Self-Stultifying Arguments (Unfinished)

[A MAJOR disclaimer!

The following material is VERY tentative at this point, was written several years ago, and lacks the rigor and precision that is in its future. I have NOT incorporated precise terminology for the components of speech acts, conversational implicature, theory of abstract objects at this point. There are many iterations of critique that I will go through on this material, but I consider the basic approach to be worthy of research. (I recently 12/94 found a published reference with a similar approach--Karl Otto-Apel "The Problems of Philosophical Foundations in light of a Transcendental Pragmatics of Language" in Philosophy: End or Transformation? (eds. Baynes, Bohman, McCarthry: MIT, 1993)

There is also material that I have just started digesting in Stephen Bartlett's Reflexivity: A Source Book in Self-Reference, North-Holland, 1992. His work validates the general approach of this 'schoolboy' paper, and has many more examples as well as a pointer to a bibliography of over 1,200 works on self-reflexivity. Many/most of these examples, btw, will be exempt from the standard Tarski-esque responses.

As such, I welcome ALL feedback, biblio references, and pointers/advice from those working in this area. Thanks.]

Let's Start with Breakfast...

Imagine the following comical scenario.
A fellow Earthman runs up to you, with glazed and feverish eyes, and proceeds to explain how he has discovered a fundamental and absolute truth about himself. When you ask him to tell you this awesome truth, he blurts out this: "The fundamental truth is that I cannot pronounce or write the word 'breakfast'!" You are not sure you heard him correctly, and so you ask him to write the truth down on a sheet of paper. He then writes legibly on the sheet: "I cannot pronounce or write the word 'breakfast'."

There is something obviously wrong here (other than the fact that the guy's elevator doesn't go all the way to the top floor!) and what the obvious wrong is is clear--the speaker contradicted himself in the process of speaking. He rendered his 'truth' ineffective--he stultified himself.

This is a special case of reductio ad absurdum -- but the absurdity was that "what he said" (the words, pronunciation, the speech act itself) contradicted the "what he said" (the content, the intention, the meaning).

Now let's generalize this type of argument.

The Nature of Self-Stultifying Statements

A self-stultifying statement is a statement that contradicts:
  1. itself;
  2. the case it advances as proof (if any);
  3. the presuppositions inherent in the subject matter being discussed;
  4. the presuppositions inherent in the speech act.
Let's illustrate these cases with a simple example. Cases 1 and 2 do not occur very often, and Case 3 often produces "standard" reductio ad absurdum refutations. Case 4, however, is not as obvious (except with the guy who couldn't say 'breakfast'). Case 4 requires 'unpacking' of the presuppositions in speaking or discourse or language or communication, to see if what is being said (explicitly in the statement) is contradicting what is being said (implicitly in the presuppositions).

One last example before we turn to history for a moment.

Try "All sentences are meaningless." The obvious question to ask here is "including this one?" The statement (no meaning) contradicts the presupposition (sentences are adequate vehicles for meaning). The position is self-stultifying. It cannot be even stated without contradicting itself--it 'pulls the rug out from under itself.' This contradiction will necessarily arise from this statement, and the obvious thing learned is that it is impossible to deny that 'some sentences are meaningful.' This is the value of looking for self-stultifying arguments -- we find undeniable truths or absolutes. We may not be able to produce an air-tight proof for the position, but the fact that they cannot be denied at all can be seen as such a proof.

This is a clear example of Type #4 (contradicting the presuppositions in the speech act). Speakers normally presuppose that their utterances convey meaning (we will explore this more when we get to Zen) before they 'go around uttering them.'

We need to articulate some of these speech-act presuppositions at this point. Each semantic act (sentence, paragraph, chapter) is a semantic unit. As such, the beginning of the unit forms the context for each succeeding part of the unit. The context must be retained (by the reader/audience), building until the end, in order for the unit to have its intended meaning. The semantic unit presupposes an audience -- either an actual audience, or a theoretical one (we fashion our word choices and structure with an intended audience in mind).

Examples From History

There are three examples in history that illustrate usage of this type of argument: Aristotle, the Verification Principle of the Logical Positivists, and the Liar's Paradox.

