Professional Documents
Culture Documents
Sharing Realities 05 The Construction of Understanding
Sharing Realities 05 The Construction of Understanding
Richard Ostrofsky
Copyright © Richard Ostrofsky, 2000
ISBN:1-894537-00-9
It will now be clear why classical logic is largely irrelevant to the problem
of argument among people who see the world in significantly different
ways. As the form of valid inference – displaying the formal patterns in
our use of words – logic has long been conceived as a kind of skeleton for
language, showing why two statements may be equivalent, or why one
statement may imply another1. In contexts where the meanings of words
are fairly constant and reliable, logic can still play this role. On the other
hand, in most real-life arguments where words mean what their users
want them to mean, and where such preferences are inseparable from the
substantive issues in dispute, the attempt to agree on definitions merely
underlines the contentious instability of language itself. The notion of
logical proof is empty whenever terms are too vague, too abstract or too
value-laden to bear precise, mutually acceptable definitions. The
predicament can be cast in a rough syllogism:
Possibilities of reason depend on the existence of logical
norms that distinguish legitimate from illegitimate claims.
In open discourse, where passions run high, and word
meanings are mutable and self-serving, such norms are not
supplied by classical logic.
Therefore:
We must either resign ourselves to the absence of reason in
all discourses that engage our passions, or else re-think the
foundations of logic on broader lines that allow for the
1 Classical logic is concerned with the inferential meanings of words and the tautologous
equivalence of word strings – with the many ways there can be of saying essentially the same
thing. If it is true that “All men are mortal”, then the mortality of any individual is part of the
meaning of his humanity. When we are told that someone is human, we have also been told
implicitly that death is part of his lot. Classical logic lets us infer his necessary death from the fact
of his being human.
6. Logic of Conversation
mutability and contentiousness of language.
So let us roll up our sleeves and begin.
6.1 Rethinking Logic
Do I contradict myself? Very well then, I contradict myself. I am large. I contain
multitudes.
Walt Whitman
The task is to understand how intelligibility and reason are possible (to the
very considerable extent we know them to be so) in domains where
viewpoint and interpretation have ultimate sovereignty – i.e. in irreducibly
polyphonic domains where no universal “truth” can marginalize dissent.
What is needed is a kind of hermeneutic logic (as we might call it) – a
logic of interpretation and understanding – to clarify the “rules of
engagement” among divergent knowledge cultures, divergent systems of
understanding. Such a logic will take divergent values and interpretations
in its stride; and it will deal justly with “hard” and “soft” disciplines alike:
with the physicist’s interpretation of experimental data, with the historian’s
interpretation of his documents, with the critic’s interpretation of a poem.
This hermeneutic logic must accept a task that classical logic sets
to one side, and that most post-modern thinkers concerned with the
problem also preferred to ignore – either in a spirit of mischief, or from
political agendas that deranged their judgment. The task is to do
philosophical justice to our intuition of a single material reality that we all
inhabit, but also to our “subjective” understandings: to Plato and to the
Sophists; to the world of science with its powerful claims to universality;
and to the diverse worlds of human cultures and interests. The rules of
logic should be canonical in some sense – not mere cultural conventions,
but necessary features of any conversation that can make collective, public
sense from the articulated viewpoints, values and perceptions of its
participants. Relativists hold that no such trans-cultural logic can exist –
that all “universal principles of conversation” are just conventions of a
culture. To the contrary, I suspect that the requirements for intelligible
discourse are the same for everyone, right across the universe, for all
cultures, all species, all ecologies – inherent in the nature of
conversational relationship.
Conversation may or may not succeed in construing an “objective”
view from nowhere,2 that would make all personal and cultural
understandings reciprocally intelligible, relative to some universal truth.
No such abstraction may be available. Even if it is, it may not attract a
consensus. All that reasoned conversation can be sure of constructing will
be a tensegrity of coherent argument – a “meeting of minds” – along the
lines of Section 3.2. Accordingly, the logic we are after will amount to a
theory of polyphonic understanding. It is a fair conjecture that such a
theory will apply to all minds everywhere, because the conditions of
2 Thomas Nagel’s term for the construed, “objective” perceptions available from an artificial
“public” viewpoint that belongs to everyone (to everyone who want to share in it), but not to
anyone in particular.
6. Logic of Conversation
intelligibility must be prior to language and culture, and to the evolution or
design of sensory organs and nervous systems. There must be common
requirements of perception, recognition and classification that shape both
biological and cultural evolution, diverse as the outcomes of these
processes are known to be.
A hermeneutic logic that can work with polyphonic truth should
reduce to classical logic in the special case that meanings are stable, and
common assumptions agreed. But it will generalize and extend this logic
in several ways. To begin with, it is concerned – as classical logic is not –
with the complex inter-relationship of theory, value and practice. Instead
of focussing solely on descriptive, value-free, true-or-false statements,
hermeneutic logic finds the source of the world’s intelligibility in the
three-fold interdependence of praxis, values and “beliefs” (i.e. cognitive
commitments), as discussed in the last chapter. While classical logic, to
uphold its conception of factual truth, must insist on a rigorous separation
of these elements, a hermeneutic logic bases itself on their
interdependence. It is a logic of values, beliefs and intentions all bound up
together.
