Conversation and Logic
Carl H. Flygt
May 2007
I think what we all want from conversation and human relationships generally is a kind of truth-functional freedom. We want the words and sentences we hear and speak to carry a certain power, to produce images and responses in our minds that give us a certain satisfaction, a certain completeness with ourselves, like the blisses that can arise in our better and sweeter dreams. The notion of truth-functionality is fundamental to the modern concept of logic, and is both commendably simple and tremendously powerful, as we see in modern computational devices. The notion of truth in natural language and discourse, by way of contrast, is somewhat ambiguous and remains undecided. The truth predicate may, for example, represent a correspondence of some kind between linguistic items such as sentences, and facts about the world, but this correspondence view of truth is fundamentally problematic. Correspondence with other sentences seems to be the only thing a sentence in general can possibly correspond with. To say that a sentence corresponds or can correspond with reality seems to be a category mistake. If facts about the world are merely other sentences, which seems right, then the notion of truth becomes empty, because it is a relationship involving the world itself that we want our semantic notions to entail.
I favor a pragmatic picture of truth, which appears to solve the problem of correspondence. Truth signifies that “ideas, which themselves are but parts of our experience, become true just in so far as they help us to get into satisfactory relation with other parts of our experience” (James, p. 197). There is a correspondence relation, but it is a correspondence between parts of ourselves. It is a correspondence of energies and feelings, and language helps us to manipulate those energies and feelings in complicated ways. Simplistically put, language gives us substantive sense of objects; feelings make relations between objects known. “Truth” applies to moments in which our sense (our experience) of some objects and relations (objects and feelings) become satisfactorily related to our sense (our experience) of other objects and relations (objects and feelings). Overall, on the pragmatic picture, experience is a species of reality. It is a species of reality by means of which reality seeks contact with itself through language and feeling.
Notwithstanding the tremendous physical complexity of what truth in human experience on the pragmatic picture seems to entail, mediated as it is by that most complex of all physical mechanisms ever devised, biological or otherwise, we all sense rather directly that truth, and only truth, has the power to set human beings free. This sense derives from the fact that freedom by way of truth is freedom by way of society. Society is the great human invention that simultaneously makes possible our comfort and progress and that suppresses our inclinations and higher impulses. Because it functions both as a safety net and as a goad to individual excellence and autonomy, we sense that perfect social freedom is the greatest freedom possible for the human being. We sense that if we are able to hear and say true things, and only true things, as a way of life, society will bend to us and order itself in the way that Heaven has decreed. “The unstable estimates of men crowd to him whose mind is filled with a truth, as the heaped waves of the Atlantic follow the moon.”
Imagine Robinson Crusoe on his island, a natural man able to supply all of his essential needs with no very great difficulty, but without any society. Compare him to a socialized man, who despite the presence of some irksome but seemingly unavoidable obligations, nevertheless manages to maintain for himself a modicum of leisure and the means to enjoy it. Which man is more free? I believe it is the socialized man, because there is a larger number of intentional actions it is possible for him to take. He, and not the natural man, is able to draw on the concerted labor and capital of many other such men, and thereby to handle a very large number of domesticated objects and situations that simply could not arise otherwise. Now imagine that socialized man in a condition of ideally constituted language use. I believe his freedom would undergo a still greater augmentation, because under those conditions, at least during the interlude in which an ideal employment of language remained tenable, the vicissitudes of dealing with one’s compatriots, who otherwise may and usually do feel compelled to act out of self-defense and self-interest, would for a time be transcended, and the tenor of social relationships characterized by equanimity and open-mindedness.
First order logic represents a central and essential step toward the ideal of truth-functional conversation, because it arrays the truth values of two-valued connectives in an unquestionable way. For example, if someone speaks a compound sentence joined by “and,” and if the two conjuncts are each true, the resulting speech act has the force of truth. A certain power of assent accompanies the utterance of
(1) Snow is white and grass is green
regardless of the context that calls forth its utterance. Or does it? Surely there are contexts in which uttering (1) would appear irrelevant, even anti-social, for example if I had said “Give me the timer code or this thing will explode” in the sincere expectation of a true response. Similarly, suppose I say to you,
(2) Either my father was Sten G. Flygt or my father was George W. Campbell.
On logical grounds, I have said something true, but conversationally we would be inclined to dismiss such an utterance as misleading, perhaps even not true. To get truth-functional conversation and perhaps even ideal social relationships we are going to need to augment logic.
