Annihilation and Hermeneutics Part III

I have Asperger’s Syndrome. An odd way to start a post on hermeneutics, but it will become apparent as to why I am starting on this note as we proceed. [In fact, part of the reason I have not blogged in a while is because I have been quite bad recently, and I wanted to take some time off.] Asperger’s Syndrome is a mild form of autism, one of the recognized autism spectrum disorders. People with Aspergers and other ASD’s often struggle with a an area linguistic meaning called “pragmatics,” which deals with things such as conversational implicature, presupposition, speech acts, and deixis. For this reason, when I first start studying pragmatics, it opened up a whole new world of meaning in language. Here was meaning in language that was not even found in the words of the text. For example, take the phrase:

I broke a leg last week.

Where does it say in the text that the leg that you broke was one of your own? It doesn’t but the speaker clearly means it.

Now, as I got more and more into pragmatics, I started to encounter Sperber and Wilson, and cognitive pragmatics. A biologist professor friend of mine at church gave me a book on Neuroscience in order to better understand the brain and its functions. As I mentioned in a previous post, he has sort of been helping me along as I have been studying neuroscience, and its relationship to the cognitive aspects of human language. That is where I ran into autism, and, imparticular, started studying Asperger’s more closely to understand why people who have it struggle with Pragmatics, unless they specifically learn those aspects of human language. I have been very grateful to have run into people who have encouraged me to study linguistic pragmatics, not only to help me with Asperger’s, but to help other people who struggle with Asperger’s to likewise understand pragmatics, and be able to better relate to people.

However, just recently, I ran across the following article which gives surprising information about people with ASD’s. According to this study, researchers did a study on the pragmatic discourse of several males with ASD ages 6-22. What they found is that the participants in the study were able to use and recognize every element of Pragmatics. Now, nevertheless, they do not go so far as to say that there are no problems with pragmatics amongst people with ASD. In fact, that is part of the problem. Obviously, people with ASD can use proper phonology, morphology, syntax, and they can use proper semantics and, apparently, know how to use every element of pragmatics as well. Why is it that they have so many problems with pragmatics imparticular?

Their answer is worth quoting:

A standard way of finding out how something works is to see what happens when things go awry. The earliest insights into what function was performed by what parts of the brain, for instance, came from observing which abilities were impaired by lesions or trauma to the part in question. The relevance of this in the context of the present volume should be clear: For those of us who are keen to find new sources of empirical evidence and methods in philosophy, the panoply of “nature’s experiments in pathology” seems a very promising resource. Of particular interest to us is an emerging method in philosophy of language and mind: taking evidence from cognitive and linguistic breakdowns as evidence for “normal” cognitive and linguistic function. It is against that backdrop that the methodological/philosophical question of this article arises. The specific application of this deficit-based methodology that will be our focus is the use of language and cognitive deficits to help trace the boundary between knowledge of language on the one hand, and knowledge of nonlinguistic facts on the other. We begin by explaining this idea, and then we consider its apparently problematic relation to the empirical findings about pragmatic abilities in ASD given above. To those not overly steeped in Quinean indeterminism in the Analytic tradition, or postmodernism in the Continental tradition, it can seem obvious that there is a divide between knowledge of language and knowledge of facts about the nonlinguistic world. Knowledge of phonology and syntax seem to be clearly linguistic. To know, for example, that “exile” can be both a noun and a verb, and that it is pronounced /egzail/, looks to be knowledge about language—specifically, knowledge about English. In contrast, knowledge about other minds (“folk psychology”), general knowledge about history, geography, physical science, and so on, and specific knowledge about the physical and social situation in which a conversation takes place, all seem to be nonlinguistic. Continuing with the example, to know that exile is most often a punishment for a political transgression, that the reason people find it insufferable is they miss their home and their kin, and that in ancient times the Jews were exiled to Babylon, seem to be bits of knowledge about the social and mental world that are not properly linguistic. Similarly, knowing who has said “I am in exile,” where they said it, and why, are all beliefs about the speech situation that are not derived from knowledge of language. When one presses a little, however, it becomes hard to see exactly which information fits where: Is the fact that exile can be self-imposed a fact known about the word “exile” or a fact known about the state of exile? Is knowledge of the etymological history of the word “exile” knowledge of language? The answer does not come pat. It is because of such hard cases that the issue of the language/world boundary arises. An especially fraught class of cases of knowledge of language versus knowledge of the nonlinguistic world involves the semantics/pragmatics boundary.