Hermeneutic logic need not and does not deny the existence of a
real world, and is consistent with the most steadfast realism. On the other
hand, it in no way precludes commitment to some particular narrative
about that world, but rather encourages a diversity of such commitments.
However, for purposes of extra-mural conversation with interlocutors who
do not share the same preferences of understanding, it suspends all
personal commitments, and takes a polyphonic, “ecumenical” approach,
remaining neutral between divergent stories in play. Its concern is with
mutual intelligibility and integrity, rather than absolute truth. Human
truth, where it exists at all, is always construed and consensual. More
commonly, it is a structure of argument. God’s-eye Truth, by definition,
lies outside of human conversation.
What we would call “hermeneutic logic” seeks to collect and
articulate the necessary working assumptions for mutual intelligibility and
knowledge-value in a cross-cultural conversation. Of these working
assumptions, the following seem fundamental:
Assumption: Each protagonist is ineluctably part of a common
situation, but lacks a god’s-eye view of it. In general, the best available
answer to a question is not a “true” statement, but a structure of reasoned
argument among competing views.
Comment: Even if I believe I have been vouchsafed a unique
revelation, it is necessary to lay this commitment to one side in dealing
with epistemologically challenged persons, less privileged than myself.
Assumption: The “beliefs” (cognitive commitments) a person comes
to hold, along with the experiences that derive from and justify these, will
have their roots in his or her life experiences, interests and values.
Comment: This relativizing assumption is necessary if I am to give
6. Logic of Conversation
a cognitive adversary the benefit of the doubt, and treat with him as
perceiving and dealing in good faith. Without it, I am compelled to think
anyone who disagrees with my own position either knave or fool. With it,
I can attempt to understand how my interlocutor’s commitments are
grounded in his life experiences from much the same motives (and with
comparable justification) as my commitments are grounded in my own
experiences.
Hermeneutic logic is consistent with the “perennial
philosophy” of contemplatives and mystics: Value (that
is to say, love in all its manifold forms) is not only the
source (or “motivator”) of our actions, but of cognition
as well. To know anything at all we must become
involved with that thing, contemplate (i.e. “spend time”
with) it, and generally give ourselves up to it in order to
receive its teachings. Doing so, what we come to know
will depend on the nature of the relationship we have
created.
Assumption: Values (to view cognition from the other end) derive
from the kind of creature one is, and the kind of life one leads.
Comment: Our patterns of valuing are certainly personal, but they
are not mere whims. Very like our so-called “beliefs of fact,” our values
and priorities represent commitments. They are formulated, accepted, and
finally habituated, so as to render experience intelligible. They represent
heuristic principles, or “rules of thumb,” to help us in making sense of the
world by orienting us within it – by pointing out to us which way is “up.”
This assumption is needed to accommodate values, especially the values
of others, within the scope of reason. Without it, reason becomes, as
Hume said, “the slave of the passions” – a means for gratifying tastes that
are themselves beyond understanding, justification and ethical or aesthetic
judgment. Thus, it becomes impossible to reason about values – to
transmute competing interests into joint concerns.
Originally, and in its etymological meaning, logic is the philosophy of the
Logos – the creative, magically powerful Word. As such, it overlaps with
the descriptive sciences of semiotics and linguistics to embrace the whole
domain of articulate meaning and understanding. This domain, we’ve
seen, can no longer be conceived as absolute or value-free. We can no
longer imagine that there is a single correct, or most nearly correct,
understanding of everything, toward which all right thinking persons must
converge – following sound arguments wherever they lead. (Well, we can
imagine what we please, and may even find agreement in some areas, but
we will find it practically impossible to convince anyone of anything,
unless he is already disposed to be convinced.) We must accept that
people, and whatever other sentient beings we encounter, can legitimately
understand their worlds in very different ways – according to their “lights”
as we say. That is, they will understand according to their life histories
6. Logic of Conversation
and experiences, their physiologies, their cultural and personal habits, their
interests, their purposes, and the moods they happen to be in. In the last
analysis, there can be no common truth or meaning that does not finally
depend on a structure of potentially divergent understandings.
In sketching the outlines of a “hermeneutic” logic that could avail
under these conditions, we must begin by lowering our expectations.
Hermeneutic logic (unlike the classical variety) cannot give mechanical
procedures for generating valid conclusions from agreed pre-suppositions.