Aristotle, in The Metaphysics IV.4, is discussing the principle of Contradiction/Non-contradiction. This principle says that one cannot say that something is both A and non-A at the same time. Aristotle argues that this principle does not need proof, because it cannot be argued against. Let's look at his argument in three parts:

Part I: "We can, however, demonstrate negatively even that this view is impossible, if our opponent will only say something; and if he says nothing, it is absurd to seek to give an account of our views to one who cannot give an account of anything, in so far as he cannot do so. For such a man, as such, is from the start no better than a vegetable."
In part I, Aristotle argues that we can demonstrate the truth of the principle of non-contradiction if the opponent will only say something. In other words, if he commits a speech-act, we can show the truth of the principle. (Aristotle closes the part with a little Ad Hominum (or should we say Ad Vegetablum) slur.)
Part II: "Now negative demonstration I distinguish from demonstration proper, because in a demonstration one might be thought to be begging the question, but if another person is responsible for the assumption we shall have negative proof, not demonstration."
Here Aristotle argues that a proper demonstration/proof of non-contradiction would itself assume the principle of non-contradiction ("begs the question"). And that, whoever brings the assumption up, provides us with the negative proof.
Part III: "the starting-point for all such arguments is not the demand that our opponent shall say that something either is or is not (for this one might perhaps take to be a begging of the question, but that he shall say something with is significant both for himself and for another..."
Aristotle finishes this argument by pointing out that it is not important what the person says (content), but that he says anything at all (commits a speech-act). This is, at its heart, a refutation by self-stultification.

The Verification Principle for the Logical Positivists is another interesting example in history. Logical Positivism was an early 20th-century philosophical school of thought that was decidedly anti-metaphysical. It scorned statements like "reality is spiritual" or "beauty is significant form" as being cognitively empty. It advances its famous Verification Principle as a way of determining whether statements were cognitively meaningful or meaningless.

The principle basically said that for a statement to be meaningful it must either be 1) analytic (tautological) or; 2) empirically verifiable. This principle was mainly used in its second case (empirically verifiable) to dismiss all "non-scientific" claims as being neither true nor false, but rather, nonsensical.

There were (and are) many, many criticisms and restatements of this principle; so many perhaps that it might be argued that it "died the death of a thousand qualifications." Our focus here, however, is simply on one of these criticisms: that the principle was self-stultifying. When the principle was 'turned back on itself,' it proved to be neither analytic nor empirically verifiable. It basically rendered itself meaningless!

The Liar's Paradox is one of the best known and most celebrated of the logical paradoxes. It shows up in various forms in intelligence tests, for example. The basic paradox goes as follows:

"The statement I am now making is false."
The 'paradox' in this can be seen by developing the outcomes. If the statement is true, then it is false. If it is false, then it must be true. This is a circularly-destructive statement.

Another form of this, closer to Aristotle's type of refutation, would be as follows:

A person who always lies, says "what I am now saying is true."
In this instance the 'contradiction' is between the character of the speaker (always speaking falsely) and the content of the statement (speaking truly this once).

(The Liar's paradox has been the study of many an epistemologist and many a semanticist. There is no commonly accepted 'solution' to it.)

Self-Stultification and Critical Thinking

The point of developing the ability to detect self-stultifying arguments is to be able to construct Aristotle's 'negative demonstrations' of truth-claims. If an epistemological position (or political or semantic or whatever position) can be shown to be self-stultifying, then it cannot be even advanced for serious consideration. We can then proceed to draw the implications of this inability as a 'negatively demonstrated' absolute. This will not help us at all in verifying or falsifying any position which passes this test, of course, but as we shall see, it will narrow the field considerably, if we use it correctly.

What this amounts to is the ability to stop an argument before it launches and to draw conclusions (negative demonstrations) from that 'stopping.' What we end up with are absolutes--in the sense of undeniables, not ultimates--in human language!

Practice Test One (or "Fun with Sentences")

Let's examine several examples to see this work out in various forms.