H.P. Grice (1975) has drawn attention to divergences between the meaning of formal devices and formulas derivable from some simple such formulas, e.g. Tarski-style meta-languages, and natural languages, where a full analysis of meaning and intension is not always possible. Grice suggests the consequence of this divergence, if it cannot be resolved, is that natural language will remain forever inadequate to “serve the needs of science.” Science, it will be granted, represents a sort of ideal in social processes, because in science an advance on truth and social consensus is plainly discernable. Grice then makes the remarkable assertion that what divergences between natural and formal languages appear to exist can probably be eliminated by giving adequate attention to the nature and importance of the conditions governing conversation. If Grice’s assertion is correct, and if my initial outline (see below; see also Flygt, 2006) of conversation theory manages to hold up, a “reformation of natural language” in favor of formalism, formality and science may be inevitable.
Grice’s approach is to define a term of art, conversational implicature, and to show how its use gives rise to meanings probably impossible to capture in two-valued logic, unless that logic can be applied to the form of conversation per se. What Grice is at pains to show is how meanings derived from conversational implicature are in fact well-formed. If Grice’s analysis is exhaustive and accurate, it would follow that if conversation could be conducted reliably as a scientific instrument or a scientific process, scientific results, probably having to do with the nature of human consciousness, would follow. If human consciousness is the real target of future science, and current advances in neuroimaging and neuroscience suggest that it is, then a reformation of language in favor of such science would, if undertaken with due deliberation and care, appear to portend some sort of social evolution.
My own approach is to define pure conversation as a situation in which everyone present is interested in maintaining a certain connection of intuitive content and emotional affect for the duration of the interlude. Following James, Dewey and the pragmatist school, intuitive content is what is present in consciousness or experience in the moment, with no postulate of a Cartesian knower or of the transcendental reality of objects. It is purely and simply what is salient in the moment to experience as such. That content is generally termed intentional, and as such is propositional. Notationally, intentional states are represented on the form,
(3) F(p)
where F is the mode or state of directedness, (e.g. belief, desire, perception, intention) and p is the proposition that the state is about. Thus,
(4) intend(I pick up this cup)
captures the immediate experience, including the state of consciousness and the propositional content, of my trying to pick up this cup.
Let’s take it as given that people in the circumstance of conversation affect each other in specific ways from moment to moment, that they cannot avoid doing so, and that it is of general interest (scientific, philosophical, social, political) to determine the structure of this causal nexus. With the foregoing notation together with sufficient measuring devices in place during conversation, including for example the digital clock, videotape, and galvanic skin response, it should be possible to represent the experience of each participant more or less exactly for the duration of a conversation, and thereby to explore what really happens in social experience generally. Quite possibly, these techniques can be extended to descriptions and perhaps even measures of emotions and their underlying brain states, which are being shown today to have a clear motivational and causal role in moral judgment and behavior.
Of course without an adequate overall theory of conversation, and its implementation on experimental lines, no social inferences about such causes and effects can be drawn. What exactly is an ideal conversation? What causal role does language play under such conditions? Which factors of human experience can and should be maximized under these conditions, and which minimized? Today it is believed that 46% of all Americans will suffer a mental disorder within their lifetime. If it can be shown that mental health can be improved through social and cultural interventions using well-formed conversation, conversation theory should probably be applied as public health policy. Similarly, if all real-world conversation can be shown plausibly to represent an ideal form, albeit typically with significant degeneracy, then the ideal form should become a matter of pressing interest, political and cultural, to the leading classes of society.
Conversation then is a sequence of conversational moments:
(5) Conversation↔CM1, CM2…CMn
where CM is a conversational moment with a certain causal structure to be made explicit.
In pure conversation, a conversational moment has the feature that everybody participating in the conversation is satisfied that they exactly understand, recognize or otherwise represent the meaning of the current content. By hypothesis, under this condition, everyone has the same content or object in mind. Thus,
(6) CM↔All x, All y(xRy^part(x))
where “R” is a two-place predicate meaning represents or possesses and “part” means participates. In other words, a conversational moment is a moment in which for anyone participating in the conversation, that person represents, possesses or otherwise intuits y, the current conversational content. Under such conditions, there is no disjunction between what someone actually experiences and what he (she) represents himself (herself) as experiencing.
As a practical matter, it is usually an intuitive, artistic affair to determine universal representation. Generally such determination is a question of the mood or vibratory mode of the group, and is usually left to tacit acceptance. By such implicit means together with certain explicit agreements and procedures, such moods or modes can be strengthened to the point of preeminence. Under such conditions, not only can states of consciousness be altered constructively, but the entire social ontology can be inverted, resulting in groups of individuals behaving as a social organism, analogous to a flock of birds or a school of fish behaving in a psychic accord.