As we’ve suggested, when meanings are unstable, the jump to mutual
intelligibility (never mind agreement) is a sympathy of resonances, not the
outcome of a rule-bound process. What hermeneutic logic might provide
is a theory of understanding – a kind of meta-psychology about the
grounds and mutual relevance of diverging cognitive commitments. Its
aims might be:
1) to explain how understandings are formed in conversation – partly
from direct suggestions of “the things themselves”, and partly in the flux
of suggestions from the conversation’s participants;
2) to elucidate the grounds and methods of reasoned argument: the
process of systematic comparison and criticism of divergent
understandings.
In short, a hermeneutic logic must tackle the central difficulties of any
serious relativism: to explain how diverse understandings are constrained,
and brought into mutual relationship by their exposure to feedback and
correction from a common world; to explain how disparate
understandings may be shared – at least to the extent of mutual
intelligibility; to explain how collaborative, more-or-less common
understandings can be developed when interlocutors are neither alike in
their interests and backgrounds, nor so unequal in power that one party
can just impose its preferred understandings on all the others. This last
point is the cardinal weakness of relativist thinking at present: namely,
that it can scarcely conceive of any social relationship or cultural
transmission except in terms of power.
Now, it would be correct to say that power issues are always present in
human relationships, and that the balance of power among interlocutors is
usually somewhat – or extremely – unequal. And Nietzsche’s point – that
understanding, by its very nature, involves some assertion of cognitive
power – is also correct. But it does not follow that power is the only, or
even the essential feature of all cognitive and social transactions, or that
every case of cultural influence is simply an imposition of the stronger
party’s cognitive structures upon the weaker. Thinking in these terms only
distorts our social priorities, and distracts us from really brutal and flagrant
abuses of power, which would keep activists busy enough.
In real life, the distribution of power is rarely so unequal as to allow a
dominant faction to impose its concepts and values in any straightforward
way. The spiritual conversion of a conquered people is more difficult than
6. Logic of Conversation
outright extermination, and even that has rarely been feasible. History
offers no end of examples of the conquest and absorption of one culture by
another, but scarcely any that have been total, entirely one-sided, and free
of active collaboration and selection by the weaker party. And the most
intangible items of cultural inventory – its concepts and its values – are
always the last to go. For all the atrocities inflicted by Euro-American
peoples on aboriginal populations and on descendants of imported African
slaves, and for all the disproportions of power involved, the cultural war
against these peoples, from a white supremacist perspective, has been a
failure. Their spiritual subjugation and disappearance from the cultural
landscape was never achieved, and now seems more unlikely than ever.
The same point is even more obvious concerning the alleged cultural
domination of men over women. Notwithstanding some very real social
injustices, a case could be made that males always and everywhere find
themselves much more thoroughly co-opted by the reproductive, “uterine”
imperatives and time-tables of their women-folk than women ever are by
the “phallocentric” agendas of their men. I doubt there is a culture in the
world so thoroughly “patriarchal” that experienced husbands are not afraid
of their wives’ ability to make their lives miserable when they themselves
are unhappy about something. Almost always, it will be simple-minded to
understand a relationship in terms of power alone, and even more so to
analyse cognitive influence in terms of power relationships, to the
exclusion of all other factors.
We do not wish to overlook the significance of power-relationships in
the cognitive sphere as elsewhere. But if we hope to understand the
sharing of cognitive structures in conversation, we have to think beyond
the Nietzschean language of imposed interpretations. We have to think in
terms of destabilisation and the evolution of differing understandings,
along the lines of Section 5.2. In short, we need something like a
hermeneutic logic, not just a sociology of ideas or a political science.
4 Nietzsche’s point that all cognition is ultimately a matter of interpretation must itself be
understood in a hermeneutic context – since it is itself a rather fancy interpretation that cannot be
true or false in the traditional sense. In every particular case, it is finally a matter of philosophical
commitment whether you regard a way of understanding as “invented” by some creative mind, or
forced upon that mind by evidence and argument, (and hence “discovered”). Moreover, the
meaning of this point only emerges in the context of some further purpose. Used by one writer to
argue for a complete relativism, it can be used by another to argue some version of pragmatism, or
by a third – as in the present essay – as the point of departure for a reconstructed, conversational
concept of reason.
6. Logic of Conversation
5
interpretation. As such, it lacks epistemic value, except insofar as it
comes from some trusted source who backs it with his reputation. As
suggestion, the statement can be no more reliable than the author who
vouches for it. The epistemic value of the suggestion, we might say,
derives from the integrity of its author’s witness to that whereof he
speaks. This is only common sense, of course, yet it is often overlooked.
To some degree, there is a willing suspension of disbelief for information
that we find in our data banks, or read in our media. We tend to accept it
without question, because it would involve so much work to question in
any serious way. We tend to accept many written statements without
looking very hard at who wrote them, for which purpose, and with what
justification.