In other words, the problem with autism in all of its forms is a difficulty conceiving of and entering into other people’s world. That can be less severe with Asperger’s, or extremely severe with full blown severe autism. However, what these authors seem to be saying is that this is the very reason for struggles in pragmatics on the part of people with ASD, because there is non-linguistic knowledge that is crucial to our understanding of meaning in language.

In other words, let us use an analogy. Let us make a distinction between a person’s ability to use oars in a canoe, and the oars themselves. A person may have the *ability* to use the oars, but if he has no oars, the boat will go left and right without steering, and the boater will look incompetent. Think of the ability to use oars as pragmatics, and the oars themselves as non-linguistic knowledge. Without that non-linguistic knowledge, even if one knows how to use that non-linguistic knowledge, the boat will steer out of control. What this points out is our understanding of meaning in language is based, not just upon the sounds, words, and phrases themselves, but upon our understanding of various worlds-knowledge that, itself, is not properly linguistic. They suggest that the semantics/pragmatics boundary is between knowledge that is linguistic [such as what you may find in semantics], and knowledge that is non-linguistic, semantics dealing with linguistic knowledge, and pragmatics dealing with the relationship between linguistic knowledge and non-linguistic knowledge. If you reject this, then, the problem is, it is impossible to explain autism, given that people with ASD have all of the major tools to form language properly.

So, how are we to understand this interface between semantics and pragmatics? The two best suggestions I have seen deal with the “pragmatically enriched said” of Recanati and the relevance theorists, and the Levinsonian notion of a pragmatic intrusion into what is said by conversational implicature. To turn a corner now, and show how all of this is relevant to annihilationism, I have a friend of mine who I met in James White’s chat channel by the name of DeoVolente. He told me that he believes that pragmatics are completely ignored in the hermeneutics of annihilationism. After thinking about it for a while, and talking to these folks, I am convinced that he is right.

I am not going to go into detail concerning these theories of conversational implicature, but they are the background to what I am going to say. The concept of conversational implicature is something I have introduced many times on this blog, especially in dealing with Matthew 1:25 and the word “until” as well as dealing with Voddie Baucham’s argument concerning the phrase “all who could understand.” As can be seen, annihilationists are not the first ones to have problems with pragmatics. Nevertheless, I will not repeat everything I wrote there. However, suffice it to say that such material will be important as we deal with the arguments from annihilationism.

For example, one of the things that these folks raised hell about [no pun intended] is the notion of the Qal stem being used as a passive. The problem is that, if you ignore the notion of pragmatic intrusion into the semantic level, it makes sense why one might say that the Qal can be passive in meaning. This is how I thought that these scholars they cited might try to argue for the notion of a Qal stem as a passive, by showing that you can find verbs where the Qal is used, and yet, the object was clearly acted upon. However, consider the verb “to go out” in English. Very clearly, in English, that verb is not a passive. Yet, note how pragmatic intrusion due to relevance can give you the meaning that someone put it out, even though the verb itself doesn’t indicate that:

John took a deep breath, and blew hard at the candle flame. Then the candle flame went out [ingressive stative].
+>The candle flame was put out. [passive]

Because of this pragmatic intrusion into the meaning of “to go out,” the meaning of the sentence is that the flame was put out, even though the text says “to go out.” Hence, the passive meaning intrudes into the semantics of the verb “to go out” due to the relevance between breathing hard on a candle flame and it going out. The question I have is how do these annihilationists or any of these scholars know that what they perceive as a passive meaning to the Qal is not, in reality, a pragmatic intrusion into the semantics of the verb? A proper division of labor between semantics and pragmatics is crucial at this point to avoid confusing issues of pragmatic intrusion with issues of the semantics of the verb itself.