Now for the theory of pure conversation, conversational contents possess a kind of necessity. When actually expressed in the way they are, they rule out certain contents, heretofore possible, and open up others as candidates for being the next one. Because universal representation can be made as precise as desired, the limiting case is that in which the subsequent content is uniquely determined. Whether this determination is purely logical, or whether physical and intensional factors also play a role, I leave undecided here. Let us suppose, for the sake of making a guess,
(7) CMm→CMn
Now y, the content of the current conversational moment is generally speaking an intension. It is a function from some scenario imagined by a speaker to the extension of that scenario. Ordinary conversation generally lacks the patience to establish with certainty what the intension of any particular utterance really is, but pure conversation does so routinely by means of the technique of conversational reproduction. Any content, if it fails to satisfy perfectly on a universal basis, is reproduced iteratively until everyone is able to assert satisfaction with it. Only then is it considered proper to move to the next content. Thus we can say,
(8) y=Int(p)
where p is the proposition represented by the current utterance, and
(9) Int(p)=({D,I,R},F)
where D is the domain of objects the speaker could be representing, I is the index or scenario the speaker has in mind, R is the accessibility relation between what the speaker has in mind and what the others have in mind and F is the satisfaction value that an interpreter derives from divining the intension of the utterance.
Conclusion
By defining semantic interpretation in terms of the satisfaction of universalized intentional states, conversation theory shifts the locus of truth conditions from actual things and situations in the world to the experience of actual things and situations in the world, including physiological indicators of such experience. According to conversation theory, this species of a Jamesian/Deweyian radical empiricism, together with the Kantian specification of moral law will be sufficient to make objective determinations about human consciousness under conditions of meaningful conversation.
Reference:
Bennett, Jonathan. A Philosophical Guide to Conditionals.New York: Oxford, 2003.
Collin, Finn and Finn Guldmann. “Intensional Semantics.” In Meaning, Use and Truth. Burlington, VT: Ashgate, 2005.
Emerson, Ralph Waldo. “The American Scholar.” In Classical American Philosophy, John J. Stuhr, ed. New York: Oxford University Press, 2000, pp. 17-26.
Flygt, Carl H. Conversation – A New Theory of Language. Great Barrington, MA: Lindisfarne Books, 2006.
Flygt, Carl H. “How to do Conversation Research.” Available at www.consciousconversation.com.
Grice, H. P. “Logic and Conversation.” In A.P. Martinich, ed. The Philosophy of Language. pp. 156-167.
James, William. “A World of Pure Experience.” In Classical American Philosophy, John J. Stuhr, ed. New York: Oxford University Press, 2000, pp. 181-193.
James, William. “The Stream of Thought.” In Classical American Philosophy, John J. Stuhr, ed. New York: Oxford University Press, 2000, pp. 161-181.
Knobe, Joshua and Jesse Prinz. (forthcoming). “Intuitions about Consciousness: Experimental Studies.” Phenomenology and Cognitive Science. Available at http://www.unc.edu/%7Eknobe/consciousness.pdf.
Peirce, Charles Sanders. “What Pragmatism Is.” In Classical American Philosophy, John J. Stuhr, ed. New York: Oxford University Press, 2000, pp. 105-115.
Prinz, Jesse. “The Emotional Basis of Moral Judgments.” Philosophical Explorations, Vol 9, #1, March 2006.
Searle, John R. Intentionality: An Essay in the Philosophy of Mind. New York: Cambridge University Press, 2003.
Weiss, Rick. “U.S. Leads in Mental Illness, Lags in Treatment.” The Washington Post, June 7, 2005, p. AO3.
The idea that facts are propositions, not sentences, leads to other difficulties, because a clear definition of “proposition” has thus far eluded philosophers. Whether they are “strong Russellian” (actual realities), “weak Russellian” (realities with subjective attributes) or “Fregean” (intersubjective agreements) is an analysis I cannot pursue here.
James pp. 170-171 can be consulted here.
Presumably under such social conditions it would not be necessary to conceal a truth about her condition from a terminal cancer patient, who will have conducted her entire life in such a way as to accept the reality of death, the hazards of poor lifestyle choices, the vicissitudes of genetic endowment, the limits of current medical knowledge and what is scientifically understood about what happens to human consciousness in death and afterward.
Here again the pragmatic view of inquiry is preferred. Charles Sanders Peirce writes, “That which any true proposition asserts is real, in the sense of being what it is regardless of what you or I may think about it…. Accordingly, the pragmatist does not make the summum bonum to consist in action, but makes it to consist in that process of evolution whereby the existent comes more and more to embody those generals which were just now said to be destined…” (Stuhr, p. 114).
Jonathan Bennett has argued coherently that the Gricean analysis, although “shiningly true,” is powerless to handle the conversational truth value, and hence the meaning, of conditionals (sentences of the form “if…then…”). The only way to handle such sentences in conversational context, is with probabilities. My theory of conversation is indifferent to this fact, and is in a position to welcome it as a way of leaving the door open to conversational art. Moreover, on my theory, many probabilistic semantic values can be detected physiologically, as functions of a kind of group consciousness or group mind that can arise in well-formed conversation.
The standard reference on intentionality is Searle, 2003.
Prinz and Knobe represent contemporary work in experimental philosophy and experimental sentimentalism.