The statement’s negation is shifted correspondingly. In classical
logic, no statement can be true and false at the same time. That would be
a contradiction; and to demonstrate a contradiction is to catch the speaker
in absurdity. In hermeneutic logic, a statement and its negation are not
mutually exclusive. To the extent that both represent tenable, serious
positions (as indeed is often the case) they are complementary poles of a
tensegrity. As such, neither is complete unto itself. It is the tensegrity of
commitments that we must re-construct when we want a complete story
about some issue.
6.4 Inference
In classical logic, inference means the drawing of valid conclusions from
accepted premises. In hermeneutic logic, we do not so much draw
conclusions as jump to understandings. This jump may or may not be
appropriate and useful, but it is not valid in the formal, classical sense.
Hume’s recognition of this point in connection with the notion of cause-
and-effect marks the beginning of the end of traditional philosophy:
Neither induction nor deduction afford compelling justification for the
inferences of science and everyday life. It does not follow that we are
wrong to draw these inferences – only that they must be deployed with
caution and justified on a different basis.
Consider the following example: From behind, we see a person whose
build, dress, hair style and walk are those of someone we know. We
recognize her; we call her name; she turns. We were expecting a certain
face, and then we see it. Or perhaps we are surprised to see a stranger.
Or consider the following excerpt from an imaginary paper on
Shakespeare’s Hamlet:
The play’s frequent references to doubt, and the stark contrast between
its indecisive protagonist and its blustering minor characters – Laertes,
Fortinbras, the gravedigger – reveal its focus on what we might call
personal epistemology: the construction and validation of personal
6 The “flash” was that he could measure the volume, and thence the density of an irregularly shaped
object – the king’s crown – by measuring the rise in the water level when the crown was immersed
in a tub of known surface area.
6. Logic of Conversation
a complicated book – exemplify hermeneutic reasoning at its richest. In
any prolonged cognitive effort, the hermeneutic circle will be traversed
many times over. Time and again, a present state of generic understanding
will be tested against specific problems, break its teeth on these, and be
thrown back in doubt and confusion. Time and again, if the student is
patient and honest in his bewilderment, a deeper, more flexible
understanding returns at last. This repeated movement is not a circle, but
an evolutionary spiral. It moves forward – albeit not in any predestined or
determinate way – and can itself be seen as a chain of reasoning, with each
successive state of understanding inferred from the one before:
1) From experience and discourse, a preliminary understanding develops.
2) In attempts to apply and articulate this understanding, further
experience is gained, and further discourse takes place.
3) New experience, in turn, destabilizes understanding, leading to periods
of confusion, (actually periods of cognitive growth), from which some
richer understanding emerges.
4) Recursively, this new understanding enables further experience and
discourse, enlarging on what had been possible before.
For the adepts of any discipline, this spiral never reaches a final state of
perfect mastery; but it does find interim states of relative stability, where
big surprises are rare. Interestingly, at this limit, understanding is usually
polyphonic. No simple structure of flat Aristotelian truths could represent
the understanding of a creative theoretical physicist or a software engineer,
let alone that of a great musician, or a master potter. The craftsman knows
too much: He knows the exceptions as well as the rules. He knows that
the mysteries of his craft can be articulated in any number of different,
superficially conflicting ways, which are all of them right, and all wrong.
He knows how to suit the explanation to the pupil, and the tool to the task.
Even his values are flexible, and tend to re-shape and adapt themselves to
his immediate situation – to the unique occasion of each performance.
Then too, the understanding of adepts in any field tend to be subtly or
even grossly divergent – depending on the backgrounds and baggage they
arrived with, the teachers they studied with, the projects and pupils they
confronted. In some ways, the masters resemble each other much less
after thirty years experience than they did as novices. Managers, in search
of predictable, competent performance, do not want to hear this; and it
terrifies beginners. But it is true none the less: Three years of training
(give or take) will teach a pupil the basic ideas and techniques of his trade.
Thirty years of good practice will teach him how much he doesn’t know,
how much (and how little) freedom he has, and how to find self-
expression in its discipline.
What does all this learning and experience have to do with Truth and
Knowledge, as philosophers traditionally conceived them? This question
goes to the core of the philosophy of knowledge; and it is one of the great
6. Logic of Conversation
questions.
The master’s understanding will not be absolute or universal. Masters
will often disagree, and there is no reason to think their understandings
must ultimately converge on all points. There are many fields of
knowledge where lore and technical skill have steadily accumulated from
the beginning, but which are still divided by the same basic issues that
vexed their founders. And it is both likely and desirable that at least some
of the current argument in any field – its current polyphony – will be
internalized by every competent practitioner, so that he himself has no
rigid opinion on the matter. Even then, when we have accepted that Truth
is polyphonic in nature, there is still no guaranty that the understanding of
any individual, or of the whole conversation, will come to map the diverse
possibilities of “things as they are” in any reliable way. What can be said,
I think, is that the spiral of experiencing and theorizing represents a form
of inference yielding results whose long-term knowledge-value will
depend on qualities of integrity to be discussed in the next chapter. We
could say simply that the “truth” of these results will depend on a certain
quality (I call it integrity) of the experience that produced them, except
that we have no independent way of knowing what truth means.