However, as I have mentioned before, the other problem with conversational implicatures is that they can be defeated. A number of what annihilationists call interpretations where they are looking at the “context” are really nothing more than conversational implicatures. However, the question remains whether these alleged implicatures can be defeated by background assumptions. For example, from the last article by Glenn Peoples, he writes:

So what would we learn by taking verb stems into account, such that it would override the observations which Chris made? For one, no interesting facts about verb stems can overturn the observation that the objects of divine judgement in Isaiah 66 are described as corpses or carcasses. This is because the word translated this way is not a verb and has no stems. It is the plural form of a construct noun (בְּפִגְרֵי, “the carcasses of”). Even if we were to grant everything that Blauser says about the verb for “be quenched” In Isaiah 66:24 (and we do not), the passage, if it explains the nature of final punishment at all, still clearly supports the annihilationist view because it portrays the enemies of God as having been slain in judgement. It might seem strange that the fire would keep burning forever afterwards, but we could not conclude that the subjects of divine wrath are alive in the fire. So a foray into verb stems is not “fatal” to the position that Isaiah 66 speaks of the death of God’s enemies rather than the eternal torment.

However, what if this interpretation of the fire going out once the corpses are consumed is simply based upon conversational implicature? If that were the case, then we would have this:

Their worm will not die.
+>Their worm will not die until the corpse is consumed.

Their fire will not be quenched.
+>Their fire will not be quenched until the corpse is consumed.

Also, one might also point out that the notion that they are slain in battle, and thus not alive in the fire is something else that comes from the notion of conversational implicature:

Isaiah 66:16 For the LORD will execute judgment by fire And by His sword on all flesh, And those slain by the LORD will be many.
+>Once they are slain, there is no life in them, and hence we are not talking about eternal conscious punishment.

One can already see how such conversational implicatures are accepted as “context” by annihilationists. The problem is such conversational implicatures are defeasable by further context and background assumptions. For example, the background assumption of the text is clearly that this slaying is imagery [is God really going to have a literal sword that he uses to literally pierce his enemies??????] They seem to think that the slaying with the sword here is a literal piercing of the enemies of God, even though the text clearly indicates that this is imagery, as our background assumptions about God would show [since God doesn’t need a sword]. Even Calvin recognized this view is way too simplistic:

These metaphorical expressions are very customary in Scripture; for we could not comprehend this dreadful judgment of God in any other way than by the Prophets employing metaphors drawn from known and familiar objects. (2 Thessalonians 1:8; 2 Peter 3:7.) By means of them the prophets endeavor to make a deep impression on our senses, that, struck with the true fear of God, we may not envy the wicked, for whom such dreadful vengeance is prepared. Hence we see how trivial and useless are the speculations of the Sophists, who dispute about the refined nature and qualities of that fire; for the design of Scripture is to point out to us under figures the dreadful judgment of God, which otherwise we could not imagine or understand. This is still more evident from the word “sword,” in the following verse; for it conveys the same meaning. [John Calvin, Commentary on Isaiah]

As one can readily see, this is also an error in speech acts. The assumption of Peoples is that the imagery is meant to be one to one with reality. Hence, if corpses must stop burning at a certain point, so must it be with the corpses used in this imagery. If the corpses have been slain, then they are not alive in the imagery, and hence, must not be alive in reality. However, such is fallacious. The statement “The Lord is my shepherd” does not mean that God is a literal shepherd with a literal robe, and even a rod and a staff. Using People’s logic, one might even conclude that God *must* have a literal staff because, if you look at the “context,” it says “Your rod and your staff they comfort me.” Imagery is meant to convey some similarities and some differences. To demand similarity in all areas is to completely violate the illocutionary force of imagery, as can be demonstrated with the statement “The Lord is my Shepherd.” However, if the background which is to be brought to this text is that the slaying and carcasses mere imagery for the defeat of God’s enemies [as one must hold unless one is going to argue that God will use a literal sword to literally pierce his enemies through], then it will defeat the conversational implicature Peoples has postulated that, once they are slain, there is no life in them. In other words, Peoples took the wrong background assumptions because he didn’t understand that this speech act was imagery. That lead to the wrong implicature.