No Truth: "There is no such thing as truth." Then, obviously, this sentence is not true, and therefore, there really might be something like truth. You should recognize this as a slight variation of the Liar's Paradox, akin to "all sentences are false." (This assumes that the statement is not about truth as having some type of ontological/physical/metaphysical existence; in which case our approach does not generally apply.) Implication: It is undeniable that some sentences are true.

No Certainty. "You can never know anything for sure." Does the speaker know that for sure? If he does, then it is self-stultifying. If the speaker doesn't know it for sure, then maybe some things can be known for sure. Implication: It is undeniable that some things can be known for sure--certainty is possible.

Sentences and Reality. "Sentences never describe reality, only the speaker's mental states." When we turn this back on itself, the question is obvious: does that sentence say anything at all about sentences or is it only about the speaker's state of mind?! (You should be able to see the pattern emerging by now: any sentence saying something about all sentences is saying something about itself.) Implication: .It is undeniable that sentences can describe reality.

Generalizations. "All generalizations are false." As a sentence of the "all X are..." form, this is itself clearly a generalization. And...we net out with a lair's paradox again--"all generalizations are false (including this one)."

The Next Level -- More Epistemological Examples

Kant-an-Sich. "Something exists external to ourselves, but they cannot be known--only phenomena from them can be known." In the Kantian world, the objects around us 'send out data' towards us as knowers. This data is called phenomena and only this data can we know--we can never get back to the real object itself. Now, applying the self-stultification theme to this statement/position, we ask: "does this statement describe the objects or just phenomena?" The position distances itself from both the things and the phenomena, and in so doing, stultifies itself--it makes knowledge claims about both phenomena and objects. Implication: It is undeniable that we can know objects, and not just their phenomena.

Dewey. "Truth is only that after verification." This is a similar to the Verification Principle. If you apply it back onto itself, it stultifies itself. Is Dewey's statement only true after verification, or is it simply an indefensible definition? Is it not making any truth claims? In other words, Dewey is trying to set up a criterion for truth--before verification! If truth is only truth after verification, then Dewey's statement is not true. Implication:. It is undeniable that there can be truth statements that are 'true' prior to verification.

Marx. "Thought is a reflex of the brain for material life processes related to economics." The problem with all types of 'brain-reflex' positions is that the thoughts produced are related only to the knower's brain, and not at all to the objects 'being known.' Theoretically, I could have a brain-reflex about a cat being in front of me at any time, whether or not a cat was actually in front of me. When we apply Marx's statement back on itself, we arrive at the conclusion that his thought has no relation to reality, but only to the state of Marx's brain at that point. The statement cannot be describing 'real thought' because it is merely about brain-reflexes, not about 'real thoughts.' Implication: It is undeniable that thought is not solely a brain reflex.

(Note also that this applies to any position that says that thought-about-something is a product of something other than the object-being-thought-about. If I construct a position "Thought is not a function of an external object, but of my X" the criticism applies to it mercilessly. The "X" can be anything other than the object-itself and the criticism applies:

If the thought is not dependent on the external object (in some direct way), then its truth claims about the object are irrelevant at best.

Universalism. "All worldviews are true." How about the worldview that says "no worldviews are true"--is it true as well? This obviously cannot be the case. This is the inverse of a form of the Liar's paradox -- 'all worldviews are false.' Notice that the generalizations 'all are true' and 'all are false' cancel one another out, leaving only their implications around. Implication: It is undeniable that some worldviews might be false, and that some worldviews might be true.

Predatory Paradigms. "In the final analysis, there is no truth, only perception. We are all trapped in the walls of our own paradigms, and there are no doors." This is a position in considerable vogue today. Paradigms are 'ways at looking at things,' a set of rules, of expectations, of frameworks. They are very similar to models or extended analogies and are extremely useful in science, in art, and in explanation ('its a lot like X'). The problem is that idol-creating man likes to take something of power (bulls, angels, money, moons, paradigms) and make them into gods with absolute power. It is no exception here--paradigms are seen as absolute in reach and almost oppressively restricting (a la Genesis 3:2-4). From a self-stultification standpoint, the problem is simple: what paradigm allows us to discuss paradigms? The above statement actually includes two problems. The first one is in its first sentence--if there is no truth, obviously that statement is not true (we have seen this before, remember). The second is in the second sentence--if we are trapped in our paradigm, how can we escape it momentarily to objectively see that paradigm? Being trapped in such a paradigm would mean access to only perceived truth, not 'real' truth -- the Kantian problem. The position ultimately destroys itself in its relativistic starting point. Implication: It is undeniable that some paradigms may give access to 'real truth' and not just to perception.