To summarize then: When we speak of hermeneutic inference we
have three patterns in mind:
∙ instantiative inference moves from generic skills and understandings
to specific moves and expectations in a particular situation.
∙ constructive inference abstracts and generalizes from detailed
experience toward a schematic grasp thereof; and
∙ recursive inference iteratively cycles through constructive and
instantiative inference, moving toward successively richer, more
comprehensive modes of generic understanding, and better grasp of
the cases that arise.
Understanding, even circumscribed or deluded understanding, confers
what we have called existential sureness: the confidence that a healthy
human animal needs to feel it is on top of things, and able to cope with its
situation. Such confidence is not surrendered lightly. Usually, an
understanding must suffer intense and long-lasting embarrassment before
it is seriously questioned. No wonder, then, that some understandings
prove so durable: Constructed so as to be immune to any possibility of
embarrassment, their adherents would sooner die than forfeit the sureness
they afford!
And yet, even the most hidebound understanding may become
unstable, and be replaced finally by some new way of thinking – when the
older way is repeatedly embarrassed by experience and discourse collected
in its own light. In this sense, a cognitive strategy may prove inconsistent
with itself, and may de-stabilize itself. At this point, some new
understanding must be inferred from the experience and discourse that the
older one made possible.
6. Logic of Conversation
6.5 The Logic of Conversation
What I would call the logic of conversation is something larger than this
recursive, hermeneutic logic. Mathematical chaos theory and the theory of
self-organizing systems suggest that all recurring interactions must
eventually settle down into stable patterns of inter-relationship. All
systems, from atoms to galaxies, seem to have formed according to this
same evolutionary logic of emergent structure. So we conjecture, at any
rate. In particular, what we are calling the conversation – the exchange of
suggestions by human suggers – settles down into the stable patterns of
social life. Artifacts of culture, both cognitive and material, evolve to
facilitate and stabilize these patterns; and, from one perspective, the logic
we have in mind is just the theory of this process, an aspect of sociology in
its widest sense. Fully developed, however, the logic of conversation
would be something more than a descriptive sociology, but tantamount to
a grammar of intelligible social relationship – explicating pre-requisites
for meaningful intimacy, play, work, trade, conflict, government, and all
else that we do. The problem is to account for the evolution of intelligible,
orderly, useful communication, and to explicate the pre-requisites of
meaning and order when it is accepted that cognition is an active process,
that interpretations are imposed by individual suggers to “get a grip” on
their private worlds, and that common understandings are the outcomes of
a evolutionary and conversational process.
What we can now surmise about this logic of conversation is most
conveniently discussed under four headings generated by a pair of natural
distinctions that cut across each other: On one hand, is the distinction
between process and structure: between the on-going or diachronic aspect
of conversation, and its static or synchronic aspect at some point in time.
On the other, there is the distinction between global and local: those
features or properties that pertain to the whole conversation and those that
pertain only to a small region of it – to a given participant, say, or a given
transaction. These distinctions divide our field into four quadrants, to be
discussed in turn:
7 For an analogy here, consider the quantized energy levels in the electron shell of an atom. With
6. Logic of Conversation
conversational relationship, we can assume a degree of self-consistency
sufficient to allow its recognition as such. Except for small variances and
cumulative changes that will transform or terminate it in the long run, we
can assume that any on-going relationship must maintain (or cyclically
restore) its own pre-conditions if it is to endure. Accordingly, we can
think of any relationship as an eigenstate8 of conversation – an almost
self-reproducing pattern in the flux of activity and suggestion. Each type
of relationship–such as cooperation or trade or conflict – can be thought
of as a semi-stable pattern of conversation. These concepts themselves,
through the suggestive influence of their very names, often suffice to
establish or sustain the relationship in question. Thus, by a logic of self-
confirming expectation, to be married means simply to think of one’s self
as married, to be at war means to think one’s self at war – and to act
accordingly.
As a relationship becomes established, its patterns of activity and
suggestion will be assimilated to and stabilized by their own past history,
and by the experience of the interlocutors in previous or vicarious
relationships. Youngsters expect from their first sexual relationships what
they have heard from their friends, read in books, or seen at the movies.
Later, with their children, they develop a style of parenting based on their
experience with their own parents. Something more than emulation is
happening in such cases. They are further examples of the assimilation of
present experience and activity to that of the past, to that of significant
others, and to the patterns of a culture.
When suggers first encounter each other, a pattern of mutual
exploration will commence, aimed at determining the possibilities of
relationship. This pattern may be more or less cooperative, more or less
adversarial; with attempts at withdrawal or concealment on one side or
the other, or both, depending on what the parties fear and hope from each
other. Or, the exploration may be more or less flirtatious, with the
advances and withdrawals meant playfully or symbolically, and not to be
taken seriously. Paradigm cases of exploratory relationship might be a
first date, or the opening sessions of a negotiation – or two dogs in the
park, sniffing each other up.