The same appeal to background assumptions defeating conversational implicature can be made in regards to the conversational implicature that these corpses will burn until the fire goes out. James White has made an excellent argument that the eternality of hell is based upon the fact that men do not stop sinning once they get into hell. The only thing that can remove a sin nature is the blood of Jesus Christ. Hell can never take away our sin nature, and hence, we will not stop sinning in hell. In fact, with the restraint of God fully removed, one can only imagine the evil that is done in that place. However, if that is the case, then they will simply be continually piling up the wrath of God, and hence, the punishment will never end, because the sin will never end. If this is the background against which we are to read the burning of the corpses and the eating of the worms, then clearly, if those punishments are punishments for sin, they will never stop, because the sin will keep coming. Thus, the implicature that the fire will only burn until the corpses are consumed is defeated by our background knowledge of reality.

One can readily see that the acknowledgement that there is a pragmatic level of language causes major problems for annihilationism, and is, as DeoVolente said, the one area of linguistics they completely ignore. Not only do annihilationists take the wrong conversational implicatures, misunderstand the illocutionary force of various passages, but all of this is a result of not allowing scripture to define its own world. All of these things must be read against the fact that man has a sin nature, that our only hope of escape from this sin nature is the blood of Christ, and that hell, therefore, is not an escape from this sin nature. It is also recognizing the illocutionary force of things like imagery, and not confusing the illocutionary force of imagery for the horrible nature of the wrath of God with the literal way God will punish his enemies when he returns. If you do not recognize these things as background, they you will take the wrong conversational implicatures to try to get out of the plain meaning of the passage. One begins to see the relevance of me studying pragmatics in order to better understand struggles with Asperger’s, and the way in which linguistic knowledge relates to non-linguistic knowledge. It is the relationship between the linguistic knowledge and the non-linguistic knowledge that, I would say, is the achillies heel of this system also. Language is meant to relate to reality. If you do not relate what you are reading to nonlinguistic knowledge, you will always have bad exegesis.

One can also see the relationship to the first two posts as well, especially the first. The notion of the analogical nature of human language is something that is crucial to the illocutionary force of imagery. It demonstrates the necessity of things like language games to the understanding of how imagery is used. Thus, this series of articles has, in a sense, come full circle.

More than that, when we turn to passages like Matthew 25:46 and apply pragmatics, the problem becomes even more devastating. The text reads like this:

Matthew 25:46 “These will go away into eternal punishment, but the righteous into eternal life.”

The obvious conclusion one comes to at this point is that, just as life is eternal, punishment must also be eternal. However, annihilationists have tried to argue that “eternal punishment” here is a result noun; that is, it describes the result of the punishment, not the punishment itself. While “result noun” is a semantic category, what they are trying to say is that the meaning is parallel to the phrase “eternal sin.” That is, we are talking about not a sin that is done eternally, but a sin that has eternal consequences. TurretinFan and Chris Date have gone back and forth on this topic here, here, and here.

In this discussion, TurretinFan points out that the semantic category of “deverbal result noun” does not always refer to the results of an action at all. For example, TurretinFan gives the following example:

His back was injured during the first quarter of last night’s game. During the injury, he also hurt his left forearm.

Obviously, we don’t mean to say that during the results of the injury, he also hurt his left forearm.

Because of this, I used to think that compositional semantics were what was crucial here, but I am starting to wonder whether there is a relationship between compositional semantics and pragmatics. Date responded to this by saying that deverbal nouns are polysemous. That is one possibility, but the problem is that the notion of context and even background assumptions crucial to the meaning of result nouns. This opens up the possibility that it is not the semantics of result nouns have changed, but that there is a pragmatic intrusion into the semantics of result nouns, and, indeed, all nouns derived from verbs.