Truth and Propositions. "Truth cannot be expressed in propositions-- it can only be experienced and/or pointed to." (This actually should be in the beginner section, but I thought I would wait to let you pounce on this one!) To utter a proposition, claiming to be true, that truth cannot be stated in propositions, is a perfect example of self-stultification. Implication: It is undeniable that truth can be expressed in propositions.

Cognition versus Emotion. "Truth is not something you know--it's something you feel." This is simply a combination of the preceding (truth is non-propositional) and of the brain-reflex case (truth is related to me, not to the object known). The self-stultification is obvious: how can you know that truth is not something you know? Is the author simply 'feeling' that truth is not knowable? Implication: It is undeniable that truth has cognitive content.

Coherence Theory of Truth. "Truth is not correspondence to some state of affairs, but simply logical consistency within a set of statements or calculus." The message here is that statements are not true because they describe something in reality (states of affairs), but because they do not contradict other statements which are assumed to be true. (We will explore this more in depth in later sections.) There are several problems with the position, but we want to focus only on the self-stultification problems here. And the problem should be obvious: to use a sentence, claiming to be true, to describe a state of affairs (sentences being true or not) and then deny that truth is not related to 'states of affairs' is viciously self-defeating. Implication: It is undeniable that statements referring to states of affairs can be true.

Deus Obscurus. "God is indescribable, beyond the reach of our puny language and logic." If this looks strangely like "Breakfast is indescribable, beyond the reach of our puny language" it is because it is! It is 'highly irregular' (if not pure self-stultification) to 'describe God' as being 'indescribable.' Enough said. Implication: It is undeniable that anything (God, breakfast, concepts) can be described, at least minimally.

Exploring the Speech Act

The preceding examples all had words like truth, sentence, knowledge, thought, worldview, and generalization. In these cases, it is clear how to 'turn the argument back onto itself.' The statement itself was a member of the class of statements being referred to in the statement. If the statement's subject was 'all sentences,' then that subject (the class of all sentences) obviously included the statement itself (as a sentence, it was a member of that class).

All of these were fairly obvious, but these examples illustrate various and different 'parts' of the speech act:

Most of these 'parts' were implicit or assumed in the examples, and actually provided us with 'a place to stand' in examining each argument for self-stultification. But what are some other assumptions in the speech act, and where might they lead us? Let's look at some more, and construct self-stultifying positions from them:
P07: All of the statements used words.
We could easily construct the following statements as examples of self-stultification: Let's expand P07 into:
P08: All of the statements used more than one word, and at least two of the words were different.
This may seem a bit trivial, but it has some surprising implications. Take, for example, the following position:
"All differences are illusionary."
The fact that different words were necessary to construct the position, and that those differences were essential to even making the statement, stultifies the position. Implication: It is undeniable that real differences 'exist.'

I remember well the first time I experienced this line of reasoning. I was in another epistemological discussion, this time with a Ph.D in anthropology. She was vigorously defending a Zen version of the Eastern philosophy of maya--that all differences were illusionary--and that these illusionary differences actually stopped us from seeing the 'real truth' of oneness. I drew her to the implication that, if that were the case, then the very language she used to articulate her position added to the problem. In other words, she multiplied these 'illusionary differences' with her every sentence and in so doing, actually hindered our progress toward "awareness of oneness." I pointed out that if language did so lead us away from the 'real truth,' then the voluminous Vedic literature that she based her beliefs on were major contributors to the problem (not to mention the confusion her lengthy doctoral dissertation must have caused!).

Now let's try another.

P09: All of the statements had an author.

This generates some interesting self-stultification positions:

The first of these has particular relevance to my very first epistemic debate. It could have gone like this:

He: How do you even know you  exist?!
Me: Because it is self-stultifying for me to assert that I do not.
(Notice this would not answer the question "how do I even know you exist?" -- although we will examine this later.)