Even at this early stage, certain necessities begin to make themselves
felt. For one thing, the ebb and flow of suggestion among interlocutors
will tend to fall into a definite rhythm. This may be driven by a natural
respect to the atomic nucleus, an electron can only exist at certain energy levels, namely those
corresponding to a wave-length that fits a whole number of times into the corresponding shell
circumference. If the wave-length does not divide evenly into the shell circumference, the
electron (in its wave aspect) will tend to cancel itself out, and cannot be found there. In quite a
similar way, observable patterns of living and relationship – family or work patterns, for example
– must also tend to reproduce themselves from one day and year to the next, since otherwise they
would not be patterns but one-time events.
8 This term derives from matrix algebra and quantum mechanics, where it indicates a state that
remains unchanged under a given mapping of the state space into itself, as in a system’s change
from time t to time t+△.
6. Logic of Conversation
source, e.g. the daily and yearly planetary rhythms, or by some party
external to the relationship as with a prospective couple meeting under a
chaperon’s supervision. Or it may emerge from the dynamic of the
relationship itself, as in physical combat or sex. The need for temporal
patterning is one basic feature of the logic of relationship. That for
approximate self-consistency from one cycle to the next is another.
Local Structure: The Logic of Context and Signification
The logic of the second quadrant, that of local structure, is concerned with
micro-arrangements for conversation among the parties to a relationship.
Just by holding each other’s attention over a period of time, interlocutors
develop a common context, both institutional and purely cognitive, setting
terms of reference for further conversation. This context does not yet
amount to a common culture, but it will make culture possible. Of crucial
importance in the present discussion, it will enable conventions of
signification to evolve in a flux of suggestions that are not yet true signs.
Words and symbols belong to the realm of language; and, like other
artifacts of culture, they evolve as global structures of conversation, as
will be discussed in the next section. But, before there can be signs, there
must be associations (as they are called) – glimmerings of connection
between two separately recognized events, which in time may (or may
not) create a stable relation between a signifier and signified. The simplest
and earliest associations are established locally, through a linking of
percepts construed as belonging to a common context. Thus, in time, for
Pavlov’s dog, the bell is associated with food, and eventually becomes a
sign of food. Nothing about the bell intrinsically suggests food, but the
bell comes to “signify” food for the dog through their repeated association
in a common gestalt (whole context). It is the logic of context formation
and association, the logic by which sensory events are able to suggest and
signify, that we need to explore in this quadrant of local structure.
The logical chain runs from stimulus to suggestion, from suggestion to
association, from association to sign. A stimulus, such as an infant might
experience in a noisy room is not yet a suggestion like the gnawing in its
belly that prompts it to cry. Contact with mother’s body is an association,
not yet a sign, that the gnawing sensation will soon be relieved. The word
and concept of “mother” is not local to the immediate conversation
between the infant and the woman, but global to the language, the culture
and the whole social conversation in which their relationship takes place.
As such, it belongs to the next quadrant, that of global structure.
A suggestion is a stimulus that suggests a certain response. A
suggestion is always a suggestion to do something. The ringing bell and
the smell of meat are both stimuli for Pavlov’s dog, but before the
experiment begins, the smell is a suggestion (to salivate) while the former
is not. Basic suggestions are possible to the sugger because its nervous
system is configured to respond to a certain kind of stimulus in a certain
way – unless this response is inhibited by competing suggestions to the
6. Logic of Conversation
contrary. Whether this configuration is hard-wired or acquired through
experience, or through some combination of these, is of no importance
here.
For such configuration to occur, however, two pre-conditions must be
satisfied. First, the stimulus itself must be treated as a discreet and
recognizable event. Second, the suggested response must somehow be
selected as appropriate to the stimulus in question. Both segments of the
arc – the perception of the stimulus and the selection of an appropriate
response – are subject to a kind of logical necessity. Logically, the
stimulus must stand out as a figure against a background, and must be seen
as similar to other events of that same kind. Logically too, the response
must be appropriate to that type of stimulus: a judgment of
appropriateness, at some conscious or unconscious or evolutionary level,
must have taken place.
The human capabilities, shared with most advanced suggers, to
distinguish figures against their typical backgrounds and to distinguish
importantly different kinds of figures, evolved via the logical process of
natural selection. Similarly, the conversion of suggestion into sign occurs
through the logical process of empirical learning. Both processes depend
on induction, a logic of associations whose rule is: “Expect what has
happened before.” Whether such inductions are justified, whether the
expectations based on them are satisfied is beside the point, which is that
the ability to recognize events as being (or not being) of a certain kind,
and the ability to remember and anticipate event configurations are logical
pre-conditions for adaptive learning. Such patterns of suggestivity and
signification comprise the local structure of conversation.
