursors to syntactic structure is a part of what “more generally…had to be done to bring propositions into existence” (see King, The Nature and Structure of Content, 67).
King, The Nature and Structure of Content, 66-67.
King, The Nature and Structure of Content, 67.
My point here does not require me to take a stand on the truth of such theories. Plausible or not, these theories are better off with respect to the Birth of Language Problem.
I use the expression “proto-thought” here because “proto-intentional state” carries the aforementioned attachment to the development of lexical items, syntactic structure, etc.
I wish to express my appreciation to an anonymous FPR referee for helpful comments on a previous version of this paper.