Let's try one more before we go on to the advanced examples.

P10: All of the statements had an intended audience.
There are a couple of points to be made about this. First, the author always has an intended audience in mind--it determines what language he uses, how technical the terms are, what the slant of the argument takes, what assumptions can be agreed on, and so forth. This is independent of whether or not the argument is actually delivered to anyone. (Even if the author were the last person alive on the planet, he would still presuppose an audience for each utterance.) Secondly, this applies only to the types of statements (philosophical positions) we are discussing here--not rhetorical questions or artistic soliloquy (although it may be perfectly applicable to those cases). Third, this makes statements into social events, subject to shared external norms of linguistic usage. Fourth, language has a dynamic character and has the means to create new expressions and phrases within itself to address virtually any possibility or concept. It can create extremely narrow/precise expressions, or very general/abstractive notions.

What are some self-stultification positions from this?

Let's look at two examples in a little more detail; one, from Lewis Carroll, and the other from older anthropology textbooks.

In Lewis Carroll's Through the Looking Glass, Alice has a linguistic skirmish with Humpty Dumpty (Acting Ultimate Reference Point for All Predication).

He: 	There's glory for you.
She:	I don't know what you mean by 'glory' there.
He:	When I use the word 'glory' I mean 'knock-down argument.'
She:	But 'glory' doesn't mean that!
He:	When I use it, it does.
She:	The question is whether you can make a word mean anything you want.
He:	No, the question is who's to be the master, that's all (he said with a contemptuous 	smile).
HD hit the issue on the head--who controls the language, the speaker or the language. The answer is obviously not the speaker-only. The language itself exerts force and control over the statements, even thought we can create the desired expression out its building block material. Implication: It is undeniable that this statement is a shared, and publicly accessible act.

The second example comes from older college anthropology textbooks. In discussing the power of culture and language over man's thought, the example is often given about some tribe somewhere that doesn't have a word for the human wrist. The most precise word they have is 'forearm.' This example is unfortunately used as proof that they do not even have a concept of 'wrist' or that they cannot 'know' anything about it.

I have two problems with this. The first is that they have obviously deified culture by ascribing absolute power and immutability to it. The second is that every language has the ability to refine and narrow a concept by continuing to heap adjectives, adverbs, and other phrases on top of one another. In the case of wrist, it is so simple to create a phrase like 'that section of the forearm at the end by the hand, in which resides the joint that allows the hand to move at an angle to the forearm, etc.' until the meaning is obvious. Implication: It is undeniable that language provides enough flexibility for an author to express his position.

Storming Small Fortresses...

With this methodology of searching the individual speech act for points of contact with the content of that individual speech act, let's try some advanced examples.

Heraclitus' Unchanging Sentence. "All things are in flux--nothing is the same from moment to moment." Heraclitus was, of course, the waterproof philosopher of ancient Greece (not to be confused with the waterproof captain of the Nautilus)--he couldn't step into the same river twice, because it never was the same river twice. Everything in Heraclitus' world changed ever-so-slightly from moment to moment, and therefore, was never the same (and couldn't be called by the same word). So Heraclitus' position was that time changes everything, and that nothing stays the same even for an instant.

How can we apply this back onto his position to test for self-stultification? Where is 'time' in the speech act of Heraclitus' position? There is 'time' in at least two elements. Speech acts occur at specific times (e.g., noon on Friday, the 3rd day of 505 B.C.) and speech acts take time to deliver, in that the words flow sequentially and after one another. It is this second element that freezes Heraclitus in this tracks.

In a sentence, each successive word builds on the accumulated context of the preceding words. This growing context narrows the range of possible meanings for each word until the end of the sentence is reached. If the words uttered in the first few micro-seconds of a sentence were ever changing, we would never get a context for the end of the sentence. We would never get closure. We would never get to a meaning for the statement. Hence, Heraclitus could never articulate his position. Self-stultification. (Notice that Heraclitus' problem is not how long a time, but the presence of time at all.) Implication: It is undeniable that some things can remain constant through some periods of time.