Global Structure: The Logic of Ecology and Culture
The logic of the third quadrant, that of global structure, is that of
evolution, whose scope is wider even than that of conversation since it is
found wherever chaos submits to order in the absence of imposed design.
In the physical sciences, evolutionary selection for dynamic stability
seems to govern the structuring of systems at all levels, from the sub-
atomic to the cosmic. In biology a similar process controls the origin of
species, and the size and distribution of populations within the whole
ecology and biosphere. In anthropology, this same evolutionary logic
constructs and maintains the patterns of language and culture. We are now
discovering that a similar logic governs the configuration of brains and the
personal learning of individuals. The basis of this logic is an utter
tautology: As Gregory Bateson put it, “Longer lasting patterns last longer
than patterns that last not so long.”
We might take language as the prime example of conversation’s global
logic, but what is true for language is true for every repertoire affording
expressive possibility – music, painting, dance, architecture, what have
you. In every case, a limited “vocabulary” of techniques and materials
supports a combinatorial explosion of utterance. Here, probably, is the
6. Logic of Conversation
answer to the English schoolboy’s great question, “Why does everything
in Spain look Spanish?” At one time, almost everything a tourist saw
would have been fabricated from characteristically Spanish
“vocabularies,” evolved for self-consistency and coherence. This is no
longer be true. At least, in the richer parts of town, we find the same
styles around the world today, because the same “vocabularies” are
available everywhere.
What we find in the logic of this third quadrant is that conversation is
a self-stimulating “production system” that lifts itself by its own boot-
straps. Constrained, to be sure, by available material resources, cognitive
structures developed in the past set the terms for further expression. Of
these twin constraints, the latter is by far the more important, once a
minimum of the former is given. This is why conversations can thrive
splendidly under conditions of relative scarcity, and why there is all the
difference in the world between real impoverishment and the “genteel”
poverty of the monk, the artist, the “distracted” intellectual, or the
dispossessed aristocrat. This is why people and whole societies can strike
us as impoverished, although rich beyond belief. Quite literally, they have
more material wealth than they know what to do with.
In sum then, we are in the structuralists’ debt for their insights into the
relative closure and self-consistency of language and all other global
artifacts of conversation. Though it is foolish to think of these as
hermetically sealed from outside influence, it is correct and important to
be aware of them as providing (at any given time), a definite structure of
expressive possibility. Whatever else constrains and shapes it, pursuant to
the logic of this third quadrant, that structure must be compatible with
itself. Ammunition must fit the gun barrels; railroad cars must fit the
track. Styles of dress, of manners, language and thought are somehow of a
piece. So long as Spain remains a more-or-less self-contained society,
everything one sees in that country will look Spanish.
Global Process: The Logic of History
The phrase “logic of history” is in bad odour today, Karl Popper’s charges
in The Poverty of Historicism being widely accepted and largely correct.
For several reasons, notably the chaotic instability of conversation, we can
never know enough about a society and its conversation to predict the
future with any assurance. It is sheer arrogance to set your favourite
ideology at the end of history, denouncing all who oppose it as enemies of
progress. Thus, it is prudent to begin this section by rejecting such tactics
as categorically as possible.
This said, we are constantly forming expectations about the likely
development and outcome of various conversations – various social
processes – and cannot live without doing so. And our record in making
such judgments is much better than chance. When seasoned advocates
guess at the likely bias of a prospective juror, when investors guess at the
development of a business or a market sector, when governments
6. Logic of Conversation
formulate policies to advance (what they conceive to be) a long-term
national interest, they are in each case making judgments about the future
course of a conversation. Admittedly, they are not making predictions in
quite the same sense that astronomers do in predicting a solar eclipse.
Accordingly, to avoid confusion, let’s reserve the word prediction for what
physicists do when they calculate the future trajectory of a dynamical
system from known initial conditions. Let’s speak of forecasting for
systems (like the weather) that are predictable in a statistical sense – for
which we can calculate a probability that there will be rain (for example)
three days from now, but cannot know whether it actually will rain. Let’s
speak of anticipation in connection with systems whose futures cannot be
calculated even in a statistical sense, but for which reasonable fears, hopes
and contingency plans are still feasible, based on the interplay of known
interests, habits and intentions. Obviously, human conversations belong in
this third category.
Anticipations for a given conversation are based partly on the stated
intentions of the players, partly on estimates of their characters and
capabilities and interests, but partly too on one’s sense of the
conversational dynamic itself. In every intense relationship one can
observe how this last sometimes works against the protagonists’ conscious
purposes and interests. The game called “Prisoner’s Dilemma” shows
how two crooks can be made to rat on each other, against their collective
best interests. A play by Jean Giradoux,9 written on the eve of World War
II illustrates how a conflict nobody wants can be inevitable. Hegel gave
us a delicious phrase, "the cunning of Reason" for this tendency of
conversations to acquire a global life of their own – to the point of
manipulating participants into foolish or self-destructive behaviour.