This can be seen in the discussion that ensued between Date and TurretinFan. TurretinFan argued that the context very clearly indicates that this punishment is eternal, because it is paralleled with “eternal life.” Hence, if you are going to argue that “eternal punishment” only means that the results of the punishment are eternal, then you are caught believing that “eternal life” means that only the results of life are eternal. Such is an awkward position to be in. Date, however, gives this response:

TurretinFan, however, appears to insist on a different element of the local context as the means by which we must determine whether “punishment” in Matthew 25:46 is a result or a process noun. He writes, “when ‘eternal punishment’ is placed in parallel with ‘eternal life,’ we are given an unmistakable clue that the ‘event’ or ‘manner’ sense is intended.” In other words, what determines one noun’s reading is that of its nearest neighbor. But this is not true. If a mechanic were to repair the engine of one’s car, guaranteeing that both the parts and labor will last for a year, one would naturally understand that while the parts themselves would function properly for a year, the laboring would not; the outcome of the labor would last for that period of time. TurretinFan’s test would render the guarantee nonsensical.

One can readily see where pragmatics falls in at this point. Because of the relevance of eternal life and eternal punishment, TurretinFan has seen the following implicature:

“These will go away into eternal punishment, but the righteous into eternal life.”
+>Eternal punishment is eternal in duration in the same way that eternal life is eternal in duration.

However, Date compares that to his statement about the guarantee on the car. He compares it to an auto mechanic saying the following with the conversational implicature obviously defeated:

I guarantee that both the parts and labor will last for a year.
~+>I guarantee that the parts will be a year in duration in the same way that I guarantee that the labor will be a year in duration.

Now, why does one elicit the conversational implicature, but the other one does not? The answer is very simple: our background assumptions about reality. Background assumptions about reality can defeat conversational implicatures. Consider the following example cited by Huang in his textbook on pragmatics [p.33]:

John and Mary bought a car last week.
+>John and Mary bought a car together, not each one separately.

Yet, note that the following example, although the same in form, does not elicit the same implicature:

The Americans and the Russians tested the atom bomb in 1962.
~+>The Americans and the Russians tested the atom bomb together in 1962, not each one separately.

Why is it that, although both statements are in the exact same form, the first elicits a conversational implicature while the second does not? The answer is very simple. The conversational implicature that would be elicited by the second statement runs contrary to our background knowledge of reality. We know, given our background assumptions, that the US and Russia were enemies in 1962, and hence, could never have tested the atom bomb together. Because of its inconsistency with our background assumptions, the implicature is defeated.

Now, all of this looks quite similar to what Date and TurretinFan were arguing about. Note the implicatures again:

“These will go away into eternal punishment, but the righteous into eternal life.”
+>Eternal punishment is eternal in duration in the same way that eternal life is eternal in duration.

I guarantee that both the parts and labor will last for a year.
~+>I guarantee that the parts will be a year in duration in the same way that I guarantee that the labor will be a year in duration.

The reason why the second statement does not elicit the same conversational implicature as the first statement is because the second statement is contrary to our background assumptions about reality. Auto mechanics do not, as a practice work on cars for a year. This background knowledge will defeat the conversational implicature that the labor will last for a year means that it will be a year in duration. However, is there anything in our background knowledge that would defeat the conversational implicature that punishment is eternal in duration? What in our background knowledge about reality is inconsistent with the notion that final punishment is eternal in duration? Nothing. Hence, in the first one, there is nothing in the context to defeat the implicature, there is nothing in our background assumptions to defeat the implicature, there are no semantic entailments that will defeat the implicature, and there is no metalinguistic negation. Hence, the implicature comes through unblocked.

What we have going on was a pragmatic intrusion into the phrase “eternal punishment.” However, such intrusions are defeasible, and the annihilationists have unwittingly picked an example where that intrusion is defeated by background assumptions. In other words, although the two texts were *semantically* parallel, they were not *pragmatically* parallel. Because of the ignoring of linguistic pragmatics, the examples that are given are simply not parallel.

As one can readily see, pragmatics is a major area of neglect by annihilationism. I must thank DeoVolente for suggesting that too me, and these series of articles have all been to build up to this notion of pragmatics. Hence, I do owe the idea for them to him. However, to ignore the pragmatics of human language is to do violence to human language. Not to mention the fact that it makes things like the verbal difficulties of autism and Asperger’s completely unintelligible. Thus, I would conclude that annihilationism’s hermeneutics are inconsistent with the way natural language operates, and thus is not exegetically sound.