Historical Knowledge. "You can never know what happened in the past." This is a basic position of skepticism in historical knowledge. There is an implicit belief that only knowledge of the immediate present is possible and certain, and that as soon as something slips into the past, it becomes less certain as less "unknowable."

Many of you will have already attacked and dismembered this one, based upon the arguments against Heraclitus. The obvious approach is to apply the 'through-time' character of the speech act back onto itself. In other words, by the time the author gets to the word "past" in the sentence, where are the preceding words (and context)?--obviously in the unknowable past!

This type of reasoning would obviously apply to similar arguments against memory. (Some might make a distinction between the immediate past and the more remote past, but I do not think these can bear up under scrutiny. The distinction between short-term memory, long-term memory, memory with memory helps, memory incarnated in writing, memory incarnated in cultural artifacts is a matter of degree, not of nature.)Implication: It is undeniable that some of the past can be known.

Self Identity through Time. "The Self has no continual existence through time; it is merely an unrelated succession of states of consciousness." Sometimes this position allows for small periods of existence before the self dissolves (Whitehead's process philosophy said the self existed for 1/16 of a second before disintegrating).

The self-stultification approach would be to restate the position is such a way as to include the author: "The author which began this sentence is not the author who finished it." The key here is to understand what gives a statement its semantic unity--it's the author. The author intends a meaning and constructs a semantic 'whole' to convey that. It is the author who 'holds together' the discrete elements of the sentence until its united expression is complete. The author is presupposed or 'embedded' in the developing context of the words of each statement. The author's meanings, purposes, contexts are present in each word--to change authors in mid-stream is to change the unchangeable past. Implication: it is undeniable that a self can exist through an interval of time.

Pushing the Borders

Let's look at some statements that are not as clear, but that might yield results after more study.

"Experience cannot be trusted." From a self-stultification standpoint, the author is obviously trusting experience in making the decisions concerning word choice, language, assumptions, and so on. So much experience is using in communicating (or even attempting to communicate) that this seems an obvious self-stultification.

It's not quite that simple, however, and the complexity shows up when we try to formulate the implication. Proposed Implication:: It is undeniable that experience can be trusted in some cases. The problem here concerns whether we said anything or not! If you don't know which cases, how can you trust experience? Some of the problem occurs with the flex in the word 'trust' and some of the problem has to do with the flex in the word 'experience.' (We will look at this in more detail in a later chapter.)

"Personality is illusionary." This sometimes is a conclusion of various types of reductionism, in which personality is dissected into its alleged components of brain impulses, hormones, trained animal behaviors, oat bran, tidal cycles, economic conditions, quantum mechanics, and whatever else researchers propose.

The self-stultification approach to this problem would be to get a definition of personality from the author, and see if semantic ability were a component. For example, if an author agreed that personality consisted of volition and cognition, then it would be rather straightforward to show that

P11: All the statements were made through an author's cognition and at his volitional discretion. (He did not have to say it at all.)
This would effectively constitute a refutation of the position. If, however, personality is defined in ways that do not have implications for speech acts (and I do not see how that could occur, quite frankly), then this approach would not be applicable.

"Communication across (pick one: languages, cultures, worldviews, religious persuasions, socio-economic status, paradigms) is impossible." This seems to be self-stultifying in that no two people have exactly the same languages, etc., but that an author presupposes that he/she can cross this divide in each position advanced. At the same time, the author does not admit that this 'translation' radically compromises his message. In other words, she/he assumes that the content can be reformulated in other languages, cultures, etc.

This has many, many implications ranging from Bible translation, hermeneutics, exposition, Van Til's discussions between Messrs White, Grey, and Black, scientific argumentation, evangelism, and position conversions.

The Ultimate Test Case?

Throughout the history of philosophical theology there have been proponents of what might be call the 'presupposed God' position. This position basically argues that God forms the assumption under all existence, predication, virtue, etc. As such, the position maintains, all statements such as all presuppose the existence of an Ultimate Personality.

(to be continued....)

My "Pushback" to this Approach

The Christian ThinkTank...[] (Reference Abbreviations)