Whenever two people get married, or two nations go to war against their
better judgments, this “cunning,” or “logic” is at work.
Generalizations about how, and toward what ends a conversation will
develop are hard to come by. In fact, there are only two patterns I’d
venture to bet on: The first is articulated in the great dictum of William
James, that “You can’t have enough of anything without having too much
of it.” Or equivalently: “In the long run, you tend to get exactly what you
want – good and hard.” The explanation is that when individuals, or
whole conversations want something, they build routines and systems to
produce it; and in the long term these systems lead lives of their own. For
every human good whatever, it is easier to build a system to manufacture
more of it, than to slow or stop that system once it is built and running.
There is a children’s story along these lines to explain why ocean water is
not drinkable: A boy once bought a magic salt mill from a sorcerer, and
then forgot the spell to turn it off . . .
A second pattern, not unrelated to the first, is the tendency of
9 Called La Guerre de Troie n’aura pas lieu, (literally, “The Trojan War Won’t Happen.” Translated
into English it was re-named, The Tiger at the Gates.
6. Logic of Conversation
conversations to invest resources in cultural facilities of all kinds, long
past the point where such investment seems advantageous to most
members. It is for this reason, probably, that marriages and civilizations
break down: The carrying costs of infrastructure, artifacts and social
arrangements become increasingly irksome in comparison to the benefits
they afford. Some obscure law seems to be at work here: Complexity
begets further complexity; and it is always much cheaper to build a
system than to maintain it once it is built.
Be this as it may, fourth-quadrant process logic, or systems logic is the
causal influence exerted by the dynamics of a conversational system on its
individual components (you and me). Not only does any particular
conversation exert formative pressures on its individual participants;
conversation as such does so. Locally and globally, diachronically (as
process) and synchronically (as structure) conversation is one “game” that
creatures cannot help but play. For this reason, it is only partly correct to
think of conversation as a rule-bound (and therefore, culturally embedded)
“language-game.” Although particular conversations can fruitfully be
studied in these terms, conversation as such transcends all boundaries of
language and culture. It has features that are not culture-dependent, but
intrinsic to the nature of the process itself. With necessary allowances
made for differing conditions and circumstances, conversation is the same
kind of process for all suggers everywhere – for all creatures, devices and
systems that must be understood and dealt with from the intentional
stance.
Subject to the same necessities wherever found, the exchange of
suggestions in conversation is a universal process that overflows all
channels, shifts all boundaries, and makes up its rules as it goes along –
across all frontiers of individuality, culture and even biology. It follows
that however difficult it may be to understand someone of the opposite
sex, or class, or culture, or religion, or species, a basis for understanding
always exists; and workable understandings are possible. In this sense,
we may conjecture that no sentient, social being can be radically
unintelligible to another. Whenever such creatures become engaged, a
conversation starts up, and it is predictable that this must eventually lead
to some degree of mutual intelligibility and accommodation amongst
viewpoints that are not eliminated outright.
The logic of conversation contradicts those extreme versions of
relativism that proclaim the mutual irrelevance of differing viewpoints and
cultures. Rather, its claim is that no autonomous, sentient creature with a
vulnerable, metabolizing body can be wholly beyond the comprehension
of another such. Whenever such creatures (and their cultures) come into
contact, they cannot just ignore each other. They must either learn enough
about each other to avoid hostilities, or they must learn enough to do battle
effectively. Either way, some conversation between them must occur.
Unless this universality is kept in mind, ideas of structuralism, culture-
6. Logic of Conversation
relativity and Wittgensteinian “language-games” are quite misleading.
The evolution of some characteristic language game – some recognizable
culture – is the predictable result whenever suggestions are regularly
exchanged by humans (probably, by all sentient creatures), however
diverse their backgrounds. This tendency of conversation to “cut channels
for itself” – to evolve facilities and rules for its own convenience – is the
central theorem of its logic, the central reason why cross-cultural
conversation is always possible. There is no culture anywhere, nor can
there be, that does not incorporate and implement the principles of
conversation. Nor is there conversation anywhere that does not link its
interlocutors with common culture.
In the last analysis, it is the logic of conversation that affords a
possibility of reason. First, (well below and prior to the verbal level), it
enables a climate of relationship in which reason can function: a pattern
of mutual engagement, intelligibility, respect, and reliability. Second, it
enables conventions of signification, that make for articulate speech and
thought. Finally, it teaches techniques of reasoning, a systematic
application of understanding, articulation and inference via the logic
we’ve been sketching. Of course, the logic of conversation cannot compel
anyone to think or argue reasonably, any more than classical logic could
compel them to do so correctly. But, like classical logic, it points toward
and demonstrates a possibility of “reasoning”; and it warns us what to
expect when the principles are violated. Through its structures,
conversation between people and with the world gradually renders that
world intelligible. When we violate those structures, our affairs become
less and less intelligible, and we become increasingly frustrated by
confusion, misunderstanding and conflict.